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 Historical content remains viewable.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

[Q] Fibrelam, what is it?

284 views
Skip to first unread message

Stephen_Chan

unread,
Jul 18, 1998, 3:00:00 AM7/18/98
to
Hello:

Could someone explain to me what is "Fibrelam".
I know the Streak Shadow uses it but haven't seen
it used elsewhere. Just what is is compose of and
how strong is the stuff?

Thanks
Stephen

Ian Barclay

unread,
Jul 18, 1998, 3:00:00 AM7/18/98
to
In article <01bdb233$825ab1e0$4bf5...@direct.ca.direct.ca>,
Stephen_Chan <scha...@geocities.com> writes
Fibrelam is a pre-cured composite sandwich of unidirectional glass fibre
skins and a nomex honeycomb. Its principal use is for airliner floors.
However a British designer, John Edgley (who designed the unsuccessful
Optica observation plane) has designed a sailplane called the EA9
Optimist that uses it for most of the structure.

Although Fibrelam comes in stiff panel form it can be cut and folded or
bent in a single plane to make up more complex forms. On the Optimist
this gives a vintage glider slab look but a more than adequate aero
shape.

However what it does allow is a simple to build, cost effective, high
strength structure. Since the Fibrelam panels are factory made they are
aerospace quality with excellent strength to weight. For the Optimist
the panels are cut and routed to shape in the factory so that the kit
builder only has to slot the panels together and once aligned then bond
together with epoxy. No messy wet lay-ups, no time consuming sanding, no
autoclaves, no special skills required. It is claimed that it is
possible for two people to build a complete fuselage in a single day!

This is a structural concept that could have a lot of mileage for
homebuilding. Maybe not for scratchbuilt designs as I think you really
need specialist equipment to cut the panels. But for kits it has
tremendous potential. Imagine a flat packed kitplane, with parts that
are produced using automated machinery, and that is as easy to build as
a model plane.

This project has had substantial funding from the UK government. I think
all the parties involved see this as but a first step into a market much
wider than just gliders.

As for the specifics on strength you might try Fibrelam makers Ciba-
Geigy/Hexel Composites for info.

--
Ian Barclay
Helensburgh, UK

scha...@rocketmail.com

unread,
Jul 20, 1998, 3:00:00 AM7/20/98
to
Thanks for explaination Ian. Now I understand.
This "Fibrelam" stuff sounds alot like what
the Formula one guys were doing with aluminum
honeycomb sandwich panels in the early 80s.
Making the central tub by cut and fold method.
I recall, it doesn't take much tools to fabricate.
Just routers to score the panel to be fold.

To the experts out there. Why aren't there more
contruction of this type in homebuilts and kits,
especially here in North America?

Stephen

=============================================================

In article <yQqIeBAU...@ibarclay.demon.co.uk>,

-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum

Robert Chilcoat

unread,
Jul 21, 1998, 3:00:00 AM7/21/98
to
Ian Barclay wrote:

> Fibrelam is a pre-cured composite sandwich of unidirectional glass fibre
> skins and a nomex honeycomb.
>

> Although Fibrelam comes in stiff panel form it can be cut and folded or
> bent in a single plane to make up more complex forms. On the Optimist
> this gives a vintage glider slab look but a more than adequate aero
> shape.
>

A neighbor is building a Pulsar. This seems to be fabricated from a
similar material consisting of a honeycomb core and fiberglass skins.
However, the fuselage clearly seems to have compound curves. What is
this material and how is it formed? Is it readily available?

Bob

Applying technology is simply finding the right wrench with which to
pound in the correct screw.

Charles K. Scott

unread,
Jul 21, 1998, 3:00:00 AM7/21/98
to
In article <6p0elk$qsl$1...@nnrp1.dejanews.com>
scha...@rocketmail.com writes:

> To the experts out there. Why aren't there more
> contruction of this type in homebuilts and kits,
> especially here in North America?
>
> Stephen

Just an opinion Stephen but it might not be cost effective. The only
company that might be interested would be Murphey, which builds a slab
sided airplane called the "Rebel" and "Super Rebel".

I don't know what this material Fibrelam costs, nor do I know what
aluminum honeycomb costs but I'll bet it costs a LOT more than a simple
sheet of aluminum and a few internal bulkheads and rivets.

I'll bet the current Murphey construction process for the Rebel weighs
less too. It might not be as stiff, but it isn't intended to race
around tracks at 190 mph either. The formula one guys don't much worry
about how much a particular part costs. If it improves performance and
gives them an advantage over the other guys, it goes into the car and
the sponsor picks up the tab.

But Murphey has to produce a kit that is cost effective. It has to be
cheap enough that people will actually be able to afford it. So using
the latest hot space age material may not be the best business
decision. At least not for routine homebuilts.

Corky Scott

Ian Barclay

unread,
Jul 21, 1998, 3:00:00 AM7/21/98
to
Corky and Stephen said,

>
>> To the experts out there. Why aren't there more
>> contruction of this type in homebuilts and kits,
>> especially here in North America?
>>
>> Stephen
>
>Just an opinion Stephen but it might not be cost effective. The only
>company that might be interested would be Murphey, which builds a slab
>sided airplane called the "Rebel" and "Super Rebel".
>
>I don't know what this material Fibrelam costs, nor do I know what
>aluminum honeycomb costs but I'll bet it costs a LOT more than a simple
>sheet of aluminum and a few internal bulkheads and rivets.
>
I'm not sure it would cost much more than more conventional
construction. The basic Fibrelam panels are being produced in reasonably
large quantities in what I assume is a pretty well automated process and
so the raw material costs are probably not that high.

But where the big savings will be is the next stages in construction. If
your kit manufacturer CNC machines the composite panels to size this is
I am sure cheaper than heat treating alloy, cutting sheets to shape,
drilling, reaming holes, welding steel frames or whatever it is that is
needed for a Rebel.

Then in the final stage of actual assembly I would imagine a simple slot
together and glue process is at least as quick (and hence as cheap) as
riveting.

Remember that it is claimed (OK I know claim and reality are not
necessarily the same thing but...) that 2 people could build a complete
Optimist sailplane fuselage from Fibrelam parts in a single day. As for
cost I believe that the basic Optimist kit in the UK is 10000 pounds
($16000) plus sales tax so its not unreasonable.

By the way, when I described in a post that the Optimist is a bit slab
sided I didn't mean THAT (Murphy Rebel) slab sided. Its like a 1950s
wood and fabric glider; faceted rather than slab sided really. With
careful design you would have no problem making an aerodynamically sound
and attractive shape.

>I'll bet the current Murphey construction process for the Rebel weighs
>less too. It might not be as stiff, but it isn't intended to race

Again I'm not so sure. The Fibrelam Optimist is lighter than the
comparable conventional glassfibre Schleicher ASK23 sailplane so
structurally it is very efficient.

>But Murphey has to produce a kit that is cost effective. It has to be
>cheap enough that people will actually be able to afford it. So using
>the latest hot space age material may not be the best business
>decision. At least not for routine homebuilts.

You said it. The key to success is getting cost out the design.
Technology for its own sake is no good. It must justify its place by
making the plane cheaper or higher performing for the same money. Mass
produced composite panels that can be easily and quickly joined together
might just do this.

Bob, you mentioned the compound curvature honeycomb composite sandwich
panels of the Pulsar. I think I am correct in saying, correct me if I am
wrong any Pulsar builders out there, that the fuselage shell is factory
produced. I think the shell is laid up as separate sheets of glass
prepeg and honeycomb in a mold and then cured in an autoclave. This
method allows compound curvature and is structurally pretty well
optimum. However it is labor intensive (for the kit manufacturer)
requiring manual layup of the composite and requires expensive tooling
(molds and autoclave). This makes for one thing : high cost. With the
glue together bent composite sandwich method you get most of the
structural efficiency but at much reduced cost.

Its a new structural concept in aircraft and might take a while to gain
widespread acceptance but I am pretty confident that over the next few
years quite a few manufacturers will take up the idea.

d...@xpsystems.com

unread,
Jul 22, 1998, 3:00:00 AM7/22/98
to
An obvious candidate for Fibrelam might be Barnaby Wainfain's Facetmobile,
which has all flat surfaces (except for the engine cowl). It would be
interesting to see his four-place design (shelved for lack of funding/time)
produced in Fibrelam.

David

In article <9weLWAAC...@ibarclay.demon.co.uk>,


Ian Barclay <i...@ibarclay.demon.co.uk> wrote:
> Corky and Stephen said,
> >
> >> To the experts out there. Why aren't there more
> >> contruction of this type in homebuilts and kits,
> >> especially here in North America?
> >>
> >> Stephen
> >

-----== Posted via Deja News, The Leader in Internet Discussion ==-----

scha...@rocketmail.com

unread,
Jul 26, 1998, 3:00:00 AM7/26/98
to
Corky:

You are most likely right about Fibrelam and other
"latest hot space age material" stuff being more
expensive, but there's another thing you have to
consider. Build time. They say more than 90% of
all homebuilds are abandon short of completion.
I don't know where this statistic come from or
whether it's true or not, but it's believable when
you consider the average plane takes 1000 to 2000+
hours to complete. If kit manufacturers can shorten
the build time to 300 hours or so, I'd bet the
completion rate would rise sharply.

I have never worked with Fibrelam or aluminium honeycomb
sheets, but it sounds like the "cut, glue and fold" method
of construction is really idea for our hobby. For
a flat panel plane like the Murphey Rebel, I'd bet a couple
of person can erect the main fuselage in a weekend and
maybe the wings and tail a few days more.

I think even if you discount the time saving, the psychologically
effect of such an easy method of building would help
builders complete their aircraft rather than pieces gathering
dust.

I am probably too optimistic here, but I think this stuff show
potential.

Stephen

PS. Anyone know how much Fibrelam cost?

====================================================================

In article <6p2aan$ak6$1...@dartvax.dartmouth.edu>,


Charles.K.Scott@**NOSPAM**.dartmouth.edu (Charles K. Scott) wrote:
> In article <6p0elk$qsl$1...@nnrp1.dejanews.com>
> scha...@rocketmail.com writes:
>

> > To the experts out there. Why aren't there more
> > contruction of this type in homebuilts and kits,
> > especially here in North America?
> >
> > Stephen
>

> Just an opinion Stephen but it might not be cost effective. The only
> company that might be interested would be Murphey, which builds a slab
> sided airplane called the "Rebel" and "Super Rebel".
>
> I don't know what this material Fibrelam costs, nor do I know what
> aluminum honeycomb costs but I'll bet it costs a LOT more than a simple
> sheet of aluminum and a few internal bulkheads and rivets.
>

> I'll bet the current Murphey construction process for the Rebel weighs
> less too. It might not be as stiff, but it isn't intended to race

> around tracks at 190 mph either. The formula one guys don't much worry
> about how much a particular part costs. If it improves performance and
> gives them an advantage over the other guys, it goes into the car and
> the sponsor picks up the tab.
>

> But Murphey has to produce a kit that is cost effective. It has to be
> cheap enough that people will actually be able to afford it. So using
> the latest hot space age material may not be the best business
> decision. At least not for routine homebuilts.
>

> Corky Scott

highflyer

unread,
Jul 29, 1998, 3:00:00 AM7/29/98
to
scha...@rocketmail.com wrote:
>
> Corky:
>
> You are most likely right about Fibrelam and other
> "latest hot space age material" stuff being more
> expensive, but there's another thing you have to
> consider. Build time. They say more than 90% of
> all homebuilds are abandon short of completion.
> I don't know where this statistic come from or
> whether it's true or not, but it's believable when
> you consider the average plane takes 1000 to 2000+
> hours to complete. If kit manufacturers can shorten
> the build time to 300 hours or so, I'd bet the
> completion rate would rise sharply.
>

I believe that is an old statistic when all homebuilts were plans
built and there was no such thing as a 'kit'.

You can't get much easier than the modern glass kits where the
airplane looks like a ReVell Model Kit. Glue together the fuselage
halves, glue on the wings, bolt on the engine, and go for it!
Just like with the Revell kit, more time is probably spent on the
details that you don't even THINK about than on the basic airframe.

I see the fiberlam as a LONGER build process than the new glass kits.

0 new messages