Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Texas Parasol Plans...

9 views
Skip to first unread message

Richard Lamb

unread,
Feb 11, 2006, 10:14:18 AM2/11/06
to
Rec.Aviation.Homebuilt special...

Eventually, these will be made available to everybody on Matronics, but for
the time being they are posted at http://www.home.earthlink.net/~tp-1/ just
for the nice boys and girls of RAH and RAU.

Enjoy...

Richard

Tater Schuld

unread,
Feb 11, 2006, 11:37:11 AM2/11/06
to
now here is a perfect example of what I am looking for. free plans, a design
that looks like it could allow a number of different engines, both two
stroke and four, both aircraft and automotive AND other.

my only complaint with it is that the weight loads are tight. looks like a
great grasshopper type plane, with the possibility of using it for a bit
more.

now to find a similar design with a higher payload and room for an extra
seat.


"Richard Lamb" <cave...@Xearthlink.net> wrote in message
news:ebnHf.9580$Nv2....@newsread1.news.atl.earthlink.net...

Le...@caspercityauto.com

unread,
Feb 11, 2006, 12:34:47 PM2/11/06
to
Bigger, faster, MORE!

They are never satisfied :-)
=================
Leon McAtee

Le...@caspercityauto.com

unread,
Feb 11, 2006, 12:34:46 PM2/11/06
to

Smitty Two

unread,
Feb 11, 2006, 12:40:02 PM2/11/06
to
In article <YooHf.81$rV6...@fe04.lga>,
"Tater Schuld" <Tate...@yahoo.com> wrote:

> now here is a perfect example of what I am looking for. free plans, a design
> that looks like it could allow a number of different engines, both two
> stroke and four, both aircraft and automotive AND other.
>
> my only complaint with it is that the weight loads are tight. looks like a
> great grasshopper type plane, with the possibility of using it for a bit
> more.
>
> now to find a similar design with a higher payload and room for an extra
> seat.
>
>


Tater,

OK, you want to fly inexpensively. And, you've deduced that rolling your
own is one way to (possibly) reduce expenses. And, you've stuck around
here and maintained your composure in the face of some snotty comments
from some of the so-called participants here. Great.

I may have missed it, but I haven't seen you state your "mission
profile." Your comments above are a good beginning, but flesh it out a
little. Doing so may help get you further suggestions.

It may help you to make a list of your objectives, including total cost,
time line, skill set you have or are willing to acquire (kit vs.
plans-built), material preference (i love composite planes but wouldn't
want to build one), speed, range, payload, etc. ad infinitum.

Put down every thing you can think of, and then start sorting them out
in order of relative importance. Then post the top five here, and see
whether someone can pull a flying rabbit out of a hat for you.

Also, as others have suggested, there are many ways to get into the air,
including hang-gliding, soaring, powered parachutes, ultralights,
renting, joining a club, etc. When someone offers suggestions like that,
some feedback from you may help to narrow the field. If you *must* own
your own "real" airplane, say so, and let's take it from there.

The cost of flying doesn't just include the cost of the vehicle, it
also includes obtaining training and possibly a PPL, and a host of fixed
and per hour costs like fuel, storage, maintenance, insurance, etc.

Don't underestimate the power of action. Research is great, but once you
get started, magic happens. Join your local EAA chapter and go to
meetings. Buy the magazines that show 200 kitplanes in one issue. Take a
workshop or two.

If you're going to build, start doing it on your kitchen table if you
have to, with a file and a steak knife. If you wait until you have a
2500 square foot climate-controlled workshop with CNC machines and a
$200,000 bank balance, you're going to be walking until you're dead.
Start building, and things will happen that are completely beyond your
control.

clareatsnyder.on.ca

unread,
Feb 11, 2006, 3:27:15 PM2/11/06
to

Got the wing spar design updated yet? Or are we trying to thin the
herd?

Richard Lamb

unread,
Feb 11, 2006, 3:49:23 PM2/11/06
to
Tater Schuld wrote:
> now here is a perfect example of what I am looking for. free plans, a design
> that looks like it could allow a number of different engines, both two
> stroke and four, both aircraft and automotive AND other.
>
> my only complaint with it is that the weight loads are tight. looks like a
> great grasshopper type plane, with the possibility of using it for a bit
> more.
>
> now to find a similar design with a higher payload and room for an extra
> seat.
>

Tater, despite what the America Bashing Third World (tm) thinks, ALL aircraft
are similarly restrictive on weights. You simply can not pile on excess
weight on any airplane and produce a successful flying machine.

There have been some twos built this way, but they are restricted to
Normal Category maneuvering limits. (That may change in the near future when
Al Robinson finishes his Texas Pete - A side-by-side two seater with a much
heavier built wing.)

This is a great little baby buggy. Inexpensive to build and fly, a great
flyer, and with a proven history.

I offer it as an excellent choice for a first time build.

Richard

Richard Lamb

unread,
Feb 11, 2006, 3:58:17 PM2/11/06
to
clare at snyder.on.ca wrote:

Nothing wrong with the wing, Clare.
But there IS something wrong with trying to overload it like you guys did.

The original agreement was that your fearless leader was to have a heavier
wing designed (by a "real" aero engineer) to handle a heavier airplane.
That was never done.

So if you think you have a bitch coming, point it back north.

Richard

clareatsnyder.on.ca

unread,
Feb 11, 2006, 4:56:37 PM2/11/06
to
On Sat, 11 Feb 2006 20:58:17 GMT, Richard Lamb
<cave...@Xearthlink.net> wrote:

If built according to plans your wing will NOT withstand the G rating
listed in your info - and even YOU did not fly YOUR plane built
according to plans.
The jury strut is also CRITICAL, and as designed is an accident
waiting to happen.
So answer the question - have you upgraded the plans?
If the answer is no, everyone on the group is entitled to know the
plane is dangerous as designed, but can relatively easily be made into
a safe airplane.

Richard Lamb

unread,
Feb 11, 2006, 7:29:46 PM2/11/06
to
Well, folks, there you have it.
The experts have again spoken.

Clare, I can understand your anger.
But it's misplaced.
You got had.
But not by me.
I did everything I could to warn you guys.
Although I wonder if Gary ever passed that on to you.

He took a design with a 350 pound recommended empty weight and
built something else. Beefed up fuselage structure, .065 wall
spars (which did nothing to add strength - just cheaper),
ALL METAL SKINS on the fuselage AND wing, and converted auto engines.

What was the final weight you guys came up with? 550? 650? Empty!

But the answer was always, "But that's the way we want it".

So Gary offered to hired an engineer to design a wing compatible
with your wants. But that gentleman died before finishing the
work - and now it's MY problem? Sorry, guy. No way.

That's why I finally bit down and asked you guys not to call
it by the Texas Parasol or Chuckbird name.
It's not - and you damned well know it.

Those drawings are straight from my first parasol.
And yes, I did fly it just as it is drawn, with the exception
of using a VW on it rather than a Rotax.

Rave if you must, Clare, but there are several dozen of these
planes _flying_ for over 20 years now. Doc, who has been
the test pilot on almost all of these, had over 650 hours on
his "Lucky Lady" when the airfield changed hands and he quit.
Doc loved to play acro with it. Loops (well, tall skinny ones),
spins, rolls. I'll trust my life to his test work because I've
seen what he can do with it.

As for you "analysis"?

So far we've seen NO structural failures, and only one fatality -
on a first flight, ran out of gas and spun it in.
(I can't tell you how hard that was to deal with.)

Changing the subject only a bit...

I went out to Kitty Hawk Airfield last weekend to look at a CGS
Hawk I was hoping to buy. The fellow I met with (Don) was very
knowledgeable about the design - AND that particular airplane.
I'm very impressed with Chuck's design, but I walked (ran?) away
from this airplane.

A few years ago some fool decided the plane needed more power
and mounted an 80 hp Rotax 912 on it. (anybody here familiar
with the Hawk?). On the first takeoff, the engine twisted plumb
off the mount, cut the tailboom off and (obviously) crashed, killing
the pilot.

Don was very up-front and honest about it - and the condition of
the rebuilt machine. The tailboom was extended, the nose also,
and a Rotax 582 installed. It weighs well over 350 pounds.
But many of the other local "experts" call it a POS death trap.

Unfortunately, it is still refered to as a CGS Hawk - and I'll bet my
bottom dollar that Chuck S absolutely hates that.

Just about the same way I feel about what you fellows have done.

Disgusted,

Richard

Richard Lamb

unread,
Feb 11, 2006, 7:33:20 PM2/11/06
to
And my apologies to the rest of the group for airing
dirty laundry in public...

Richard

Richard Lamb

unread,
Feb 11, 2006, 7:40:39 PM2/11/06
to

Richard Lamb

unread,
Feb 11, 2006, 7:41:52 PM2/11/06
to

Ron Wanttaja

unread,
Feb 11, 2006, 7:43:26 PM2/11/06
to
On Sun, 12 Feb 2006 00:33:20 GMT, Richard Lamb <cave...@Xearthlink.net> wrote:

>And my apologies to the rest of the group for airing
>dirty laundry in public...

Absolutely no problem, Richard, I appreciated hearing your side of it.
Congratulations on getting the plans online for free downloading... wish we
could do that with the Fly Baby.

Ron Wanttaja

Richard Lamb

unread,
Feb 11, 2006, 8:31:39 PM2/11/06
to
Ron Wanttaja wrote:

Thanks, Ron.

It would be no problem, technically, to put the plans in machine form.
Well, other than the legality issues...
Find a solution to that one and I can have them ready in a week or two.

I think I was about 12 years old when Air Progress Homebuilt issue presented
the Fly Baby. I wanted one so bad I could taste the spruce.

I still think it's one of the all time best all wood amateur built designs.
A real classic.

Always will be too.

Richard

Message has been deleted

veed...@isp.com

unread,
Feb 11, 2006, 8:45:31 PM2/11/06
to

Richard Lamb wrote:

>
> Those drawings are straight from my first parasol.
> And yes, I did fly it just as it is drawn, with the exception
> of using a VW on it rather than a Rotax.
>

------------------------------------------------------------------------------------

That's a lie and Richard knows it.

The set of plans Richard has posted are the same ones he sold to me
(and many others) for $80. As of February, 2003, I had identified
numerous errors in the plans and provided corrections to several other
builders. Richard's contribution was to refer to my questions as
'yammering.'

I was trained to audit blueprints for errors but the errors in the
Parasol drawings will be obvious to all -- compare the dimensions of
the cabanes to the width of the fuselage, or the dimensions shown for
the landing gear/strut carry-throughs.

Richard built a plane. And Richard made some drawings. But the
drawings depict parts that COULD NOT FIT TOGETHER. The extent of the
errors made it clear that they were not simple typos -- they appeared
to be for a fuselage OTHER than the one shown in the drawings. This
lead to an interesting series of exchanges between Richard and I in
2003, because if you modified the carry-throughs to attach to the
fuselage, it caused interference between the landing gear or the
struts. So which did he adjust? He could offer no explanation, making
it painfully clear that he had NOT designed the airframe and had taken
measurements from at least two airframes in creating 'his' drawings,
which he then sold to the public at eighty bucks a whack. Waytago,
Richard.

Since the parts don't fit, the plane won't fly, which makes any
question of safety moot. But in the process of discovering that you'll
piss away a lot of money on metal that will end up being unusable.

I still have the file of drawings, should anyone be interested. But
which way you go with the corrections is up to you -- I abandoned the
project when it became obvious it was a scam.

-R.S.Hoover

clareatsnyder.on.ca

unread,
Feb 11, 2006, 9:32:17 PM2/11/06
to
On Sun, 12 Feb 2006 00:29:46 GMT, Richard Lamb
<cave...@Xearthlink.net> wrote:

>Well, folks, there you have it.
>The experts have again spoken.
>
>Clare, I can understand your anger.
>But it's misplaced.
>You got had.
>But not by me.
>I did everything I could to warn you guys.
>Although I wonder if Gary ever passed that on to you.

No- Gary built a wing according to plans and sandbag tested it under
an engineer's supervision. The engineer stopped the loading before
failure because of excessive deflection IIRC.


>
>He took a design with a 350 pound recommended empty weight and
>built something else.

No fuselage was included in the testing

>Beefed up fuselage structure, .065 wall
>spars (which did nothing to add strength - just cheaper),
>ALL METAL SKINS on the fuselage AND wing, and converted auto engines.
>

No, the fuselage was NEVER FINISHED NO ENGINE WAS EVER INSTALLED.


>What was the final weight you guys came up with? 550? 650? Empty!
>
>But the answer was always, "But that's the way we want it".

No, you are fantacising.


>
>So Gary offered to hired an engineer to design a wing compatible
>with your wants. But that gentleman died before finishing the
>work - and now it's MY problem? Sorry, guy. No way.
>
>That's why I finally bit down and asked you guys not to call
>it by the Texas Parasol or Chuckbird name.

>It's not - and you damned well know it.
>
>Those drawings are straight from my first parasol.
>And yes, I did fly it just as it is drawn, with the exception
>of using a VW on it rather than a Rotax.

And sleaves in the spars by your own admission. Which are NOT in the
plans.

*** Free account sponsored by SecureIX.com ***
*** Encrypt your Internet usage with a free VPN account from http://www.SecureIX.com ***

Richard Lamb

unread,
Feb 11, 2006, 11:12:08 PM2/11/06
to
veed...@isp.com wrote:

> Since the parts don't fit, the plane won't fly, which makes any
> question of safety moot. But in the process of discovering that you'll
> piss away a lot of money on metal that will end up being unusable.
>

> -R.S.Hoover
>

Well, if it won't fly then I guess it just won't fly...

Glad you opted out, Robert.

ChuckSlusarczyk

unread,
Feb 11, 2006, 11:50:47 PM2/11/06
to
In article <_jvHf.14086$rH5....@newsread2.news.atl.earthlink.net>, Richard
Lamb says...

>I went out to Kitty Hawk Airfield last weekend to look at a CGS
>Hawk I was hoping to buy. The fellow I met with (Don) was very
>knowledgeable about the design - AND that particular airplane.
>I'm very impressed with Chuck's design, but I walked (ran?) away
>from this airplane.
>
>A few years ago some fool decided the plane needed more power
>and mounted an 80 hp Rotax 912 on it. (anybody here familiar
>with the Hawk?). On the first takeoff, the engine twisted plumb
>off the mount, cut the tailboom off and (obviously) crashed, killing
>the pilot.

Rich any more info on this crash? I haven't heard about it .Usually any time a
Hawk goes in I hear about it.

>
>Don was very up-front and honest about it - and the condition of
>the rebuilt machine. The tailboom was extended, the nose also,
>and a Rotax 582 installed. It weighs well over 350 pounds.
>But many of the other local "experts" call it a POS death trap.
>
>Unfortunately, it is still refered to as a CGS Hawk - and I'll bet my
>bottom dollar that Chuck S absolutely hates that.


Unfortunately it happens more then I'd like it to. Every wants to be a designer
and "improve" your airplane,then when there's problems they drop it on your lap.
They will always call it a Hawk so they can sell it .Then some poor unsuspecting
guy buys it and blames me for the problems because he was told it was a Hawk.

>
>Just about the same way I feel about what you fellows have done.

Understood completely.

See ya

Chuck S
>
>Disgusted,
>
>Richard

veed...@isp.com

unread,
Feb 12, 2006, 12:43:20 AM2/12/06
to

Richard Lamb wrote:

>
> Well, if it won't fly then I guess it just won't fly...
>
> Glad you opted out, Robert.

---------------------------------------------------------------------

And that is exactly the kind of flippant, smart-assed response builders
got when they sought information about the errors in the drawings.

Nice job, Richard. You do yourself proud.

-R.S.Hoover

Morgans

unread,
Feb 12, 2006, 2:59:21 AM2/12/06
to

<veed...@isp.com> wrote
snip>

> I was trained to audit blueprints for errors but the errors in the
> Parasol drawings will be obvious to all -- compare the dimensions of
> the cabanes to the width of the fuselage, or the dimensions shown for
> the landing gear/strut carry-throughs.
snip>
> Since the parts don't fit, the plane won't fly, which makes any
> question of safety moot. But in the process of discovering that you'll
> piss away a lot of money on metal that will end up being unusable.

Do how do these plans look? Does everything look like it would work, now?
--
Jim in NC

veed...@isp.com

unread,
Feb 12, 2006, 9:03:09 AM2/12/06
to
Morgans wrote:
>
> Do how do these plans look? Does everything look like it would work, now?
>
---------------------------------------------------------------------------------------

Dear Jim,

The drawings Richard posted are the same garbage he was flogging for
$80 a copy four years ago. The cabanes and the carry-throughs for the
LG & struts appear to be for a fuselage that is 24" wide whereas the
fuselage as shown is only 22". The drawings do not contain enough data
to develop the angles of either the axle-carrier cluster or the rear LG
'gooseneck' where it attaches to the carry-throughs. The LG yoke does
not match the hole locations shown for its attachment to the front
carry-through and if the carry-throughs are attached as shown then
either the location for the landing gear attachment OR the strut
attachment will be in error since doing it one way violates the rule
for edge-distance whilst doing it the other requires re-locating the
carry-through... which throws out the locations for all of the other
structural members in that portion of the fuselage.

The point here is that Richard's statement about flying a plane built
from those drawings is pure Texas bullshit. He built an airplane and
it flew but when the errors became evident and several of us asked him
to provide various dimensions from that airplane he simply refused to
do so. Rather curious behavior for the 'designer' of an airplane,
don't you think?

Work it out for yourself, Jim. Take your yo-yo and a piecea cardboard
and simply lay-out the forward carry-through. The errors are immediate
obvious and on the surface, don't look too serious. Now try resolving
them. You've got the point where the carry-thrus attach to the lower
longerons and that's pretty much fixed because you've already
fabricated the side-frames. Now you've got to accommodate the LG yoke,
the landing gear leg and the strut attachment. That's where you'll run
into conflict, espeically so if you've already drilled the
carry-throughs... which are now junk because the holes are in the wrong
locations. (Along with those four cabanes, if you bent them according
to the plans.)

So what are you going to move? The fasteners for the landing gear legs
must ALIGN between the front & rear carry-throughs, otherwise the legs
won't pivot. But the flanges of the forward carry-throughs are NOT
parallel to each other because of the curvature of the lower longeron
-- you'll run out of edge-distance before you get the legs to align
WITHOUT interference with either the longeron or the strut-end. Adjust
any one of the errors to fit and the result will create a conflict with
the other two points of attachment. And we're looking at some
significant loads here; forward wing strut, forward landing gear leg,
all of which goes into the carry-throughs then into the longeron
attachment. You wanna GUESS at the dimesions? Because that's what it
boils down to.

This is all simple geometry, Jim, right there in front of you on the
drawings. Richard finally admitted that he more-or-less built the
landing gear in-place, which means he KNEW the drawings were bullshit.
So what did he change? What were the dimensions of the finished
structure? And that's where he goes all coy and sez he'll leave it up
to you to figure out. Now isn't that cute.

Will it fly? Of course it will fly! Lookit how many 'Chuck Birds' are
already flying. But the plans Richard drew up simply don't make sense
and he's obviously incapable of correcting them.

Designer my ass.

-R.S.Hoover

ChuckSlusarczyk

unread,
Feb 12, 2006, 11:10:22 AM2/12/06
to
In article <1139752989.2...@g14g2000cwa.googlegroups.com>,
veed...@isp.com says...

>Will it fly? Of course it will fly! Lookit how many 'Chuck Birds' are
>already flying. But the plans Richard drew up simply don't make sense
>and he's obviously incapable of correcting them.

One small question what's a "Chuck Bird" ? :-)

See ya

Chuck S

Richard Lamb

unread,
Feb 12, 2006, 12:35:17 PM2/12/06
to

Reproduced (with permission) from Experimenter magazine.

Chuck Beeson and his Texan Chuckbirds

by Dick Cavin
June 1988

Unless you attended the EAA South-west Regional Fly-In at
Kerrville, Texas the past six years you have probably never heard of
Chuck Beeson and his ultralight-type Texan Chuckbirds. One of the
reasons for this dearth of publicity was that Chuck simply wasn't
ready for the spotlight.

Chuck is a rugged individualist. He has
spent 35 of his 52 years in aviation as a pilot and A&P. His
childhood years were spent designing and building models. In addition
to over a couple of dozen J-35, Champs and the like that he has
completely restored from bare bones, he has designed and built 29
airplanes in the ultralight weight class (or just beyond).

Of that 29, no two are exactly alike. Before ultralights came on the
scene, Chuck busied himself with the building of seven Pietenpols!
One was powered with a Chevrolet Vega engine conversion, effectively
Making it a single place with its overly heavy engine. Most of his 29
flyweights have been parasol monoplanes, but there were six biplanes
of varying configuration, one low-wing and six two-placers. In every
case except one, Chuck has designed, built and flown each airplane
(for several hours at least) before selling them to local pilots who
had the experience to take care of themselves and their airplanes.
Fourteen of them are still flying regularly from a grass roots style
airpark in suburban San Antonio where all are based. Others are
spread out over Texas. Some of the first airplanes didn't quite suit
Chuck, so he would take them apart and make a modification to correct
the deficiency, sometimes building a new fuselage, wing, or tail
group. Sometimes it meant building a new aileron, or perhaps putting
a new engine or prop on, but always it was experiment, experiment and
perfect, improving each model a little from the last one in an
ongoing evolutionary process.

Not only has Chuck had the opportunity to experiment and improve on
various airframes, but he has also evaluated several engines. In
addition to the popular Rotaxes he has tried three Zenoah engine
models and most recently the Subaru. He has also used Continental
A-65s in a couple of models, but removed them in favor of the two
cycle engines because the lighter weight two cycles give improved
performance. With their gear boxes the torque of the two cycles was
multiplied so that their thrust was as good or better than the A-65,
even though their horsepower was less. Chuck favors the Zenoah 40 hp
liquid cooled engine for his parasol kits. He has several of them
flying with this engine and four of them have gone over 400 hours
with only plug changes and no other problems. Not only are they
light, but they're also cheap, since they are a surplus engine 10,000
were on the market at one time. Combined with a Rotax 2.58:1 gearbox
and a Tennessee 60&quot; x 28&quot; wood prop, the Zenoah engine
makes a dependable powerplant package that's hard to beat for thrust,
weight . . . and cost. It also is a very smooth running engine from
idle rpm to full power. Chuck originally used the two-cylinder
opposed Zenoah that was air cooled, but after three over-temperature
seizures he said, "Enough is enough", and went to liquid-cooled Zenoahs
in the 40 and 50 hp range.

He is exceptionally pleased with the Subaru engine, too, having flown
it on his latest parasol and a new biplane prototype, which features a
swept upper wing for ease of entry to the cockpit. Not only is it smooth
as silk, it is super easy to start. He has flown it direct drive and with a
cog belt speed reduction unit that machinist friend Tom Arnold made
for it. Presently Chuck leans toward the direct drive unit, as the
excess of horsepower and low wing loading give him adequate take off
and climb performance and better cruise. For those tiny fields,
though, the increased thrust from the belt drive reduction is the
logical choice. It will pull a healthy static thrust of 287 lbs., so
this will also be a plus factor for high altitude take offs.

Chuck's philosophy towards kit purchasers is to eliminate every
material or process that demands much in the way of skills, tools,
and time for the first-time builder. Of course, the builder will
still have to complete over 51 percent of the work if the aircraft is
to be registered in the experimental amateur built category, which
most certainly will. The simplicity and straightforwardness of
Chuck's basic design not only reduces the required hours of building
for the beginner, but also does the same thing at the manufacturing
level. Chuck says he has been able to eliminate about 100 man hours
of labor with his progressive design refinements. </P>
<P ALIGN=JUSTIFY>
This philosophy dictates the use of extruded aluminum angle for the
fuselage structure, which not only is lighter, but also is actually
stronger than a welded steel tube structure. He uses 3/4 inch x 1/8
inch 6061 T-6 angle for the basic fuselage structure, with larger
one-inch by one-inch by 1/8 inch angle used in concentrated load
areas. It is joined together with regular 1/8 inch universal head
solid AN rivets (not pop rivets) with two rivets per joint minimum.
How strong is the fuselage? With it supported on both ends, Chuck
loaded fifteen 80 lb. sacks of cement in the middle of the fuselage
(1200 lbs. total) without measurable deflection. Rotating it 90
degrees produced the same result. If one supports a soda straw at
each end it doesn't take much pressure to collapse it in the middle,
while an angle is much more resistant to buckling, explained Chuck.
While welded steel tube structures may be adequate in strength for
the loads imposed on them in service, there is no comparison in the
required building time.

Chuck says he can easily build a fuselage in two days, using simple
table top assembly jigs, while a welded structure would require five
to eight times as many man hours. Besides costing more in labor and
materials, the steel tube structure would still be subject to rust
and corrosion. If stringers are added for shape it would also mean
welding stub tubes with angle fittings, where on the aluminum
structure it only means a simple angle clip to be riveted on. Basic
fuselage weight is a svelte 24 lbs. In his basic &quot;no
weldment&quot; concept Chuck only uses one welded assembly - at the
apex of each landing gear vee, as it is tied to the axle and center
strut with 1/4 inch AN bolts. Even the engine mount uses the heavier
one-inch by one-inch by 1/8 inch extruded angles, again bolted with
AN bolts. Here he uses two parallel angles per side, running fore and
aft, and separated with two Lord mounts per side. These in turn are
attached to bracing vees that tie them to the fuselage structure via
bolts.

The landing gear is strictly J-3, with shock cord on each center
strut absorbing landing loads. For simplicity, Chuck uses a single
curved tube (instead of a vee) to feed loads from the shock cord
units into the transverse double angles that also pick up compression
loads from the front vee strut on the side. These double angles are
in parallel, with the tube in between secured with a bolt that ties
all three pieces together. This also permits some rotational movement
of the vee struts at their top end. Wheels are 5 inch plastic go-kart
wheels. No brakes are used on the ultralight version. Wherever
aluminum tubing is used on the airplane Chuck ties it into the
fuselage structure with either angles or flat straps on each side of
the tube and secures it with bolts.

In higher load areas, such as attachment points of the lift struts to
the fuselage, he uses one or more doubler tubes of .058 wall inside
the outer tube the last six inches. At the other end where these
1-3/8 inch by .035 lift struts tie to the 2 inch by .049 tubular
spars he uses two .090 stainless steel plates on each side, slightly
joggled to mate with both tube diameters, again using 1/4 inch bolts
through the neutral axis of both the spars and lift struts. The jury
struts are one inch by .049 inch tubing, tied to the spars and lift
struts with stainless steel straps that go over and under the tubes
and are blind riveted. Chuck originally had vee type lift struts
(like a J-3) on his first one, but removed them and went to parallel
struts when he found the arrangement too flexible with the pylon type
cabane he uses. To pick up drag loads he supplemented the lift struts
with 3/32 aircraft cable in an &quot;X&quot; configuration between
the struts, which also picks up torsion loads at the outer strut
attach Points.

Chuck has used two different wing spans in his parasols. The kit
model has a 26 ft. span, while some earlier models had a 22 ft. span.
Chord is four feet on either. Wing area is 104 square feet on the kit
airplane. With empty weights of 240 to 250 lbs and an average payload
of 210 lbs. (five gallons of fuel and an 180 lb. pilot) the wing
loading is a modest 4.42 lbs per square foot. This translates to a
&quot;stall&quot; speed of about 20 mph. The airfoil Chuck uses is
his modification of a Clark Y. This gives a 14 percent thick wing
section. Ribs are all 1/2 inch Klegecell polyvinyl foam and are cut
to shape on a high speed router. Each rib takes only seconds to make
, he says, and requires no edge sanding. There are no compression
struts in the wing, as each rib has a two inch wide bent up C channel
of .025 glued alongside the rib to take compression loads. This
channel also serves to attach the ribs to the spars at each end via
blind rivets. There is a single one-inch by .049 inch drag strut
inside the wing that runs from the root end to a point about four
feet outboard.

The aileron is also simplicity itself. It is a bent up piece of .020
aluminum that is pop riveted to the 1.5 inch x .035 inch tubular
spar, top and bottom. Three piano hinges are used per side and
attached via five pop rivets and epoxy adhesive to the wing's rear
spar. Ailerons are full span and have a formed rib at each end. At
the customer's option he can have full span ailerons, half span
ailerons with half span flaps, or full span flaperons. Ailerons are
operated via push-pull tubes with self-aligning bearings that run
from a fitting at the root end of the aileron to a bellcrank at the
aft end of the stick's torque tube. No differential throw is built
in, but Chuck says that like most ultralights you have to lead turn
entry with the rudder anyway, so adverse yaw isn't a problem. With
3-1/2 degrees of dihedral and the generously sized rudder the pilot
can quite easily pick up a low wing with rudder alone.

The Chuckbird's cabane is basically a pylon where wings are attached
without a center section, thus eliminating eight fittings in the
process. The vertical cabane struts are one inch by .058 inch tubes,
which are attached to the fuselage structure with 1/8 inch thick
straps and four bolts. Coming together at the top, heavy angles in
the horizontal plane on the fore and aft sides provide a secure
attachment for the wing spars. Spars are clamped in position and
match drilled. Parallel heavy angles run fore and aft on each side of
the cabane tubes and are bolted to them and the attach angles,
closing the force box. A 1/8 inch drag strap runs diagonally forward
and down from the spar attach point to further stiffen the pylon fore
and aft.

The parasol fuselage is 16 ft. long from nose to trailing edge of the
rudder. The cockpit width is 19 inches, which would be tight in a
closed version, but since it is a fresh air machine there is no
restriction to a person's shoulders, so it is quite comfortable for
even large pilots. Entrance and egress is easy, since the upper
longeron is low enough to step over easily. Head room looks to be
adequate for even a seven footer. The canvas seat is unique. It is
adjustable fore and aft and is so designed that a person's tailbone
never comes in contact with a cross tube. Visibility over the nose is
excellent, especially since the cowling narrows down to slightly over
a foot in width at the front. The control stick is a husky one inch
by .058 inch tube, as is the torque tube. One-eighth-inch straps are
bolted to the bottom of the stick for the proper gear ratio to the
elevators, giving generous flare power at the 20 mph landing speed
without undue sensitivity at cruise. Eyebolts through the stick are
used to attach swaged ends of the 3/32 inch control cables. Teflon
bearings support the torque tube at both ends. Rudder pedals are bent
up .090 aluminum and are piano hinged to the floor. The floor itself
is a beefy sheet of .040 2024 T-3. It is riveted to the parallel
one-inch angles crossing under the fuselage at the front and rear
spar attach points, transferring the higher loads of the lift struts
and landing gear to the fuselage structure, giving a &quot;wider
footprint&quot; to load paths.

The five-gallon tank sits forward of the instrument panel and is an
off-the- shelf PVC snowmobile tank. There is a half-round shaped
piece of .020 over it that is pop riveted to a 1/2 inch tube at the
top of the firewall. The cowling itself is all metal and attached to
the firewall top and bottom and to a 1/8 inch by 1/2 inch strip
standing away from the sides for engine air exit. Even though the
engine is liquid cooled and the radiator is hung below the firewall a
generously sized opening in the front of the cowl is a big help in
keeping engine temperature well within bounds on those blistering
95-100 degree Texas summer days (a big plus in engine reliability).
The tail group is built from one inch by .049 tubes, with the curved
ends made in a forming jig. The stabilizer/elevator spans 8 ft. The
elevator spar is one inch by .058 inch, while other tubes are one
inch by .049 inch. Formed .020 aluminum ribs give shape between the
tubes. Two double eyebolts are used for each elevator hinge. Bracing
is via a 3/32 inch cable that ties horizontal and vertical tail spars
together. No turubuckles are used on any part of the airframe, thus
saving 15-20 lbs. of weight and at about $15.00 each, it saves
significant money, too. Chuck uses a simple trick to tauten the
cables. He uses &quot;master&quot; cable with a turubuckle to make up
a finished cable. With a &quot;tang&quot; fitting and a long bolt he
can pull them up tight. Slick!

Chuck uses Stits HS90X fabric for covering the entire airplane,
attaching it to the ribs and other structure with Super Seam
adhesive. He likes Stits because the warp and woof weave are equal, a
help in heat shrinking. The turtle deck uses Klegecell formers and
one inch x 1/4 inch wood stringers and he attaches the fabric to it
in the same manner, so no rib stitching is used on the entire
airplane, making the 100-hour building time from the kit a reasonable
figure.

After taking in all these top flight construction details, would you
believe that Chuck has built all of these 29 airplanes without ever
drawing a single line of plans? He has done it all from what was
stored in his memory and from experience! They have all flown well,
but the latest parasol flies the best yet, he says. Chuck's
philosophy on kitting an airplane is on the conservative side. He
thinks it is wrong for someone to build a prototype and then
immediately start selling plans and kits. His airplanes have now
accumulated well over 3000 hours with no structural problems, so he
now feels he is safe in offering it in kit form. Even so he has hired
a consulting engineering firm to do a computer verification of the
structure. They do corroborate his static tests of a + 3.5 and -1-1/2
G loading. On one wing test to 3.5 Gs the spars were deflected 18
inches before the load was released, and they returned to position
without any permanent set. The engineers are also drawing up an
excellent set of (dimensionless) assembly drawings to guide the new
builder. He will also do a step-by-step video of the entire building
process for each kit purchaser. The video will include the final
covering of the wing and fuselage, plus rigging and test flying. You
might well ask what Chuck's next kit will be. It might be his new
Subaru powered biplane - after sufficient testing. He is now building
his version of a 1909 Bleriot just for the fun of it, something he
values highly.

Because he treasures the joy of sport flying around the
grass-root-type airports, he doesn't want to greatly expand his
production and be saddled with all kinds of big business money
worries. He just wants to make still another RHLA (Real Honest Light
Airplane) available for the average guy who wants to fly. Chuck
prices the parasol airframe kit at $3995, which includes everything
aft of the firewall except instruments, dope or glue. He will furnish
a firewall forward kit for $1,000, which includes the water cooled 40
hp Zenoah, muffler, cowling, engine mount and propeller. He will
continue to build the ultralight version, ready to fly, for $6995 for
those who have no place or time to build. All hardware is AN quality.
I think it is safe to say that we'll be seeing Chuckbirds in a good
many other parts of the country as their reputation spreads.

Richard Lamb

unread,
Feb 12, 2006, 12:39:52 PM2/12/06
to
ChuckSlusarczyk wrote:
>
> Rich any more info on this crash? I haven't heard about it .Usually any time a
> Hawk goes in I hear about it.
>

That's what I wanted to contact you about.

I couldn't raise you via email, hence the (old fashioned)
airport page. (Guess ya had to be there...)


This is probably best discussed off-line?

Richard

flybynightkarmarepair

unread,
Feb 12, 2006, 1:37:06 PM2/12/06
to

This construction method was developed by Chuck Beeson, thus the "Chuck
Bird". When Richard started drawing up the construction plans and
selling them, the name changed to the Texas Parasol.

R.S. Hoover, Juan Lozano and I had some dialog with Richard Back In The
Day about the way the wing struts were attached to the spars, and the
use of fasteners in tension rather than shear, and he more or less blew
us off.

I think the design is generally OK (it's the DETAILS that look hinky),
and a good starting point for an experienced builder to improvise, but
as a design capable of construction by a first time builder straight
from the plans, it's on par with a Teenie Two, which is damining with
the faintest of praise.

Still....

I respect Richard for the effort of drawing up the plans and the
manual, and for, after an unsatisfactory experience with Sirius
Aviation (regrettably not uncommon), putting the plans into the public
domain. Perhaps a Open Source effort can iron out the difficulties
R.S. Hoover and others have pointed out.

ChuckSlusarczyk

unread,
Feb 12, 2006, 1:43:07 PM2/12/06
to
In article <IpKHf.14391$rH5....@newsread2.news.atl.earthlink.net>, Richard Lamb
says...

>
>This is probably best discussed off-line?

Hi Richard
Makes no difference to me except it probably wouldn't be of interest to many
here on RAH. Drop me a note at < chuck...@adelphia.net> with any info you
have.

See ya
Chuck S

Richard Lamb

unread,
Feb 12, 2006, 5:13:51 PM2/12/06
to
Mr. Slusarczyk,

When you indicated that you had not heard of this alleged incident
I had a horrible feeling I had done something very wrong.

I checked the local news paper for accidents at the field in question
and turned up nada.

So I went back out to the field this afternoon and had a short
conversation with the fellow who had told me about this.
Kind of a very polite version of WTF, over?

We quickly got to the heart of it, and I got a different read on the story.
No, it didn't happen HERE, and this particular aircraft was not involved.

So I took the (well known) name he gave me and did a quick google search
(something I had neglected to do before relating this story to this group).

Unfortunately, even that turned up negative. SAM died of cancer.

I widened the search, adding the term "ultralight" and finally found something
that may be related. "John", as it turns out.

The story I found on NWA news had no details about the cause of the accident -
only that an investigation was underway - and I found no follow up that shed
any additional light to the subject.

I fear I may have propagated an unfounded rumor as an alleged fact.

My only feeble defense is that the level of detail described to me gave me
the impression that I was hearing a first hand account from an eye witness.
I'm convinced this fellow did not intentionally misled me.
But I should have done my homework before bringing it up in public.


Sir, please, if you can, accept my sincere apology for any negative publicity
or embarrassment that I may have caused you by posting what may turn out to be
misleading information.

It was certainly not my intention to discredit you or your machine.


Sincerely,

Richard Lamb


Richard Lamb

unread,
Feb 12, 2006, 5:16:33 PM2/12/06
to
Since I brought it up here, I thought it best to correct it here.

Again. my apologies.

Richard

Monty Graves

unread,
Feb 12, 2006, 6:39:49 PM2/12/06
to
I was going to let this slide, but every time I read it. It just
pisses me off more. I am not one of the Canadian group. But I bought
a set of plans from Richard with his recomendations that it was for a
first time builder, and a good choice for a direct drive VW engine.
Come to find out its not either...!!!!!!!!

The problem isn't the Canadian group that Richard blames for all his
problems. Its Richard himself. He is his own worst enemy. And this
recent post is another example of how Richard's view of reality is
shall we say cloudy at best..........

For those that don't know Richard,,,,, He is on mental disabilty for
his service in Vietnam. And has been for years.

Richard didn't DESIGN the Texas Parasol, Chuck Beason did along with
a group of other builders flyers in the San Antonio area of Texas.....
Richard wasn't one of them. Since Richard was on disabiltiy he
started hanging around Chuck's shop and Chuck took a liking to him and
let him help build complete planes to sell to people. Somewhere along
the line. Richard became computer literate, and did CAD drawings from
measurements of Chucks planes. I don't know if Richard screwed Chuck
B. But Chuck B. told me over the phone that Richard did. And several
people in the area I talked to personally on the phone feel Richard
screwed Chuck... For the first few years Richard sold plans, those on
the internet he let everyone believe he was the one that designed the
plane. When in fact there wasn't a single part of the plane that
Richard designed. And to top it off, when the spar debate came up.
Richard sold the rights to the Texas Parasol to Ted F. of sirus
aviation. All in all a sick joke kind of thing.


Richard in January of this year on this newsgroup said he calculated
the max gross load of 650 lbs. and the thread is in the archives.

Richard's problem and the problem with the spars is Chuck designed the
plane to be buillt light using the smaller 2 stroke engines.......
Richard wanted the Texas Parasol to be all things to all
builders........

And thats were the controversy comes in.

The plans, call for 2 inch by .058 inch thick front tube spars.....
Herbert Beaujon a ligitimate designer says this wing is a 500 lb max
gross wing..... NOT 650 lbs that Richard first claimed to his plans
holders, and future VW engine users. THATS a HUGE difference.... When
confronted with this information. Richard would not reveal how he came
about with the 650 lb number. But INSTEAD showing how he came up with
the number Richard reduced the max gross weight of the Texas Paraso to
600 lbs instead.... Also Richard refused to do a wing load test.
And he gave the exact same response as above. ", but there are
several dozen of theseplanes _flying_ for over 20 years now. Doc, HARR


who has been the test pilot on almost all of these, had over 650 hours
on his "Lucky Lady" when the airfield changed hands and he quit.
Doc loved to play acro with it. Loops (well, tall skinny ones),
spins, rolls. I'll trust my life to his test work because I've

seen what he can do with it" END QUOTE. WTF kind of answer is that
when there is a legimate saftey concern. When I personally asked how
Doc Harr's wings were built. Richard WOULD NOT reply, I assume
because he didn't know how Harr's wings were built.....

Come to find out. Several of Chuck B.s orginal birds had been built
with a longer wing span, and shorter chord. Other pilots had noticed
the "gulling" of the front spar and had added flying wires to that area
of the front spar much like the king post ULS do.....which adds
signicant strenth to the spar. Richard never mentioned this to any of
his builders. He also never mentioned untill pressure was put on him.
That just maybe, not a single airplane had ever been built with the
wing design that was called for in the plans... All were
different.....Richard didn't like Chucks longer wing. He didn't like
the flying wires either, hence he didn't tell anyone.

So not being able to get any strait answers from Richard, about
exactly how Doc Harrs wings were built, and not being able to get
Richard to do a load test on his own wings, to confirm his 600 lb mas
gross weight.

The Candian group set up there own spar load test, under the direction
of an aero engineer. The aero engineer said the wing is a 500 lb max
gross wing. but if its loaded to 2 G's assuming a 600 lb max its going
to fail........... And low and behold it failed at 2 G's... and
PROVED the wing shown in Richard's Texas Parasol drawings are 500 lb
max gross

Confronted with this knowlege. Richard threw up his hands discussed
and just disappeared from the internet until reappering here a couple
of months ago. spouting the same bullshit about "his design" grossing
650 lbs

What this dirty laundry is all about is a guy that took someone elses
design for a lite 500 lb gross plane using light rotax engines and,
claiming it as his own, and marketing it to a group of folks claiming
it now can use a VW engine, 12 gal of fuel and fly a 200 lb pilot on
the same 500 lb gross wing........


And just another fact Richard never built a plane from his plans as he
said...... he built a longer plane by one bay., and maybe with these
wings, maybe with different wings..... I have documented persoanl
emails saying he built it two different ways. I think it makes a
difference as to what he has had to smoke, drink or medications he is
was taking at the time a question was asked......

For reference on why the 2 inch tube spars are a concern

Chuck Slusarczyk CSG HAWK uses 2.25 inch front spar, and so does the
Rans S-4. Both high wing single place planes..... Chuck S. can
verifiy his own max gross on those 2.25 inch spars, and how that was
derived..... Both planes use max Rotax 503 engines weights some 60 lbs
lighter than the direct drive VW.

BTW Chuck S. I think I would stay as far away from Richard Lamb as
possible... In doing research on Richard I came accross a post
referring to HIS LATEST DESIGN posted to the Romance Chat newsgroup. in
January of 2006. Sounds a lot like to me one of Bruce Kings BK 1.1
also from the San Antonio TX area.
Richard seems to use that word DESIGN rather loosely

Monty Graves

Quote from Richard.
"I haven't cut any metal on the latest design.
But I've done a whole lot of drawing on it.
(Not much else to do at the moment)


Some of the sketches are posted at
http://www.home.earthlink.net/~cavelamb/


I've got two details left to clean up.


One is the attachment of the landing gear.
I was trying to work it out so that the legs
stay on the airframe if the wings are removed.
It would make things a lot easier to live with.


But there turns out to be a lot of mechanical
complexity in doing it that way. The center
part of the wing would have to protrude far
enough outside the airframe that we'd have a
challenge in making the wing pieces line up
exactly right (OK, not all that hard, but a
lot of extra complication).


So it looks like the gear legs will be under
the wing, and a simple plywood cradle for the
fuselage solves the "move it" problem.


The other challenge involves the design of the
tail. This one ain't no baby buggy! As is, I
am projecting 160 MPH cruise (on 4 gallons per
hour fuel burn!)


I'm working out how to build the tail so that the
stabilizer (normally the fixed surface) can be
adjusted (in flight) to provide longitudinal trim
force without a lot of extra drag to slow it down.


This is not unusual on larger aircraft, but on one
so small (and it is TINY), it is a real challenge.
(Extra weight that far back is a killer!)


Estimated cost for the thing is about 4 grand.
(except for the bubble canopy - might be another
500-800 or so there. No valid quotes yet.)
Not too bad though.


Well, enough of that.


Let's go find a sucker to sploosh!


Richard

ChuckSlusarczyk

unread,
Feb 12, 2006, 6:52:24 PM2/12/06
to
In article <zqOHf.14513$rH5....@newsread2.news.atl.earthlink.net>, Richard Lamb
says...

Richard

I appreciate the correction and your strength of character to make it. Apology
not required but accepted. I never questioned your intentions for a second.
Now back to the fun stuff :-)

See ya

Chuck S

ChuckSlusarczyk

unread,
Feb 12, 2006, 7:14:54 PM2/12/06
to
In article <1139787589.3...@g14g2000cwa.googlegroups.com>, Monty Graves
says...

>
>Chuck Slusarczyk CSG HAWK uses 2.25 inch front spar, and so does the
>Rans S-4. Both high wing single place planes..... Chuck S. can
>verifiy his own max gross on those 2.25 inch spars, and how that was
>derived..... Both planes use max Rotax 503 engines weights some 60 lbs
>lighter than the direct drive VW.


That's correct,our Single seat Hawk has 2 1/4" dia spars and our 2 seaters have
a 3" dia spar. Back when the original was designed we first designed the wing
and sized the spars based on a plus 6 g ultimate load factor.We then
substantiated the designed wing with actual load tests.We used a lift
distribution for a rectangular wing to determine the loads spanwise .Then since
it was a 2 spar wing the load was divided chordwise at 70% on the front spar and
30% on the rear spar. After the sandbag tests we found we were within about 3%
of the designed load.

In my humble opinion the reason we don't see many wing failures with wings that
I consider under designed. Is the fact that the light wing loading and inability
of these planes to maintain energy prevents them from pulling more then about 3
g's before they stall. But that's just my opinion I don't want to start a "spar
wars" episode :-)

See ya

Chuck S

Richard Lamb

unread,
Feb 12, 2006, 10:03:10 PM2/12/06
to
ChuckSlusarczyk wrote:
> In article <1139787589.3...@g14g2000cwa.googlegroups.com>, Monty Graves
>
> In my humble opinion the reason we don't see many wing failures with wings that
> I consider under designed. Is the fact that the light wing loading and inability
> of these planes to maintain energy prevents them from pulling more then about 3
> g's before they stall. But that's just my opinion I don't want to start a "spar
> wars" episode :-)
>
> See ya
>
> Chuck S
>
Yep.
And I have pointed that out before, but nobody wanted to hear (or believe) it.
That's not how "real" airplanes work, so it was just another "lie".

Bottom line is that I'm simply sick and tired of the bickering and back
biting, and do not wish to play any more...

I'm not going to go into a long tirade against Chuck Beeson.
He was once my best friend. Those days are long past.
I respect him for what he has accomplished.
It's a hell of a piece of work he's done.

But ALL of us (Me, Sonny, Doc, Paul, Charlie, the Longs, his wife, and every
body else) have all walked away from him. And for good reason. Since I'm the
only one of the bunch who is an "emotionally disturbed Viet Nam Vet", maybe,
just maybe, there's more to it than "me".

Whether that fits with anybody's close held beliefs is simply not important
to me any more.

Now that's about as plain as I can put it folks.

Richard

Stealth Pilot

unread,
Feb 12, 2006, 11:56:35 PM2/12/06
to
On Sat, 11 Feb 2006 15:27:15 -0500, clare at snyder.on.ca wrote:

>On Sat, 11 Feb 2006 15:14:18 GMT, Richard Lamb
><cave...@Xearthlink.net> wrote:
>
>>Rec.Aviation.Homebuilt special...
>>
>>Eventually, these will be made available to everybody on Matronics, but for
>>the time being they are posted at http://www.home.earthlink.net/~tp-1/ just
>>for the nice boys and girls of RAH and RAU.
>>
>>Enjoy...
>>
>>Richard
>>
>>
>>
>>
>Got the wing spar design updated yet? Or are we trying to thin the
>herd?


clare you and vdubber have performed a service here beyond
calculation. I wonder if we'd ever realise how many lives you have
saved.
pat on the back boys.

gods aint the internet great!

Stealth Pilot
Australia

Jean-Paul Roy

unread,
Feb 17, 2006, 9:29:05 AM2/17/06
to
Monty, I'm glad you made this intervention here. The Texas Parasol always
attracted me as a project after my actual plane (CH701) is finished, just
for the sake of enjoying building something.

I downloaded the plans from Matronic last week and I was just going to send
Richard $30.00 for his CD or DVD.

As of NOW the files are DELETED and you just saved me $30.00.

Thanks
Jean-Paul (back to the 701 construction) Roy


"Monty Graves" <mgr...@usmo.com> wrote in message
news:1139787589.3...@g14g2000cwa.googlegroups.com...

clareatsnyder.on.ca

unread,
Feb 17, 2006, 10:07:47 PM2/17/06
to
On Fri, 17 Feb 2006 09:29:05 -0500, "Jean-Paul Roy"
<jean-pa...@sympatico.ca> wrote:

>Monty, I'm glad you made this intervention here. The Texas Parasol always
>attracted me as a project after my actual plane (CH701) is finished, just
>for the sake of enjoying building something.
>
>I downloaded the plans from Matronic last week and I was just going to send
>Richard $30.00 for his CD or DVD.
>
>As of NOW the files are DELETED and you just saved me $30.00.
>
>Thanks
>Jean-Paul (back to the 701 construction) Roy
>

And just to make things clear, I don't have a pony in this race.
I did not buy plans. I did not invest ANY time or money in the
project, and there was no reason for my Friend Gary to say anything
negative about the plane EXCEPT the fact he is VERY dedicated to SAFE
flying. He, and a lot of the other fellows, were hoping it would be a
good, cheap plane, and easy to build and fly. The building method is
VERY interesting, but to get proper hole edge clearances the longerons
in the cabin area should be 1" angle.
If built "inside out" with the flats of the angles in, instead of out,
you would not have issues with the fabric at the rivet heads, and you
would also have a smoother interior. Lots of other little
"improvements" that would make it a better plane - .

Really just needs a good designer to go ever it and fix the little
details - the ones that make the current rendition difficult to build
and less than adequate structurally. As they say, the devil's inthe
details - and they will kill you.

Richard Lamb

unread,
Feb 18, 2006, 1:35:38 AM2/18/06
to
clare at snyder.on.ca wrote:
>>
>
> And just to make things clear, I don't have a pony in this race.
> I did not buy plans. I did not invest ANY time or money in the
> project, and there was no reason for my Friend Gary to say anything
> negative about the plane EXCEPT the fact he is VERY dedicated to SAFE
> flying. He, and a lot of the other fellows, were hoping it would be a
> good, cheap plane, and easy to build and fly. The building method is
> VERY interesting, but to get proper hole edge clearances the longerons
> in the cabin area should be 1" angle.
> If built "inside out" with the flats of the angles in, instead of out,
> you would not have issues with the fabric at the rivet heads, and you
> would also have a smoother interior. Lots of other little
> "improvements" that would make it a better plane - .
>
> Really just needs a good designer to go ever it and fix the little
> details - the ones that make the current rendition difficult to build
> and less than adequate structurally. As they say, the devil's inthe
> details - and they will kill you.

Your loyalty to your friend is noted and admired, Clare.

But for just a moment, stop and reread what you wrote above.
Sure, the inside-out approach has some interesting merit, but is that
really and improvement, a radical modification, or a new design?

The first step in designing something like an airplane is
to carefully define the (dreaded) Mission Requirements Statement.

This one, as it is, fulfills the mission requirements set out for it.
And it has done so safely for many years.

As far as I could tell from the photos and article published in the Canadian
Recreational Aviation magazine the only thing the Canadian projects have in
common with this one is that they both use extruded aluminum angle for the
fuselage truss.

You can claim these were only improvements if you want.
But what it really was is a completely new, unproved, and much heavier
machine. (I'm curious why they didn't go to a 2-1/4" front spar as we
discussed repeatedly. I know it an expensive piece of tube, but it would
have solved the problem quite adequately.)


My friend, Al Robinson is doing exactly the same thing! But man, what a
difference in attitudes.

His Texas Pete is a two-seat side by side with a Geo Metro of power.
Gross weight will be right about 900 pounds.
(His pics and details are posted on the Texas Parasol group at Yahoo Groups)
(as are reports of some of those who finished and have flown their (real)
Texas Parasols)

The modifications he has made to the wing structure were supervised and
blessed by none other than the late Lt. Graham Lee. I don't think Graham
had a degree, but he was one hell of an engineer.

Al is getting close to being ready to static test his wing - and I intend to
be there to help when he does. He kindly invited me, and I wouldn't miss is.
If it holds ok, we'll cover it and go flying.

If it doesn't look safe to BOTH of us, we'll come up with something else.
Most likely (if necessary!) an I beam main spar built up using extruded
aluminum angle front and back of an aluminum sheer web.
At least that's our fall back plan.
That type construction allows us to custom tailor the load factor allowance
to what ever the builder desires.

I've got the thing drawn up, but I've not built it and tested it yet, so
it hasn't been published. And it's not going to be unless it IS tested.

I have personally flown both of my parasols (well duh!) and several of the
others.

Doc has flown damned near all of them and scared the pee outta me several
times in the process.

Sonny is building his FOURTH original design based on this stuff.

Paul Hammond flew his every weekend for years.

Doc is home taking care of the kids.

I'm sitting here trying to be patient and not pull my hair out.


So if is possible, can we call a truce and go make fun of milli-amp
for a while?

Richard

Stealth Pilot

unread,
Feb 18, 2006, 9:07:45 AM2/18/06
to
On Sat, 18 Feb 2006 06:35:38 GMT, Richard Lamb
<cave...@Xearthlink.net> wrote:


>
>I have personally flown both of my parasols (well duh!) and several of the
>others.
>
>Doc has flown damned near all of them and scared the pee outta me several
>times in the process.
>
>Sonny is building his FOURTH original design based on this stuff.
>
>Paul Hammond flew his every weekend for years.
>
>Doc is home taking care of the kids.
>
>I'm sitting here trying to be patient and not pull my hair out.
>
>
>So if is possible, can we call a truce and go make fun of milli-amp
>for a while?
>
>Richard

either take the plans off the internet
or
publish all the details relating to their structural deficiencies
along with the plans
or
get the design structurally fixed

then you can have a truce.
another death from those plans and you will be crucified.

Stealth Pilot

Richard Lamb

unread,
Feb 18, 2006, 11:45:08 AM2/18/06
to

> either take the plans off the internet
> or
> publish all the details relating to their structural deficiencies
> along with the plans
> or
> get the design structurally fixed
>
> then you can have a truce.
> another death from those plans and you will be crucified.
>
> Stealth Pilot

HOW DARE YOU SUGGEST THAT A.N.Y.O.N.E HAS DIED AS A RESULT OF
ANY STRUCTURAL PROBLEMS

I RESENT THE HELL OUT OF THAT, YOU BLOODY STALKING MORON!

POST YOUR LONG LIST OF THOSE WHO HAVE DIED FROM _ANY_
PROBLEMS RELATED TO THE STRUCTURAL INTEGRETY OF THE PLANE.

MAY I KINDLY SUGGEST YOU GO PISS UP A ROPE.

fredf...@spamcop.net

unread,
Feb 18, 2006, 4:13:54 PM2/18/06
to

Jean-Paul Roy wrote:
> Monty, I'm glad you made this intervention here. The Texas Parasol always
> attracted me as a project after my actual plane (CH701) is finished, just
> for the sake of enjoying building something.
>
> I downloaded the plans from Matronic last week and I was just going to send
> Richard $30.00 for his CD or DVD.
>
> As of NOW the files are DELETED and you just saved me $30.00.
>

The Sky Pup:

http://www.skypup.net/

is a scratch built ultralight. Pretty good plans are available and it
has a good safety record. There are several flying, several more
under construction, and an active builders/users/ club (note, not
a group, a club on Yahoo:

http://groups.yahoo.com/group/Skypup-club/

First time builders claim to have comkpleted theirs within the
400 hour estimare. Cost depends mostly on how much you pay
for the engine and prop.

The dry weight of the prototype was 200 lbs and it was designed
for 400 lbs gross, by real engineers.

--

FF

clareatsnyder.on.ca

unread,
Feb 18, 2006, 4:31:42 PM2/18/06
to
On Sat, 18 Feb 2006 06:35:38 GMT, Richard Lamb
<cave...@Xearthlink.net> wrote:

Richerd,
You are getting ahead of yourself.
Forget the pictures you saw in the Rec Av magazine. Those were
projects some guys built. They were NOT the tests done by Gary. Gary
did the tests on a wing BUILT ACCORDING TO PLANS, and IT FAILED THE
TEST.

The other improvements I noted should/could be made HAVE NEVER BEEN
IMPLEMENTED to the best of my knowledge.

As for the extruded aluminum truss construction, using the sizes
listed in the plans, and the rivets specified, it is difficult, if not
impossible, to achieve proper hole edge clearances in MANY locations.
I know, for an ultralight there are no inspections, so you can get
away with it - but you are NOT building to acceptable aircraft
standards if proper edge clearances can not be maintained.


>
>You can claim these were only improvements if you want.
>But what it really was is a completely new, unproved, and much heavier
>machine. (I'm curious why they didn't go to a 2-1/4" front spar as we
>discussed repeatedly. I know it an expensive piece of tube, but it would
>have solved the problem quite adequately.)

Then put the D@%&D thing in the plans, already.


>
>
>My friend, Al Robinson is doing exactly the same thing! But man, what a
>difference in attitudes.
>
>His Texas Pete is a two-seat side by side with a Geo Metro of power.
>Gross weight will be right about 900 pounds.
>(His pics and details are posted on the Texas Parasol group at Yahoo Groups)
>(as are reports of some of those who finished and have flown their (real)
>Texas Parasols)
>

The Texas Parasol is just an angle aluminum Baby Ace in concept -
shortened and with bigger tailfeathers to compensate.
Apparently you built yours one station longer, making it the same
dimensions as the Ace. The wings are the same dimensions and planform
as well, from what I understand. So the CONCEPT is a good one.
The plane CAN be built as a safe, economical, fun-to-fly plane - but
NOT as per plans.
Fix the inadequacies (which means admitting to them first) and make
the plans accurate enough to build from, and you'll have all kinds of
support.


>The modifications he has made to the wing structure were supervised and
>blessed by none other than the late Lt. Graham Lee. I don't think Graham
>had a degree, but he was one hell of an engineer.

Then document the modifications and put them in the plans.


>
>Al is getting close to being ready to static test his wing - and I intend to
>be there to help when he does. He kindly invited me, and I wouldn't miss is.
>If it holds ok, we'll cover it and go flying.
>
>If it doesn't look safe to BOTH of us, we'll come up with something else.
>Most likely (if necessary!) an I beam main spar built up using extruded
>aluminum angle front and back of an aluminum sheer web.
>At least that's our fall back plan.
>That type construction allows us to custom tailor the load factor allowance
>to what ever the builder desires.
>
>I've got the thing drawn up, but I've not built it and tested it yet, so
>it hasn't been published. And it's not going to be unless it IS tested.

That sounds like a good idea.


>
>I have personally flown both of my parasols (well duh!) and several of the
>others.

But tell everyone how you built yours. It was not strictly to plans.
What about the leave in the spar??


>
>Doc has flown damned near all of them and scared the pee outta me several
>times in the process.
>
>Sonny is building his FOURTH original design based on this stuff.
>
>Paul Hammond flew his every weekend for years.
>
>Doc is home taking care of the kids.
>
>I'm sitting here trying to be patient and not pull my hair out.
>
>
>So if is possible, can we call a truce and go make fun of milli-amp
>for a while?

A FULL truce is as close as the corrections to the plans. Just because
nobody's killed themselves YET does not mean the plans, as they exist
today, are safe. By your admission, many builders ARE making changes.
>
>Richard

Morgans

unread,
Feb 18, 2006, 6:50:51 PM2/18/06
to

<fredf...@spamcop.net> wrote

> The dry weight of the prototype was 200 lbs and it was designed
> for 400 lbs gross, by real engineers.

I could not fly it, and I'm by no means, a lard butt. Under 170 lbs for
the pilot, it looks like to me.
--
Jim in NC

Richard Lamb

unread,
Feb 18, 2006, 7:25:35 PM2/18/06
to
clare at snyder.on.ca wrote:

Clare, thanks for taking a reasonable position and tone.
Muchly appreciated.
I'll reply as best I can, ok?

> Richerd,
> You are getting ahead of yourself.
> Forget the pictures you saw in the Rec Av magazine. Those were
> projects some guys built. They were NOT the tests done by Gary. Gary
> did the tests on a wing BUILT ACCORDING TO PLANS, and IT FAILED THE
> TEST.

That was not my take from the article, but if I'm wrong on that point,
I'll offer an apology.

The Test:

I've never been offered any description of the set up - only the conclusions.

As I understand it, it was assumed that the front spar would take 100% of the
load. That is, of course, true IF the spar is located at the center of
pressure - as a normal wing is arranged.

But the wing in question is obviously NOT arranged that way.
The spars are at the leading and trailing edges while the center of pressure
remains back around 30% to 40% of the chord.

The airfoil used is a Clark-Y. There is a minor deviation at the leading edge
due to the 2" diameter of the leading edge being a bit larger than the radius
shown in the tables.

At zero degrees AoA the CP is at 40%.
At 12 degrees AoA it has moved forward to 30%

In my work, I've ASSUMED that the front spar would only receive 70% of the
total load. That makes a tremendous difference in the amount of deflection
of the spar, and, seems to be much closer to what we observe in the actual
structure in flight.

I think this also explains why Beeson could build a wing using .035 wall spars
tubes. My crude work on those indicates they wouldn't survive 2 G's at 100%
load. But they do - although with noticeable deflection under load in flight.

Just to be through, the wing structure should also be mounted at, in this
case, 12 degrees nose down (it's inverted, remember) to account for the
angle of attack at the test load. This will have the effect of pulling the
load vector forward some, but the magnitude is tiny compared to a 30% offset.


> The other improvements I noted should/could be made HAVE NEVER BEEN
> IMPLEMENTED to the best of my knowledge.

Hey, I'm not saying it's a bad idea.
Like you pointed out there are some real benefits to it.

Rivet heads under the fabric are a pain to work around.
On my first parasol (PBJ) I use a thin piece of wood molding on the
outboard side of the top longeron to avoid that.

On the second one I just used flush rivet - and avoided the problem
without adding the extra weight.


> As for the extruded aluminum truss construction, using the sizes
> listed in the plans, and the rivets specified, it is difficult, if not
> impossible, to achieve proper hole edge clearances in MANY locations.
> I know, for an ultralight there are no inspections, so you can get
> away with it - but you are NOT building to acceptable aircraft
> standards if proper edge clearances can not be maintained.


The edge margins for thin sheet are well known and easily found.
I'm not having much luck locating the tables for Lugs on short notice.

The difference, as I recall, is the relative thickness of the metal being
bolted or riveted. Those extrusions are five times thicker that .025 sheet
metal.

But I'm still looking for it and will post it here when I find it.

Examining the results of some of the accidents, even the one straight in
on the nose fatality, you do not find broken rivets. The angles bent, but
seldom even break. In the bad one, the first bay of the fuselage truss was
crumpled up like an accordion.

>
> The Texas Parasol is just an angle aluminum Baby Ace in concept -
> shortened and with bigger tailfeathers to compensate.

It's a Texas airplane, guys. Texans just GOTTA have a lot of tail!

> Apparently you built yours one station longer, making it the same
> dimensions as the Ace. The wings are the same dimensions and planform
> as well, from what I understand. So the CONCEPT is a good one.
> The plane CAN be built as a safe, economical, fun-to-fly plane - but
> NOT as per plans.

It's still quite a bit smaller than a Baby Ace.

The added 15 inches was for weight and balance purposed only.
A full dressed 2180 VW can weigh well over 200 pounds.
Compared to a light weight 2 stroke, that can create a - situation!

It was nearly TOO much extra tail arm as it was tough getting the CG
far enough _forward_ to stay in an acceptable range. I think we've
got it worked out though. Just had to come up with a lighter tail wheel
and leg.

> But tell everyone how you built yours. It was not strictly to plans.
> What about the leave in the spar??

I don't know where that rumor came from.
The wing on the new plane was built mainly according to the plans,
with the exception of boxing in the compression struts at the strut
attach bay. My reason for that was that the struts on this plane angle
in a bit at the fuselage end, creating a small compression load at the spar
ends. Probably not really necessary, but it made me feel better.

The internal sleeves are as per plans.

I also added a short .017 sheet metal cover over the leading edge.
I'd not think that's a structural thing - purely cosmetic.

> A FULL truce is as close as the corrections to the plans. Just because
> nobody's killed themselves YET does not mean the plans, as they exist
> today, are safe. By your admission, many builders ARE making changes.

Well, I think that was more of an accusation than an admission, Clare.

I have had one friend who was killed in his.
It was a hard loss to deal with.
And it gave me pause to seriously reconsider.

Please believe me folks, if we had ANY experience that indicated there were
real structural problems with the thing, I'd be the first one to address them.

But the physical evidence simply does not support that claim.

At least - as shown in the plans - and not "improved" to be faster, heavier, etc.


Mission Requirements:

It is intended to be a very inexpensive, very light (ALmost UL), very slow,
SINGLE seat baby buggy.

Hand tools only - with very little welding (the main gear axle clusters) and
little if any machine work needed.

Relatively tolerant of "workmanship" issues.

Protection for the pilot in the advent of an accident. (be real, folks!)

Good flying qualities, and fun to fly.


If that's not what you want, this is not the airplane to build.

And I'd appreciate it if people would NOT try to make it into something it was
never intended to be.
There in lies the real danger.


Richard

Peter Dohm

unread,
Feb 18, 2006, 7:38:45 PM2/18/06
to
"Morgans" <jsmo...@chJUarNKer.net> wrote in message
news:JwNJf.3468$4v7....@fe04.lga...
I admit the same problem, due mostly to height. It would need to be scaled
up, which is not always a good idea, since it is difficult to scale myself
down. I am making some progress on the width and gross weight part of the
problem. :-)

None the less, an interesting design--even though it apears to be so
"traditional" as to use a tail skid to double as a brake.

Peter


fredf...@spamcop.net

unread,
Feb 18, 2006, 8:19:41 PM2/18/06
to

Peter Dohm wrote:
> "Morgans" <jsmo...@chJUarNKer.net> wrote in message
> news:JwNJf.3468$4v7....@fe04.lga...
> >
> > <fredf...@spamcop.net> wrote
> >
> > > The dry weight of the prototype was 200 lbs and it was designed
> > > for 400 lbs gross, by real engineers.
> >
> > I could not fly it, and I'm by no means, a lard butt. Under 170 lbs for
> > the pilot, it looks like to me.

To be within design gross with 5 gallons of fuel, yes.

> > --
> > Jim in NC
> >
> I admit the same problem, due mostly to height. It would need to be scaled
> up, which is not always a good idea, since it is difficult to scale myself
> down. I am making some progress on the width and gross weight part of the
> problem. :-)
>

The designer was tall ans skinny, over 6' IIRC. Evidently Height is
not the problem that, er, girth is.

> None the less, an interesting design--even though it apears to be so
> "traditional" as to use a tail skid to double as a brake.
>

>From what I read on the Yahoo club it flies fine with pilots up to 200
lbs
(not sure how much fuel, probably less than 5 gallons). A heavier
pilot erodes the factor of safety. The stress analysis on the airframe
was done by finite element analysis by competent engineers so
you can be confident in the factor of safety, you know how much you
have left if you fly over gross.

--

FF

Richard Lamb

unread,
Feb 18, 2006, 9:57:31 PM2/18/06
to
Not the crumbling yellowed pages I remembered, but give this a try...

Try here: http://trs.nis.nasa.gov/archive/00000254/01/asm-B200.pdf

clareatsnyder.on.ca

unread,
Feb 19, 2006, 3:30:50 PM2/19/06
to
On Sun, 19 Feb 2006 00:25:35 GMT, Richard Lamb
<cave...@Xearthlink.net> wrote:

>clare at snyder.on.ca wrote:
>
>Clare, thanks for taking a reasonable position and tone.
>Muchly appreciated.
>I'll reply as best I can, ok?
>
>> Richerd,
>> You are getting ahead of yourself.
>> Forget the pictures you saw in the Rec Av magazine. Those were
>> projects some guys built. They were NOT the tests done by Gary. Gary
>> did the tests on a wing BUILT ACCORDING TO PLANS, and IT FAILED THE
>> TEST.
>
>That was not my take from the article, but if I'm wrong on that point,
>I'll offer an apology.
>
>The Test:
>
>I've never been offered any description of the set up - only the conclusions.

The wing was supported as it would be on the plane, but inverted. It
was then loaded with sandbags to iminent failure, with the load
properly distributed. It was tested both with and without the jury
strut. The jury strut is VERY CRITICAL (and it's attatchment according
to the plans was not sufficient) and even with the jury strut upgraded
the wing did not meet the specified G rating for the design specified
gross weight. IIRC it is a 500 lb wing, not a 600 - but my memory of
that detail may be fuzzy. What I DO know, is it did NOT meet the
published spec, or the required G load for the published maximum gross
weight.


>
>As I understand it, it was assumed that the front spar would take 100% of the
>load. That is, of course, true IF the spar is located at the center of
>pressure - as a normal wing is arranged.

There were mo assumptions made. The wing was properly supported and
loaded.


>
>But the wing in question is obviously NOT arranged that way.
>The spars are at the leading and trailing edges while the center of pressure
>remains back around 30% to 40% of the chord.
>
>The airfoil used is a Clark-Y. There is a minor deviation at the leading edge
>due to the 2" diameter of the leading edge being a bit larger than the radius
>shown in the tables.

Which also causes aproblem when a larger spar diameter is used. IIRC
it makes the stall more abrupt -which you do NOT want.


>
> At zero degrees AoA the CP is at 40%.
> At 12 degrees AoA it has moved forward to 30%
>
>In my work, I've ASSUMED that the front spar would only receive 70% of the
>total load. That makes a tremendous difference in the amount of deflection
>of the spar, and, seems to be much closer to what we observe in the actual
>structure in flight.
>
>I think this also explains why Beeson could build a wing using .035 wall spars
>tubes. My crude work on those indicates they wouldn't survive 2 G's at 100%
>load. But they do - although with noticeable deflection under load in flight.

And Chuck BUILT the .035 wing, but did not fly it, from what I
understand, as the wing was too flimsy even for HIS level of comfort.

1.5 times the hole diameter rings a bell.

*** Free account sponsored by SecureIX.com ***

clareatsnyder.on.ca

unread,
Feb 19, 2006, 3:34:35 PM2/19/06
to

I wasn't tlking lugs. I was talking fastening uprights etc to
longerons, and the firewall area as well. Any place 2 aluminum parts
are joined by drilling and riveting or bolting.Specifically the
extruded members of the truss. Up here we are encouraged (required) to
maintain the same edge margins as on sheet.

Richard Lamb

unread,
Feb 21, 2006, 4:45:44 PM2/21/06
to
Clare,

You seem to be pretty well connected up there.
And you concerns are (believe it or not) well taken.

May I ask if it might be possible to collect whatever information
IS available from that test? Initial assumptions, load distribution,
deflections noted (preferably at any given load, if possible), etc.

Last night there was a message on the Texas Parasol list that indicated
that the spar tube in question had actually kinked.

The only thing I'd heard before is that the rig was in danger of
imminent collapse - but no details as to what was meant by that.


I've never been offered even a scrap of this kind of information - only
the final conclusion. And that rather loudly...

If you really believe there is an issue here, what would it hurt?


Thanks,

Richard

Richard Lamb

unread,
Feb 25, 2006, 8:32:14 AM2/25/06
to

All right, back to the beheadings!


Four days, and not a mumblin' word.
Makes a body wonder, don't it?

Seriously, if the intent is to protect the innocent and unwary, why NOT
respond (and show your work?).

Or? Have I misinterpreted obviously altruistic motives?
(huh!)


I've been trying to raise the RAA web site for several days to see if they
post their articles on the web. But it seems to be snowed under and won't be
back up until the spring thaw.

Fortunately (or otherwise, depending on your point of view?), I just happen
to HAVE a copy of the magazine. It was sent to me by a Canadian fellow (who
I'll not name to avoid allegations of international espionage).

But here it is, and we'll let the reader decide for him/her/it self...

http://www.home.earthlink.net/~tp-1/
page-1.jpg to page-6.jpg

I particularly like the ripe irony of the last paragraph of the article.

There are also a few of the stress analysis reports in pdf format.

ul-spar.pdf The original preliminary report (corrected! The original
original showed an 8 G limit(!), but it was a simple error)

ul-redo.pdf Suggested upgrade to 4.4 Gs at 650 lbs

fastner.pdf rivets, bolts, etc

For what it's worth, none of these are my own works.
They were posted to the Fly5k list by the author and are presented here
for enlightenment and entertainment of (any?) interested readers...

Richard

Robert Schieck

unread,
Feb 25, 2006, 11:29:26 AM2/25/06
to
Richard Lamb wrote:
> I've been trying to raise the RAA web site for several days to see if they
> post their articles on the web. But it seems to be snowed under and
> won't be
> back up until the spring thaw.

It works for me .....

And they don't post their articles.

Rob


clareatsnyder.on.ca

unread,
Feb 25, 2006, 5:36:34 PM2/25/06
to
On Sat, 25 Feb 2006 13:32:14 GMT, Richard Lamb
<cave...@Xearthlink.net> wrote:

>
>All right, back to the beheadings!
>
>
>Four days, and not a mumblin' word.
>Makes a body wonder, don't it?
>
>Seriously, if the intent is to protect the innocent and unwary, why NOT
>respond (and show your work?).

I'm not involved, so I have no work to show.
As for the "Canadian project" - it is basically DEAD.


>
>Or? Have I misinterpreted obviously altruistic motives?
>(huh!)
>
>
>I've been trying to raise the RAA web site for several days to see if they
>post their articles on the web. But it seems to be snowed under and won't be
>back up until the spring thaw.

>
>Fortunately (or otherwise, depending on your point of view?), I just happen
>to HAVE a copy of the magazine. It was sent to me by a Canadian fellow (who
>I'll not name to avoid allegations of international espionage).
>
>But here it is, and we'll let the reader decide for him/her/it self...
>
>http://www.home.earthlink.net/~tp-1/
> page-1.jpg to page-6.jpg
>
>I particularly like the ripe irony of the last paragraph of the article.

The artical was written by a Canadian who was looking for a good,
cheap, safe way to get into the air. He was reporting on the early
stages of the "Canadian project" and none of the problems with the
wings had been uncovered yet, or at least the testing had not yet been
done. There is also no indication that they were building a heavier
version than yours, other than the mention of designing an all
aluminum wing.

The artical was written more about the Canadian ultralight regs than
about the Parasol.

As for the other attatchments, I'm not an engineer, so I can't speak
for their accuracy - but at least it looks like whoever wrote them did
have a very good understanding of the engineering involved. Myself ? -
it was over my head. I have no idea who wrote them, or what his
qualifications are. Neither do I particularly care.

It's YOUR design. It's up to YOU to either make it safe, and provide
accurate information, or to do one of two things:

Remove the plans from circulation, or mark them clearly as being
"proof of concept" plans requiring some extra engineering - and NOT
FOR FIRST TIME BUILDERS.

OK - 'Nuff said - I'm out here.
I've let everyone on this list know the concerns that were raised, and
the status of the plans. That's all I intended.


>
>
>
>There are also a few of the stress analysis reports in pdf format.
>
>ul-spar.pdf The original preliminary report (corrected! The original
> original showed an 8 G limit(!), but it was a simple error)
>
>ul-redo.pdf Suggested upgrade to 4.4 Gs at 650 lbs
>
>fastner.pdf rivets, bolts, etc
>
>For what it's worth, none of these are my own works.
>They were posted to the Fly5k list by the author and are presented here
>for enlightenment and entertainment of (any?) interested readers...
>
>Richard

*** Free account sponsored by SecureIX.com ***

Richard Lamb

unread,
Feb 25, 2006, 7:14:19 PM2/25/06
to

Such a shame, Clare, because the 4.4G upgrade is
exactly what you've been asking for?

Richard

clareatsnyder.on.ca

unread,
Feb 26, 2006, 12:22:56 AM2/26/06
to


I've forwarded the references to Gary, see if he says the issue is
adequately addressed. If it is, I'll pass it on back to the group.

clareatsnyder.on.ca

unread,
Feb 27, 2006, 2:28:03 PM2/27/06
to

You want more input - you got it.

OK, we are NOT engineers, but looking at this with a few more
knowledgeable guys than myself the following observations and
recommendations age just the beginning.

The last PDF with the spar calculations is a problem. On the first
page there is a basic math mistake. We did not go into the
calculations to see if the rest was correct or not. 144 pounds per
wing is 288 pounds total, not 244. The author then assumes that the
lift distribution is equal down the length of the wing, and it is
admitted that this assumption was made to simplify the math.
Unfortunately this removes some of the lift load from the inner spar
and makes the calculated G's quite optimistic. In real life the lift
is usually assumed to center about 44% out from the root. A proper
load test will have an elliptical distribution of sandbags for this
very reason. The equal distribution assumption transfers a lot of sand
out beyond the lift strut attach. This has two effects that give
optimistic conclusions. There is less sand (lift force) at the weak
point 40" out from the root. Also the equal distribution tends to
straighten the inner section of spar, like a teeter totter, so that it
would take more weight to get the spar to buckle between the root and
the lift strut attach. The conclusion is therefore optimistic. Also it
is premature to state from this spar analysis that the wing is OK. The
calculation, even if correctly done, address only the strength of the
front spar.


The testing of the wing performed at Gary's hangar addressed this
non-linear loading.

Just a little more input from another more knowledgeable than
myself. The material that the plans and the airplane use is 6061t6
correct? The calculations show the material or the identifier as being
6061. The problem in the shown calcs is that the numbers he is using
at 68000psi tensile is higher than the ultimate strength of 2024-t4
which is given at 64000. 6061t6 is only 45,000psi ultimate and 39,900
yield.

It's no wonder the wing was failing at 2g.

Curtis Scholl

unread,
Feb 27, 2006, 3:14:24 PM2/27/06
to
Hi Group:

Clare:

Looking at the calculations for the spar, spar.pdf I do not see your
interpretation. The 67000 psi value is Bearing Ultimate, not tensile.

A MATWEB lookup is in order:

6061-T6

Tensile ultimate: 45000
Tensile yield : 39000

Bearing ultimate: 88000
Bearing Yield : 56000

The values presented in the spar calculation are correct, and the
bearing ultimate of 67000 is below spec.

So where is the spar calc using bad values except, as you say, in the
initial assumptions of 244 pounds versus 288?

Curtis Scholl

clareatsnyder.on.ca

unread,
Feb 27, 2006, 4:02:51 PM2/27/06
to
On Mon, 27 Feb 2006 15:14:24 -0500, Curtis Scholl <csc...@izzy.net>
wrote:

>Clare:
>
>Looking at the calculations for the spar, spar.pdf I do not see your
>interpretation. The 67000 psi value is Bearing Ultimate, not tensile.
>
>A MATWEB lookup is in order:
>
>6061-T6
>
>Tensile ultimate: 45000
>Tensile yield : 39000
>
>Bearing ultimate: 88000
>Bearing Yield : 56000
>
>The values presented in the spar calculation are correct, and the
>bearing ultimate of 67000 is below spec.


Thjat may well be true, but is "bearing strength" the proper
charachteristic of the material to be used in this calculation?

As I said - not an engineer - just asking.
Please explain for the rest of us how the bearing ultimate and / or
yeild values impact the calculation, vs tensile.

clareatsnyder.on.ca

unread,
Feb 27, 2006, 4:29:53 PM2/27/06
to

bearing strength

The maximum bearing load at failure divided by the effective bearing
area.
In a pinned or riveted joint, the effective area is calculated as the
product of the diameter of the hole and the thickness of the bearing
member. ASM, 1

How does this impact the failure of the wing structure by buckling and
failure of the main spar???

Curtis Scholl

unread,
Feb 27, 2006, 4:38:44 PM2/27/06
to
Clare:

Bearing properties are used when designing mechanically fastened joints.
The purpose of a bearing test is to determine the the deformation of a
hole as a function of the applied bearing stress. The test specimen is
basically a piece of sheet or plate with a carefully prepared hole some
standard distance from the edge. Edge-to-hole diameter ratios of 1.5 and
2.0 are common. A hardened pin is inserted through the hole and an AXIAL
load applied to the specimen and the pin. The bearing stress is computed
by dividing the load applied to the pin, which bears against the edge of
the hole, by the bearing area (the product of the pin diameter and the
sheet or plate thickness).

Different axis and on edge rather than in tension or compression. The
flat surface described by the hole in the part as it passes through the
thickness of the tube or plate of the hole. The stress is spread through
the material differently.

Curtis S.

And I am not an engineer either, I make rocket motors, and need this
information constantly.

clareatsnyder.on.ca

unread,
Feb 27, 2006, 7:27:04 PM2/27/06
to
On Mon, 27 Feb 2006 16:38:44 -0500, Curtis Scholl <csc...@izzy.net>
wrote:

>Clare:
>


>Bearing properties are used when designing mechanically fastened joints.
>The purpose of a bearing test is to determine the the deformation of a
>hole as a function of the applied bearing stress. The test specimen is
>basically a piece of sheet or plate with a carefully prepared hole some
>standard distance from the edge. Edge-to-hole diameter ratios of 1.5 and
>2.0 are common. A hardened pin is inserted through the hole and an AXIAL
>load applied to the specimen and the pin. The bearing stress is computed
>by dividing the load applied to the pin, which bears against the edge of
>the hole, by the bearing area (the product of the pin diameter and the
>sheet or plate thickness).
>
>Different axis and on edge rather than in tension or compression. The
>flat surface described by the hole in the part as it passes through the
>thickness of the tube or plate of the hole. The stress is spread through
>the material differently.
>
>Curtis S.
>
>And I am not an engineer either, I make rocket motors, and need this
>information constantly.
>
>

I understand what the bearing strength is - I'm just saying it is the
wrong strength to design to when designing the wing, as the
compressive, shear, and tension strengths more realistically represent
the loads in the wing.

If you substitute the tensile strength numbers into the calculations,
I'll bet you find the wing fails around 2 Gs, which is what the test
found.

Curtis Scholl

unread,
Feb 28, 2006, 9:50:32 AM2/28/06
to
Clare:

Bearing strength is for the Connecting Points of all those items you
mentioned. And the engineer stopped the test before complete failure
from what I read.

Excessive deflection is not a failure mode unless the engineer states it
is out of spec. And again, what was the sandbag testing weight at which
the testing was stopped? And again, excessive deflection where? In the
skins? In the Spars? in the struts?

You stated:
**
"No- Gary built a wing according to plans and sandbag tested it under
an engineer's supervision. The engineer stopped the loading before
failure because of excessive deflection IIRC."
**

And I have a set of plans, and I recall seeing the sleeves being called
out in the manual. And I don't see where aluminum skins are in the
plans. Again, deflection where?

"And sleaves in the spars by your own admission. Which are NOT in the
plans."

See wing plan D-WING3. It clearly shows what to do on the wing as for
the sleeve inserts to reinforce the connecting points, wing to cabane,
and strut to wing.

Look, this aircraft flies, in it's designed weight class, it has worked
for many years. The designer flies it. There are many copies flying, and
the death rate is not "significant" considering what happened in the
accident reports.

As you said, Clare, you are not an engineer. And you don't have current
validated DATA to back you up. I have stated instances of you being
incorrect in your assumptions. And in one of the instances, given you
the reference for corrections of your statements.

We are done. No response is necessary or wanted. This will be an agree
to disagree situation and no further contact is warranted.

Curtis Scholl

Richard Isakson

unread,
Feb 28, 2006, 4:02:31 PM2/28/06
to
I've never liked powered ultralights that use the US part 103 definition of
ultralight. The FAA limited the empty weight to far too light a weight.
They could have added a hundred pounds to the empty weight and kept the
other limitations as they are. This would have produced a real viable
airplane class. As it is, the structures are designed without a lot of
redundancy. In the airplane in question here, the design is fine when it's
new but if there's much hangar rash on the leading edge of the wing the load
that the wing can take is greatly reduced.

I read all of the screaming and shouting about this plane's wing and I
finally decided to take a quick and dirty look at the stresses on the front
spar. I choose the front spar arguing that it's going to take most of the
loads anyway. Using a single spar unravels a knotty little problem of
resolving the reactions from the fore and aft lift struts and the two flying
wires. This was only a quick look after all. I just wanted to see if the
wing was in the ball park. Please understand that the real airplane needs
both the fore and aft spars and the fore and aft lift struts in order to
react to the wings chordwise torsion. For convenience, I used a level
spanwise wing loading. This is a conservative approach. We speak of
conservative and non-conservative. A conservative error is one which leads
to an over designed structure. A non-conservative error is one which leads
to an inadequate structure.

I used the drawings that are available on the web. They include the spar
inserts. I have a slight problem with these inserts. Since they aren't a
tight fit, its possible for the edge of the insert to cause a stress riser
on the inside of the spar tube. It's something to keep an eye on.

With my quick and dirty assumptions, I found that the spar would yield at
4.4 g's at 600 pounds gross weight. That is looking at bend moment stresses
only. A betters analysis would raise that number. This includes the
inserts. Without the inserts the spar yield at 2.3 g's at 600 pounds and
2.8 g's at 500 pounds.

Per MIL-HDBK-5 the tensile yield strength 6061-T6 tubing is 35K PSI

Rich


clareatsnyder.on.ca

unread,
Feb 28, 2006, 7:12:54 PM2/28/06
to
On Tue, 28 Feb 2006 09:50:32 -0500, Curtis Scholl <csc...@izzy.net>
wrote:

>Clare:


>
>Bearing strength is for the Connecting Points of all those items you
>mentioned. And the engineer stopped the test before complete failure
>from what I read.
>
>Excessive deflection is not a failure mode unless the engineer states it
>is out of spec. And again, what was the sandbag testing weight at which
>the testing was stopped? And again, excessive deflection where? In the
>skins? In the Spars? in the struts?

The wing was tested to 2 G's at 600 lb gross. At that point the spars
were ready to fail if I recall correctly.


>
>You stated:
>**
>"No- Gary built a wing according to plans and sandbag tested it under
>an engineer's supervision. The engineer stopped the loading before
>failure because of excessive deflection IIRC."
>**
>
>And I have a set of plans, and I recall seeing the sleeves being called
>out in the manual. And I don't see where aluminum skins are in the
>plans. Again, deflection where?

The wing was tested with NO ALUMINUM SKINS. There are doublers at the
arttach points in the plans, but the spars are not doubled between the
root and the struts, which is where the spar is not up to the job from
what I gathered. I'll be talking to Gary tomorrow


>
>"And sleaves in the spars by your own admission. Which are NOT in the
>plans."
>
>See wing plan D-WING3. It clearly shows what to do on the wing as for
>the sleeve inserts to reinforce the connecting points, wing to cabane,
>and strut to wing.
>
>Look, this aircraft flies, in it's designed weight class, it has worked
>for many years. The designer flies it. There are many copies flying, and
>the death rate is not "significant" considering what happened in the
>accident reports.
>
>As you said, Clare, you are not an engineer. And you don't have current
>validated DATA to back you up. I have stated instances of you being
>incorrect in your assumptions. And in one of the instances, given you
>the reference for corrections of your statements.
>
>We are done. No response is necessary or wanted. This will be an agree
>to disagree situation and no further contact is warranted.
>
>Curtis Scholl

*** Free account sponsored by SecureIX.com ***

Bob Kuykendall

unread,
Feb 28, 2006, 7:34:13 PM2/28/06
to
Earlier, Richard Isakson wrote:

> With my quick and dirty assumptions, I found that the spar would yield at
> 4.4 g's at 600 pounds gross weight. That is looking at bend moment stresses
> only. A betters analysis would raise that number. This includes the
> inserts. Without the inserts the spar yield at 2.3 g's at 600 pounds and
> 2.8 g's at 500 pounds.

Interesting. When I run the moment of inertia for 2" tubing of .058"
wall, I get 0.1667 in^4. Using that number and a yield strength of 35
ksi I get a yield moment of 5833 in/lbs. Do those numbers agree with
yours? Of course, those figures disregard cripling or buckling, which
I've not seen mentioned in this thread.

I suspect that this whole thing will come down to a somewhat subjective
matter of distributions and deflections. The distribution of loads
between the forward and aft spars will make a big difference, and I
think that the wing deflection will start to look scary before the spar
tubes reach yield. But those are just more non-engineer's guesses, and
there's been plenty too much of those already.

Taking this out on a tangent, one thing about little airplanes like
this that I don't understand is why so many of them use tubular spars.
It seems to me that you can get so much better strength/weight and
stiffness/weight using a built-up I-beam or C-section spar. Yeah, it's
a bit more trouble. But the result is either better strength and
stiffness for the same weight, or the same strength for less weight.
But again, that's just my non-engineer wing developer perspective.

Thanks, and best regards to all

Bob K.
http://www.hpaircraft.com/hp-24

Richard Isakson

unread,
Feb 28, 2006, 10:31:30 PM2/28/06
to
"Bob Kuykendall" wrote ...

> Interesting. When I run the moment of inertia for 2" tubing of .058"
> wall, I get 0.1667 in^4. Using that number and a yield strength of 35
> ksi I get a yield moment of 5833 in/lbs. Do those numbers agree with
> yours? Of course, those figures disregard cripling or buckling, which
> I've not seen mentioned in this thread.

For the spar alone those numbers are right and result in the poor spar
performance without the inserts. The maximum bending moment happens at the
strut attach point. This is where the insert sits and that increases the
moment of inertia to 0.3155. This results in the improved load handling
ability.

In the interest of laziness, I didn't look at spar buckling nor did I look
at negative loading. There is a potential for column buckling of the spar
between the root and the strut attach point. As the wing is lifted, the
strut is placed in tension. This places the inboard portion of the spar in
compression. The combination of the compression load and the lift load
could potenially cause buckling. Maybe I'll look at that sometime.

> I suspect that this whole thing will come down to a somewhat subjective
> matter of distributions and deflections. The distribution of loads
> between the forward and aft spars will make a big difference, and I
> think that the wing deflection will start to look scary before the spar
> tubes reach yield. But those are just more non-engineer's guesses, and
> there's been plenty too much of those already.

Reading what little has been said about the load testing, I suspect there's
a problem in the way the wing was held. It almost sounds like they didn't
have the rear lift strut attached

> Taking this out on a tangent, one thing about little airplanes like
> this that I don't understand is why so many of them use tubular spars.
> It seems to me that you can get so much better strength/weight and
> stiffness/weight using a built-up I-beam or C-section spar. Yeah, it's
> a bit more trouble. But the result is either better strength and
> stiffness for the same weight, or the same strength for less weight.
> But again, that's just my non-engineer wing developer perspective.

I think it's to keep the labor costs down. You might ask Chuck.

Rich


Montblack

unread,
Mar 1, 2006, 12:24:38 AM3/1/06
to
("Richard Isakson" wrote)
[snip]

> I've never liked powered ultralights that use the US part 103 definition
> of ultralight. The FAA limited the empty weight to far too light a
> weight. They could have added a hundred pounds to the empty weight and
> kept the other limitations as they are. This would have produced a real
> viable airplane class.


Agreed, almost.

350 lbs would have been great (without floats).

Low stall number is fine, but let's remove the speed limit on the upper end.
If it weighs X and stalls at Y, carries one person and (8<g>) gallons of
fuel ...who cares about its top-end speed!


Montblack
Hell ...I'M not 103 legal !!! :-)

clareatsnyder.on.ca

unread,
Mar 1, 2006, 1:14:04 AM3/1/06
to

It WAS attatched. It was attatched as if it was on the plane but
upside down.

>> Taking this out on a tangent, one thing about little airplanes like
>> this that I don't understand is why so many of them use tubular spars.
>> It seems to me that you can get so much better strength/weight and
>> stiffness/weight using a built-up I-beam or C-section spar. Yeah, it's
>> a bit more trouble. But the result is either better strength and
>> stiffness for the same weight, or the same strength for less weight.
>> But again, that's just my non-engineer wing developer perspective.
>
>I think it's to keep the labor costs down. You might ask Chuck.
>
>Rich
>

*** Free account sponsored by SecureIX.com ***

Robert Schieck

unread,
Mar 1, 2006, 7:04:24 AM3/1/06
to
Richard Isakson wrote:

> In the interest of laziness, I didn't look at spar buckling

Without the jury struts, IIRC, it cripples long before anything else fails.

Deflection at the jury strut attach point to the strut , during the load
test was, 2 or 2.5 inches when really loaded down.

hth

Rob

Peter Dohm

unread,
Mar 1, 2006, 8:00:01 AM3/1/06
to
"Montblack" <Y4-NOT...4monty4blacky@yvisiy.com> wrote in message
news:120abus...@corp.supernews.com...
I'm not sure that the subject is worth discussing further at this late date.
But, since we are--the stall speed number is definitely *not* fine!

The problem with the unreasonably low stall speed is that very modest
surface gusts can easily upset an ultralight while taxiing; or worse yet,
while taking off or landing.

350 lbs, one seat, and the speeds now authorized for LSA would have
*dramatically* improved safety with only very modest training--although any
maximum speed of at least 80 Kts would have worked.

Peter

Flagellation of a deceased equine is unsatisfying!


fredf...@spamcop.net

unread,
Mar 1, 2006, 12:28:54 PM3/1/06
to

Curtis Scholl wrote:
> Clare:
>
> Bearing strength is for the Connecting Points of all those items you
> mentioned. And the engineer stopped the test before complete failure
> from what I read.
>
> Excessive deflection is not a failure mode unless the engineer states it
> is out of spec. And again, what was the sandbag testing weight at which
> the testing was stopped? And again, excessive deflection where? In the
> skins? In the Spars? in the struts?

I disagree. Excessive deflection is a failure mode if it has
sufficient
adverse impact on function or safety, without regard to whether or
not any particular engineer realized it and said so.

That is more a criticism of the wording you used, than it is of the
implied question--How do we know the deflection was excessive?

(And what was it that deflected?)

--

FF

Richard Isakson

unread,
Mar 1, 2006, 1:49:31 PM3/1/06
to
<clare at snyder.on.ca> wrote ...

> It WAS attatched. It was attatched as if it was on the plane but
> upside down.

I'd like to take a look at the deflections for this wing from a theoretical
veiwpoint. This discussion seems to have been on-going from another group.
Which group? What was the final loading and deflections when the test was
called off? Did the wing that was tested have the inserts as shown in the
current plans or was this an earlier wing without them?

Thanks

Rich


clareatsnyder.on.ca

unread,
Mar 1, 2006, 2:51:48 PM3/1/06
to
On Tue, 28 Feb 2006 09:50:32 -0500, Curtis Scholl <csc...@izzy.net>
wrote:

Clare wrote:

>"No- Gary built a wing according to plans and sandbag tested it under
>an engineer's supervision. The engineer stopped the loading before
>failure because of excessive deflection IIRC."
>**

>As you said, Clare, you are not an engineer. And you don't have current

>validated DATA to back you up. I have stated instances of you being
>incorrect in your assumptions. And in one of the instances, given you
>the reference for corrections of your statements.

Just an update on the testing procedure and results on the TP wing
tested at Kitchener Waterloo International Airport last year.

The wing was instrumented with strain guages. The main spar was
bending significantly between the Cabane mounting and the strut, and
the strain readings were beyond the limits the engineer was
comfortable with at 2Gs testing for a 600 lb plane.Failure was
IMMINENT.

Morgans

unread,
Mar 1, 2006, 4:31:13 PM3/1/06
to

<clare at snyder.on.ca> wrote

> Just an update on the testing procedure and results on the TP wing
> tested at Kitchener Waterloo International Airport last year.
>
> The wing was instrumented with strain guages. The main spar was
> bending significantly between the Cabane mounting and the strut, and
> the strain readings were beyond the limits the engineer was
> comfortable with at 2Gs testing for a 600 lb plane.Failure was
> IMMINENT.

I don't understand why the testing did not continue. Seems to me that
having someone say that failure was imminent, based on strain gauges, leaves
a lot of room for opinion to creep in.
--
Jim in NC

Richard Lamb

unread,
Mar 1, 2006, 4:18:07 PM3/1/06
to

Somehow, I get the impression that they would not have approved of Voyager's
wings either.


FWIW.

Richard

clareatsnyder.on.ca

unread,
Mar 1, 2006, 5:42:59 PM3/1/06
to
On Wed, 1 Mar 2006 16:31:13 -0500, "Morgans" <jsmo...@chJUarNKer.net>
wrote:

They didn't want to have to shovel up the sand when the sandbags
split on the floor.

UltraJohn

unread,
Mar 1, 2006, 7:54:29 PM3/1/06
to
clare at snyder.on.ca wrote:

> On Wed, 1 Mar 2006 16:31:13 -0500, "Morgans" <jsmo...@chJUarNKer.net>
> wrote:
>
>>
>><clare at snyder.on.ca> wrote
>>
>>> Just an update on the testing procedure and results on the TP wing
>>> tested at Kitchener Waterloo International Airport last year.
>>>
>>> The wing was instrumented with strain guages. The main spar was
>>> bending significantly between the Cabane mounting and the strut, and
>>> the strain readings were beyond the limits the engineer was
>>> comfortable with at 2Gs testing for a 600 lb plane.Failure was
>>> IMMINENT.
>>
>>I don't understand why the testing did not continue. Seems to me that
>>having someone say that failure was imminent, based on strain gauges,
>>leaves a lot of room for opinion to creep in.
> They didn't want to have to shovel up the sand when the sandbags
> split on the floor.


I don't have much interest on either side of this but have been reading with
curiosity. This last statement was really very lame! If someone's purpose
was to prove a design was deficient you would have thought they have tested
to failure. I doesn't cost alot to get a couple high school kids to do the
shoveling.
My question, How much permanent deflection in the wings after the test?
John

Morgans

unread,
Mar 1, 2006, 10:17:13 PM3/1/06
to

<clare at snyder.on.ca> wrote

> They didn't want to have to shovel up the sand when the sandbags
> split on the floor.

IF that is really the reason, IMHO, that is pretty damn sorry.
--
Jim in NC

clareatsnyder.on.ca

unread,
Mar 1, 2006, 10:46:22 PM3/1/06
to
On Wed, 1 Mar 2006 22:17:13 -0500, "Morgans" <jsmo...@chJUarNKer.net>
wrote:

>

Look, I was not there when the test was done, but I know the guys who
were. I know the engineer who instrumented the test.

The shovelling was in jest. From what I understand, they were hoping
to make some modifications to the wing to make it a better wing, so
did not test to destruction. After doing more analysis they decided to
cut their losses and forget the project. They designed an all metal
wing, which MAY eventually be used on a TP-ish plane.

The strain guage information, along with the distortion under load was
adequate information to tell the engineer "enough"

Bob Kuykendall

unread,
Mar 2, 2006, 10:27:34 AM3/2/06
to
Earlier, UltraJohn wrote:

> My question, How much permanent
> deflection in the wings after the test?

I think that that is one of the most central questions in this
exchange. If there was no permanant deflection, then the material never
reached its 35 ksi (some books like Machinery's Handbook say 40 ksi)
yield stress, and without deflection records we'll never really know
how close the test wing came to actually failing.

Overall, I'm satisfied with the descriptions that a fairly valid static
test was done. However, I know from watching videos of static tests and
from running deflection estimates for various wings that tests even to
just the design limit can result in some seriously spooky deflections.
And to add to that, the spar under consideration here is very shallow,
and that will also tend to give it a lot of deflection.

One thing I would have expected for a test like this is a comparison
between the wing under test and some sort of deflection reference like
a template. Since the spar in this case has a constant depth of 2", the
same template can be applied anywhere on the spar to check the
deflection and deduce the stress from that.

If I've run the numbers right, for a 2" tall aluminum (Young's modulus
of 10 meg) element, at 35 KSI in the extreme fibers the element will
have a curvature radius of about 20 feet. That may sound like a pretty
shallow curve, but over four feet it results in a bend of about 9.5
degrees and a vertical deflection of about 4".

But don't take my word for it, I'm _not_ an engineer...

Robert Loer

unread,
Mar 4, 2006, 1:00:21 AM3/4/06
to

<SNIP>>

>>I don't understand why the testing did not continue. Seems to me that
>>having someone say that failure was imminent, based on strain gauges,
>>leaves
>>a lot of room for opinion to creep in.
> They didn't want to have to shovel up the sand when the sandbags
> split on the floor.
>

That was a stupid childish answer.

I would say Mr Scholl kicked butt with facts.


Richard Lamb

unread,
Mar 4, 2006, 9:11:23 AM3/4/06
to

Yes, but let it go now, Robert.

Not worth getting our collective panties wadded up any more.

Thanks everybody.

Richard

Richard Lamb

unread,
Mar 4, 2006, 10:56:49 PM3/4/06
to

Texas Parasol CD Archive:

http://www.matronics.com/photoshare/cave...@earthlink.net.03.04.2006/

Thanks Matt.

Richard Lamb


Archive by:
Matt G Dralle | Matronics | PO Box 347 | Livermore | CA | 94551
925-606-1001 V | 925-606-6281 F | dra...@matronics.com Email
http://www.matronics.com/ WWW | Featuring Products For Aircraft

Richard Lamb

unread,
Mar 12, 2006, 2:19:47 AM3/12/06
to
0 new messages