http://busybeetools.ca/cgi-bin/picture10?NTITEM=B2227L
Trying to find out more about it I looked at the manual but it is quite
useless. One of the questions I had was how is the chuck attached to the
spindle. The dealer tells me that there is a "flange" but any further
details have not been forthcoming so far.
Are there any standards that would describe such flange attachment? Are
there standard sizes? I am concerned about accessories (5C collet chuck??).
This machine is BTW only superficially similar to the Grizzly G0602 which I
understand has a 1-3/4 - 8TPI spindle.
--
Michael Koblic,
Campbell River, BC
> Michael Koblic,
> Campbell River, BC
Michael
Probably attaches with threads as you mention. Don't buy from this
dealer if he does not know or does not find out.
Bob AZ
Looking at the closeup pics of the chuck and headstock, shows the flange to
be the same size as the chuck.
There may be some common chucks in the 5" diameter range (and possibly a
little larger) that can be fitted/machined to adapt to the flange on this BB
machine.
Chucks much larger than 5" will likely require an adapter plate that the
user will need to perform some machining to mate the chuck with the spindle.
Oh, I noticed that an adapter plate is shown at the bottom of the
description, and separate from the optional 5" 4-jaw chuck.
The machine features/specs show that the spindle bore is 1" and has a MT4
taper.
The 1" bore excludes using a 5C collet draw tube.
Collet choices would be Morse tapers used with a drawbar, or adapt the MT4
to 3C collets, and use a draw tube.
The 3C collets allow feedthru of material (feeding long stock from the left
side), but the workpiece/stock diameter limit is smaller than 5C.
Morse taper collets don't allow workpiece/stock passthru. Collets sized MT4
may be relatively expensive. MT3 collets with an adapting MT3/MT4 sleeve
would allow cheaper MT3 collets to be used. A few MT4 collets in the larger
sizes might be desirable.
There are adapters for MT3 to 3C collets, but I don't know if there are
adapters that go directly from MT4 to 3C.
The draw tube for the 3C collets may be the same one available for the 9x20
models, or possibly adaptable with (maybe) minor modifications.
--
WB
.........
metalworking projects
www.kwagmire.com/metal_proj.html
"Michael Koblic" <mko...@gmail.com> wrote in message
news:hkti4...@news3.newsguy.com...
Enco makes (or made) a nice lathe with cam-lock chuck, that can fit a 5C
snap handle collet setup.
If you plan on switching often between lathe chuck and collets, cam-lock is
a dream.
The lathe you show is definitely not camlock, as you would see 3 add'l
square socket ditties, for the cams.
I wonder if 10" is large enough for a 5C collet system..... your 1" bore
would not be large enough for 5C.
--
EA
If it is the same as many of the 3-in-1 machines, it is a standard metric
100mm nosepiece that fits into a recess in the back of the chuck. 4 bolts
pass through the flange and are threaded into the chuck.
--
Dennis
You will be tempted to buy others, don't. Be patient, get it right the first time.
Steve
"Michael Koblic" <mko...@gmail.com> wrote in message news:hkti4...@news3.newsguy.com...
Good advice, except I don't think taper attachments are so common. My
buddy, who goes thru quite a few lathes, hasn't seen a taper attachment in a
decade or two.
Are taper attachments wedded to a given lathe, or can they be generic, like
a toolpost?
--
EA
"Existential Angst" <UNfi...@UNoptonline.net> wrote in message news:4b731571$0$22533$607e...@cv.net...
I just thought it might be an option as this guy installed a 5C collet on
his G0602 which also has 1" spindle bore:
In any case it was just to illustrate a point. I wondered if the "flange"
thing would severely limit me in the choice of attachments. For a start I do
not know what quality the chucks offered with this machine are and it would
be nice to have the option to go with another manufacturer for replacements.
Furthermore, the 4-jaw chuck available for this machine is only 5" which is
a bit small. Does this mean one would have to manufacture a whole adapter
for a bigger chuck? One has to believe that there are adapters available for
this type of spindle commercially somewhere, but without further details how
does one go looking for them?
Incidentally their 7X8 lathe also has a flange which I understand is common.
LMS sell adapters for *3" flange*. That is the sort of standardization I was
looking for. Things would be even simpler for a 9X20 with a 1.5"-8TPI or
M39X4mm spindle. Adapters for 3,4,5, or 6" chucks are freely available.
Meanwhile I got a response from the dealer and here it is:
"We do not have exact dimentions of the spindle flange, but it is 5" in
diameter and is machined to fit behind the chuck which comes with the lathe.
the holes are approximately 3.723" apart. One side of the adaptor is
machined in a similar design as the back of the 3 jaw chuck and the other
side is machined to accept the 4 jaw chuck. Since you want to use a
different chuck we suggest you wait until you acquire the lathe and the
chuck then machine the adaptor accordingly. "
I am not sure what to make of it.
Really? I thought a 13X40 would be much longer than 47". They do 12X26 at
double the price but that one has issues, too...
You can make a better adapter plate than you can buy because you can
fit it as carefully as you want to the exact sizes of the spindle
flange and chuck back recess.
If you measured the diameters of the spindle flanges on the production
line I suspect you'd see them increase continuously as the tool wore,
then jump back to the minimum when it was replaced. An adapter plate
to fit them would have to be at the large end of the tolerance in
order to fit all lathes, and a sloppy fit on most, while you can make
one exactly the right size for yours.
Look at lathe chucks in the MSC catalog and you'll see separate back
plates for the common mounting styles. You fit the plate to the
spindle, then shape it into a flange that fits snugly into the recess
on the back of the chuck.
http://www1.mscdirect.com/CGI/NNSRIT?PMPXNO=1784923&PMT4NO=0
jsw
"Michael Koblic" <mko...@gmail.com> wrote in message news:hl2ca...@news7.newsguy.com...
<snip>
> You can make a better adapter plate than you can buy because you can
> fit it as carefully as you want to the exact sizes of the spindle
> flange and chuck back recess.
>
> If you measured the diameters of the spindle flanges on the production
> line I suspect you'd see them increase continuously as the tool wore,
> then jump back to the minimum when it was replaced. An adapter plate
> to fit them would have to be at the large end of the tolerance in
> order to fit all lathes, and a sloppy fit on most, while you can make
> one exactly the right size for yours.
Ah, I was under the impression that they are all made oversize for the final
turning down on each particular machine. At least that is what the LMS
adapters are.
>
> Look at lathe chucks in the MSC catalog and you'll see separate back
> plates for the common mounting styles. You fit the plate to the
> spindle, then shape it into a flange that fits snugly into the recess
> on the back of the chuck.
> http://www1.mscdirect.com/CGI/NNSRIT?PMPXNO=1784923&PMT4NO=0
This is kind of my point: There are 297 adapters listed in that section.
Which one do you pick for this machine? Personally I would prefer to get one
that at least somewhat matches the spindle than have to make the whole thing
from scratch. With a known common type such as "10 inch 3-jaw with a A1-6
mount" I know what everyone is talking about. The one adapter you linked
would clearly not fit this machine. I looked at all 297 and could not decide
which one would.
Thanks. This gave me a point of reference. I went and had a look at the
Grizzly web site. Their G9729 has (presumably??) the same spindle. I read
the manual (their manuals are far superior to anything out of Busy Bee). It
seems this flange will not take any chuck bigger than 6". And there were
other caveats.
The more I read the less I like it...
All you need is a chunk of steel or iron the right size. I've made
adapters out of scrap hydraulic cylinder rod, a cast-iron pipe coupler
and a locating pin from an old Segway fender mold.
Barbell weights are the right shape, though the iron may be difficult
to cut. A few days of cooking inside the wood stove might soften then
up enough, it works well with flame-cut hot rolled steel.
Fitting an adapter to the spindle is easier if the adapter is on a
mandrel between centers so you can check the fit and replace the
adapter exactly concentric, to remove another half thousandths all the
way around. You could bolt it to the flange to cut it almost to size
quickly.
5" is a reasonable size for the chuck on a 10" lathe. You don't want
it to hit the carriage. I have a 5" 3 jaw and a 6" 4 jaw and have to
be very careful with the larger one, or set a stop.
jsw
That presupposes a degree of skill on my part :-) I was doing a dry run in
my mind and got stuck at stage one: Getting the right holes drilled in the
right places if you cannot remove the flange and use transfer punches. But I
guess careful measurement would suffice. The next stage would involve a very
interrupted cut which presumably is not an issue with a lathe this size. BTW
can you part the piece out by having the parting tool at 90 degrees to the
usual direction? Cutting into the piece at the right angles with the tool
lined up along the lathe bed? I have seen it done with a wood lathe.
The good thing is that one would not have to cut large internal threads for
this spindle.
>
> 5" is a reasonable size for the chuck on a 10" lathe. You don't want
> it to hit the carriage. I have a 5" 3 jaw and a 6" 4 jaw and have to
> be very careful with the larger one, or set a stop.
Here is a question then: Say you have a 9" swing lathe and a 8" diameter,
0.25" thick plate of mild steel than needs a 6.5" hole bored in the centre.
You cannot hold it on the outside in a chuck. You cannot clamp it outside on
a face plate, even a wooden one. Not enough space for the clamps. What do
you do? Would making an 8" sacrificial face plate and super gluing the work
piece to it work? Would you cut the hole with a mill on a rotary table? None
of the above?
[ ... ]
>> All you need is a chunk of steel or iron the right size. I've made
>> adapters out of scrap hydraulic cylinder rod, a cast-iron pipe coupler
>> and a locating pin from an old Segway fender mold.
>>
>> Barbell weights are the right shape, though the iron may be difficult
>> to cut. A few days of cooking inside the wood stove might soften then
>> up enough, it works well with flame-cut hot rolled steel.
>>
>> Fitting an adapter to the spindle is easier if the adapter is on a
>> mandrel between centers so you can check the fit and replace the
>> adapter exactly concentric, to remove another half thousandths all the
>> way around. You could bolt it to the flange to cut it almost to size
>> quickly.
[ ... ]
> BTW
> can you part the piece out by having the parting tool at 90 degrees to the
> usual direction? Cutting into the piece at the right angles with the tool
> lined up along the lathe bed? I have seen it done with a wood lathe.
No! The blade of the parting tool would need to be curved like
a parenthesis on a radius to match the cut to be made. This is called a
trepaning tool. A standard parting tool would bind on the outside
below the cut.
[ ... ]
>> 5" is a reasonable size for the chuck on a 10" lathe. You don't want
>> it to hit the carriage. I have a 5" 3 jaw and a 6" 4 jaw and have to
>> be very careful with the larger one, or set a stop.
>
> Here is a question then: Say you have a 9" swing lathe and a 8" diameter,
> 0.25" thick plate of mild steel than needs a 6.5" hole bored in the centre.
> You cannot hold it on the outside in a chuck. You cannot clamp it outside on
> a face plate, even a wooden one. Not enough space for the clamps. What do
> you do? Would making an 8" sacrificial face plate and super gluing the work
> piece to it work? Would you cut the hole with a mill on a rotary table? None
> of the above?
What I would do is to mount the faceplate (which is probably
larger in diameter than the chucks), coat the surface with double-sided
tape, approximately center the workpiece, and using a live center with a
flat pusher in it -- or a turned piece of aluminum with a large center
hole drilled not all the way through, and use the tailstock and this to
press the workpiece against the faceplate. Better if your pusher is
turned to press near the OD and relieved a bit inside that.
This is used to press the (slightly oversized) workpiece firmly
against the plate to get the tape to grip firmly. Then (using very slow
speeds) drill through the center with a bit which fits your lathe --
perhaps a 1/2" one unless you can find a Morse taper shank drill to fit
the tailstock taper.
Then, using a boring bar, bore through the workpiece slowly
increasing the size until it is the right ID.
Then -- put on the 3-jaw chuck (assuming that you don't need the
OD to be truly concentric with the bored hole) expand the jaws inside th
bored hole to grip the workpiece so you can turn the OD to something
close to concentric.
If you need the two to be truly concentric, turn the OD while
still holding the workpiece against the faceplate with the live enter
and pusher adaptor. Then shift to boring the OD.
Yes -- the super glue (or a bearing mount Loctite) would work in
place of the double-sided tape. You will need to use heat to release
it.
And -- while boring, it would probably help to have an extra
sacrificial aluminum plate between the workpiece and the faceplate so you
don't cut the faceplate.
Of course if the plate which you are machining can tolerate some
holes in its surface -- perhaps in places which would be machined larger
later -- drill and tap for bolts to hold it to the faceplate..
And yes -- you could use a rotary table and the double-stick
tape or glue again on the mill to do the same thing. The finish of the
bore will probably not be as nice, however.
Note that you can probably use a larger diameter 4-jaw on the
lathe than the 3-jaw because the jaws are easier to adapt to not stick
out as far. My 12x24" Clausing has a 6-1/4" 3-jaw, and a 10" 4-jaw.
In your particular described situation, a 9" swing lathe, and an
8" diameter workpiece, *if* your jaws for an 8" 4-jaw chuck have the
last step only 1/2" thick, you could just barely grip the 8" diameter
workpiece with the jaws extending only partially beyond the OD of the
chuck. Be very careful when doing this:
1) Make sure that it does not hit the bed or the wings of the carriage,
and that the tool can reach fully through the workpiece without
bringing a part of the carriage into contact with the jaws.
2) Don't crank down too tight on the jaws. Extended even a single
step beyond the OD of the body puts an extra stress on them.
Or -- you could perhaps make extended soft jaws for the 3-jaw
chuck if it has two-piece jaws. You can extend them far enough to grip
the OD of the workpiece, and make them of aluminum so you are less
likely to damage the bed if you don't adjust things quite right.
Good Luck,
DoN.
--
Email: <dnic...@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---
> [ ... ]
>
>> BTW
>> can you part the piece out by having the parting tool at 90 degrees to
>> the
>> usual direction? Cutting into the piece at the right angles with the tool
>> lined up along the lathe bed? I have seen it done with a wood lathe.
>
> No! The blade of the parting tool would need to be curved like
> a parenthesis on a radius to match the cut to be made. This is called a
> trepaning tool. A standard parting tool would bind on the outside
> below the cut.
Makes sense! Like making a hole in the head. Do I take it that you need
various sizes trepanning tools to cut different radii?
I take it at this stage you have removed the pusher and relying solely on
the double sided sticky tape.
> Then, using a boring bar, bore through the workpiece slowly
> increasing the size until it is the right ID.
>
> Then -- put on the 3-jaw chuck (assuming that you don't need the
> OD to be truly concentric with the bored hole) expand the jaws inside th
> bored hole to grip the workpiece so you can turn the OD to something
> close to concentric.
From then on it is a piece of cake - it is that first step that I could not
solve without engaging in what I considered a rather sporting activity. Glad
to know it is an acceptable way to do it.
> If you need the two to be truly concentric, turn the OD while
> still holding the workpiece against the faceplate with the live enter
> and pusher adaptor. Then shift to boring the OD.
>
> Yes -- the super glue (or a bearing mount Loctite) would work in
> place of the double-sided tape. You will need to use heat to release
> it.
Right. I allowed for that. I thought the holding would be better with SG
than the tape.
> And -- while boring, it would probably help to have an extra
> sacrificial aluminum plate between the workpiece and the faceplate so you
> don't cut the faceplate.
I was thinking a recess as I have done with wood. But this is simpler.
>
> Of course if the plate which you are machining can tolerate some
> holes in its surface -- perhaps in places which would be machined larger
> later -- drill and tap for bolts to hold it to the faceplate..
>
> And yes -- you could use a rotary table and the double-stick
> tape or glue again on the mill to do the same thing. The finish of the
> bore will probably not be as nice, however.
That is what I thought. And using my boring head to 6" on my mini-mill is
probably pushing it :-) I did manage 3-1/8"...once!
>
> Note that you can probably use a larger diameter 4-jaw on the
> lathe than the 3-jaw because the jaws are easier to adapt to not stick
> out as far. My 12x24" Clausing has a 6-1/4" 3-jaw, and a 10" 4-jaw.
>
> In your particular described situation, a 9" swing lathe, and an
> 8" diameter workpiece, *if* your jaws for an 8" 4-jaw chuck have the
> last step only 1/2" thick, you could just barely grip the 8" diameter
> workpiece with the jaws extending only partially beyond the OD of the
> chuck. Be very careful when doing this:
>
> 1) Make sure that it does not hit the bed or the wings of the carriage,
> and that the tool can reach fully through the workpiece without
> bringing a part of the carriage into contact with the jaws.
>
> 2) Don't crank down too tight on the jaws. Extended even a single
> step beyond the OD of the body puts an extra stress on them.
OK, it sounds like additional reason to use the faceplate method.
I picked that example so you *could not* use the 4-jaw. I have run into that
problem several times. I really wanted to know what is an accepted way of
dealing with pieces that have a diameter approaching the lathe swing.
>
> Or -- you could perhaps make extended soft jaws for the 3-jaw
> chuck if it has two-piece jaws. You can extend them far enough to grip
> the OD of the workpiece, and make them of aluminum so you are less
> likely to damage the bed if you don't adjust things quite right.
Thanks a lot. I shall try this on a smaller scale (4") and be ready to duck
if things fly at 550 rpm :-)
One of my old books shows an offset spindle added to a lathe to turn
oversized disks. A gear or pulley on the main spindle turns the
faceplate on the new spindle at a reduced speed. The geometry is
difficult with pulleys but the planetary reduction off a burned-out
electric winch might fit well, bolt the large internal ring gear to
the faceplate and drive it with the small sun gear.
The added spindle goes above and behind the main one so the tool post
is still in the right front-rear position.
jsw
BTW, I forgot to mention earlier, wrt machining chuck adapter backplates for
spindles with flange mounting, the flanges generally have a protruding
shoulder (raised center section) which registers the chuck body to be
centered.
The shoulder also adds to the security of the mounting, so that the chuck
isn't just held in place by studs or bolts passing through the spindle
flange.
Machining a chuck backplate (or other accessory) would include cutting a
matching undercut recess to closely fit the flange feature.
--
WB
.........
metalworking projects
www.kwagmire.com/metal_proj.html
"Jim Wilkins" <kb1...@gmail.com> wrote in message
news:a3fbddd5-d08c-4344...@g23g2000vbl.googlegroups.com...
> On 2010-02-13, Michael Koblic <mko...@gmail.com> wrote:
... [re making a ring, 8" OD, 6.5" ID, from .25"-thick steel
plate, on a 9" lathe]
>> can you part the piece out by having the parting tool at 90 degrees to
>> the usual direction? Cutting into the piece at the right angles with
>> the tool lined up along the lathe bed? I have seen it done with a wood
>> lathe.
>
> No! The blade of the parting tool would need to be curved like
> a parenthesis on a radius to match the cut to be made. This is called a
> trepaning tool. A standard parting tool would bind on the outside below
> the cut.
>
Geometrically, a lot of other shapes would work, such as some
triangle and wedge cross sections. Of course, the fixed-radius
curved-cross-section trepanning tool that you mention probably
would be strongest and best for thin cuts. However, at radius
3.25" cutting .25" deep with a .5"-high cutoff blade, it would
work to grind a 5-degree [*] side relief (and 1 degree of back
relief, per <http://yarchive.net/metal/parting_off.html> toward
the end). Also, you might want to trepan to a slightly smaller
radius than the finish ID and then finish by boring, depending
on what gives a better or quicker result.
[*] For a trepanning blade of height h at radius r, the
interference at the outer bottom edge is about (h^2)/(2*r).
With a .5"-high blade, this comes to 0.038" at 3.25" radius
and to 0.042" at 3".
...
> Of course if the plate which you are machining can tolerate some
> holes in its surface -- perhaps in places which would be machined larger
> later -- drill and tap for bolts to hold it to the faceplate..
...
I've snipped DoN's other comments; take them as predecessors
to following suggestion: If the ring is for a sun-dial, holes
on the back won't show, so you could drill holes .15" deep to
seat upon pins in a mandrel plate attached to the faceplate.
The pins would withstand shear forces, allowing much heavier
cuts than you could take when depending on glue or tape to
withstand those forces. One could tap the holes, use loctite
and studs, etc, but presumably not worthwhile to do so.
--
jiw
Yes -- you could make one by grinding a HSS tool bit to the
proper radii -- including clearance.
[ ... ]
Yes. Keep the lathe speed slow, and the cut depth small to keep
the forces down.
Hmm ... for the drilling, you could make a special holder which
held a thrust bearing assembly with a bore larger than the drill which
you would use to start the hole. You could crank the carriage into
position (with the bearing well centered) to maintain extra pressure
while drilling, since you don't need the carriage for the actual
drilling.
Hmm ... another way to hold it fairly concentric (assuming that
your stock is truly round and near the desired size) would be a ring
which slips over the OD of the workpiece and (with less than 1/2" radial
size around the workpiece) and drill and tap four holes at 12:00
o'clock, 3:00 o'clock, 6:00 o'clock, and 9:00 o'clock and clamp it to
the faceplate with two steel bars going from 12:00 o'clock to 3:00
o'clock and from 6:00 o'clock to 9:00 o'clock so they don't interfere
with the spindle -- and the ends have to be cut to clear the bed. This,
perhaps augmented by the double-sided tape, would hold things firmly in
place.
Hmm ... two rings -- one which is a fit on the faceplate, and
the other on the (larger) workpiece, held together with bolts might make
getting things centered a bit easier.
Anyway -- with these approaches, you could use higher spindle
speeds than you would find safe with just the tape.
In any case -- use alcohol or acetone to scrub down the surface
of both the workpiece and the faceplate to make sure that they are both
free of oil and grease for maximum adhesion of the tape.
>> Then, using a boring bar, bore through the workpiece slowly
>> increasing the size until it is the right ID.
>>
>> Then -- put on the 3-jaw chuck (assuming that you don't need the
>> OD to be truly concentric with the bored hole) expand the jaws inside the
>> bored hole to grip the workpiece so you can turn the OD to something
>> close to concentric.
>
> From then on it is a piece of cake - it is that first step that I could not
> solve without engaging in what I considered a rather sporting activity. Glad
> to know it is an acceptable way to do it.
One of many ways. I've suggested some others above which take
more time making fixtures, but which make sense if you are making
multiples of these (e.g. the scales for your sundials).
>> If you need the two to be truly concentric, turn the OD while
>> still holding the workpiece against the faceplate with the live enter
>> and pusher adaptor. Then shift to boring the OD.
>>
>> Yes -- the super glue (or a bearing mount Loctite) would work in
>> place of the double-sided tape. You will need to use heat to release
>> it.
>
> Right. I allowed for that. I thought the holding would be better with SG
> than the tape.
It depends. super glue forms a rather brittle joint, so if you
have an interrupted cut or otherwise bump things you could break the
bond. The double-sided tape -- especially sufficient area of
double-sided foam tape -- will give a very good grip, and be more immune
to shocks.
[ ... ]
>> Of course if the plate which you are machining can tolerate some
>> holes in its surface -- perhaps in places which would be machined larger
>> later -- drill and tap for bolts to hold it to the faceplate..
>>
>> And yes -- you could use a rotary table and the double-stick
>> tape or glue again on the mill to do the same thing. The finish of the
>> bore will probably not be as nice, however.
>
> That is what I thought. And using my boring head to 6" on my mini-mill is
> probably pushing it :-) I did manage 3-1/8"...once!
Yes -- the torque needed goes up fast with increasing diameter.
The rotary table and a "slot drill" (two flute endmill) of
perhaps 1/4" diameter will probably work fairly well.
>> Note that you can probably use a larger diameter 4-jaw on the
>> lathe than the 3-jaw because the jaws are easier to adapt to not stick
>> out as far. My 12x24" Clausing has a 6-1/4" 3-jaw, and a 10" 4-jaw.
>>
>> In your particular described situation, a 9" swing lathe, and an
>> 8" diameter workpiece, *if* your jaws for an 8" 4-jaw chuck have the
>> last step only 1/2" thick, you could just barely grip the 8" diameter
>> workpiece with the jaws extending only partially beyond the OD of the
>> chuck. Be very careful when doing this:
>>
>> 1) Make sure that it does not hit the bed or the wings of the carriage,
>> and that the tool can reach fully through the workpiece without
>> bringing a part of the carriage into contact with the jaws.
>>
>> 2) Don't crank down too tight on the jaws. Extended even a single
>> step beyond the OD of the body puts an extra stress on them.
>
> OK, it sounds like additional reason to use the faceplate method.
Yes.
> I picked that example so you *could not* use the 4-jaw.
Actually -- you *could* -- barely -- with modified jaws. It is
just more difficult to make jaws for a 4-jaw than soft jaws for a 3-jaw
which is already fitted with two-piece jaws.
Think of the special soft jaws available for the Taig/Peatol
lathe. Those will handle a workpiece significantly larger than the
chuck body. For a 5" chuck -- just make large pie jaws to replace the
hardened top jaws -- *if* the machine comes with a chuck with two-part
jaws.
Hmm ... also, this sounds like an argument for a gap-bed lathe.
The first few inches of the bed can be unbolted and lifted out to allow
working with a larger diameter workpiece. There is some debate whether
the removed section can ever be put back as precisely as it was
positioned when the bed was first ground, but there are things where the
gap bed is a winner.
> I have run into that
> problem several times. I really wanted to know what is an accepted way of
> dealing with pieces that have a diameter approaching the lathe swing.
Lots of ways to do it -- depending on the lathe, and how many
of the same size workpiece you are making.
What you are making is relatively thin so you don't have to
worry about it clearing the carriage.
>> Or -- you could perhaps make extended soft jaws for the 3-jaw
>> chuck if it has two-piece jaws. You can extend them far enough to grip
>> the OD of the workpiece, and make them of aluminum so you are less
>> likely to damage the bed if you don't adjust things quite right.
>
> Thanks a lot. I shall try this on a smaller scale (4") and be ready to duck
> if things fly at 550 rpm :-)
The larger the diameter, the more grip surface you have between
the double-sided tape and the workpiece and faceplate.
But in any case, I would not consider 550 RPM to be a good idea.
First -- if the tape does let go, it could throw things rather
hard.
Second, 6.5" diameter means 935 SFM, which strikes me as a bit
fast unless you are using something like carbide tooling and a free
cutting mild steel like 12L14.
Based on the cutting speeds chart for HSS tooling in the
Jorgensen steel book:
Steel SFM
---------------
1018 125
12L14 300
4140 120 (annealed)
Stainless steels below
Steel SFM
---------------
304 75
416 160
Enjoy,
> Hmm ... for the drilling, you could make a special holder which
> held a thrust bearing assembly with a bore larger than the drill which
> you would use to start the hole. You could crank the carriage into
> position (with the bearing well centered) to maintain extra pressure
> while drilling, since you don't need the carriage for the actual
> drilling.
That is getting tricky...
> Hmm ... another way to hold it fairly concentric (assuming that
> your stock is truly round and near the desired size) would be a ring
> which slips over the OD of the workpiece and (with less than 1/2" radial
> size around the workpiece) and drill and tap four holes at 12:00
> o'clock, 3:00 o'clock, 6:00 o'clock, and 9:00 o'clock and clamp it to
> the faceplate with two steel bars going from 12:00 o'clock to 3:00
> o'clock and from 6:00 o'clock to 9:00 o'clock so they don't interfere
> with the spindle -- and the ends have to be cut to clear the bed. This,
> perhaps augmented by the double-sided tape, would hold things firmly in
> place.
I like that.
> Hmm ... two rings -- one which is a fit on the faceplate, and
> the other on the (larger) workpiece, held together with bolts might make
> getting things centered a bit easier.
>
> Anyway -- with these approaches, you could use higher spindle
> speeds than you would find safe with just the tape.
>
> In any case -- use alcohol or acetone to scrub down the surface
> of both the workpiece and the faceplate to make sure that they are both
> free of oil and grease for maximum adhesion of the tape.
Acetone for the faceplate. Alcohol for me...
[ ... ]
> It depends. super glue forms a rather brittle joint, so if you
> have an interrupted cut or otherwise bump things you could break the
> bond. The double-sided tape -- especially sufficient area of
> double-sided foam tape -- will give a very good grip, and be more immune
> to shocks.
I have not considered that aspect...
>
> [ ... ]
> Think of the special soft jaws available for the Taig/Peatol
> lathe.
I do. All the time :-)
Those will handle a workpiece significantly larger than the
> chuck body. For a 5" chuck -- just make large pie jaws to replace the
> hardened top jaws -- *if* the machine comes with a chuck with two-part
> jaws.
The problem arises with the Taig also. None of the chucks will hold a 4"
piece. I do the insides with a boring head on the mill and finish on the
Taig.
>
[ ... ]
>> Thanks a lot. I shall try this on a smaller scale (4") and be ready to
>> duck
>> if things fly at 550 rpm :-)
>
> The larger the diameter, the more grip surface you have between
> the double-sided tape and the workpiece and faceplate.
>
> But in any case, I would not consider 550 RPM to be a good idea.
>
> First -- if the tape does let go, it could throw things rather
> hard.
>
> Second, 6.5" diameter means 935 SFM, which strikes me as a bit
> fast unless you are using something like carbide tooling and a free
> cutting mild steel like 12L14.
No, no! I did say I shall try with *4"*. Yes, 550 rpm (the lowest the Taig
will go) is way too fast for that too, but I have managed so far. The only
way it works for facing is a sharp HSS tool with a small point radius. And a
very slow and steady hand as one gets to the outside. To do the inside I
need to bore only about 2" diameter. OTOH there is probably no point doing
that as the mill method works just fine and it will provide me with only
limited information about how the big piece would behave.
>
> Based on the cutting speeds chart for HSS tooling in the
> Jorgensen steel book:
>
> Steel SFM
> ---------------
> 1018 125
> 12L14 300
> 4140 120 (annealed)
>
> Stainless steels below
> Steel SFM
> ---------------
> 304 75
> 416 160
>
Granted. But if I followed that I would be cutting nothing bigger than 1"
diameters. The frightening thing about it is that to face the piece (1018)
in question (8" diameter) one should have a lathe with low speed of 60.
Back to cranking the handle...
<snip>
> Geometrically, a lot of other shapes would work, such as some
> triangle and wedge cross sections. Of course, the fixed-radius
> curved-cross-section trepanning tool that you mention probably
> would be strongest and best for thin cuts. However, at radius
> 3.25" cutting .25" deep with a .5"-high cutoff blade, it would
> work to grind a 5-degree [*] side relief (and 1 degree of back
> relief, per <http://yarchive.net/metal/parting_off.html> toward
> the end). Also, you might want to trepan to a slightly smaller
> radius than the finish ID and then finish by boring, depending
> on what gives a better or quicker result.
>
> [*] For a trepanning blade of height h at radius r, the
> interference at the outer bottom edge is about (h^2)/(2*r).
> With a .5"-high blade, this comes to 0.038" at 3.25" radius
> and to 0.042" at 3".
I will save this for future reference, thanks. Right now my plates are
pre-fabricated but who knows, in future this may be the way to go.
'
>> Of course if the plate which you are machining can tolerate some
>> holes in its surface -- perhaps in places which would be machined larger
>> later -- drill and tap for bolts to hold it to the faceplate..
> ...
>
> I've snipped DoN's other comments; take them as predecessors
> to following suggestion: If the ring is for a sun-dial, holes
> on the back won't show, so you could drill holes .15" deep to
> seat upon pins in a mandrel plate attached to the faceplate.
> The pins would withstand shear forces, allowing much heavier
> cuts than you could take when depending on glue or tape to
> withstand those forces. One could tap the holes, use loctite
> and studs, etc, but presumably not worthwhile to do so.
I routinely attach the armature to the back of the face just that way. It
took me a few attempts to reliably drill a 0.15" hole without going right
through (the current faces are 0.189" thick) and to drill deep enough to get
at least two threads of 10-32. In fact this was another way I contemplated
but I did not think of the shear pins which would certainly simplify
matters.
Much obliged, this is helpful.
I am rapidly going off the idea of purchasing a lathe with a flange, or
indeed anything "Craftex". Ask me why. Hint: I just bought a Craftex band
saw.
> "Jim Wilkins" <kb1...@gmail.com> wrote in message
> news:a3fbddd5-d08c-4344...@g23g2000vbl.googlegroups.com...
> On Feb 12, 9:09 pm, "Michael Koblic" <mkob...@gmail.com> wrote:
>> ...
>> Here is a question then: Say you have a 9" swing lathe and a 8" diameter,
>> ...
>> Michael Koblic,
>
> One of my old books shows an offset spindle added to a lathe to turn
> oversized disks. A gear or pulley on the main spindle turns the
> faceplate on the new spindle at a reduced speed. The geometry is
> difficult with pulleys but the planetary reduction off a burned-out
> electric winch might fit well, bolt the large internal ring gear to
> the faceplate and drive it with the small sun gear.
>
> The added spindle goes above and behind the main one so the tool post
> is still in the right front-rear position.
>
I should have kept the old RedNeck lathe...:-)
I've been reluctant to suggest the way I machine large arcs, but you
seem to have some experience now.
I attach a pivot post to the table of the mill, drill the blank to fit
on it, attach a long handle and manually rotate the blank into the end
mill. Yes, the end mill is likely to grab and try to spin the work,
more so if there is any play at the pivot. That isn't too serious
while the blank is still a polygon because the end mill quickly
reaches a gap. Once the arc is nearly continuous the depth of cut has
to be very small. I use a small endmill at low speed, extend the quill
all the way down and don't lock it so I can knock the handle upward,
and set the drive belts quite loose, but it's still as dangerous as
working freehand on a drill press. The worst one I've done this way
was a circular tee slot, after the milling force damaged my undersized
rotary table.
I mostly do this to round the ends of linkage bars and stop before the
arc is tangent to the straight edge.
http://picasaweb.google.com/KB1DAL/HomeMadeMachines#5265133136395165634
The wider hinge part was first clamped upright in the milling vise,
with a shaft through the hole to position it atop the jaws, and milled
almost to size across the top. A larger diameter piece could be
clamped to an angle plate. The disk will vibrate if not well supported
close to the cut.
If the disk can be clamped firmly to the mill table, raised on
spacers, it can be cut nearly to size much more safely by rotating it
slightly between cuts. The frequent reclamping is incredibly tedious
but it would get your 8" disk close enough to finish with a light cut
on the 9" lathe.
For even larger circles I set them up on a shaft and round them with
an angle grinder, held so it crosses at an angle and spins the disk as
it cuts. That's how I shaped the front tire for the sawmill. I haven't
tried yet, but I think a disk could be turned fairly accurately on a
wooden frame with a lathe bit on an X-Y table which moves crosswise,
parallel to the axle. Think of an old foot-pedal grindstone frame. You
could spin the disk with a sanding drum in a drill.
The right way is probably to buy the largest rotary table that fits on
your mill. The disadvantage compared to a gap-bed lathe is that you
can't detail the edges as much, you'll be limited to end mill profiles
like vee grooves and corner rounders.
jsw
Those would be the GT18 steering sectors?
> If the disk can be clamped firmly to the mill table, raised on
> spacers, it can be cut nearly to size much more safely by rotating it
> slightly between cuts. The frequent reclamping is incredibly tedious
> but it would get your 8" disk close enough to finish with a light cut
> on the 9" lathe.
I have to digest this...
> For even larger circles I set them up on a shaft and round them with
> an angle grinder, held so it crosses at an angle and spins the disk as
> it cuts. That's how I shaped the front tire for the sawmill. I haven't
> tried yet, but I think a disk could be turned fairly accurately on a
> wooden frame with a lathe bit on an X-Y table which moves crosswise,
> parallel to the axle. Think of an old foot-pedal grindstone frame. You
> could spin the disk with a sanding drum in a drill.
I think you mentioned this to me before. In fact I believe it was the
inspiration for this marvel of technology:
http://www.flickr.com/photos/27683124@N07/sets/72157607743618739/
I did not have as much luck facing with the grinder. I made some plastic
jaws for a chuck on my woodlathe and used an angle grinder with a sanding
attachment but for some reason it did not work too well either. I have a
feeling that the thing has to spin at a lower rate than 300 rpm to give the
grinder chance to work (I am talking face work here).
Still, the on-going problem is the *inside* of the doughnut.
> The right way is probably to buy the largest rotary table that fits on
> your mill. The disadvantage compared to a gap-bed lathe is that you
> can't detail the edges as much, you'll be limited to end mill profiles
> like vee grooves and corner rounders.
I am certainly considering it as a viable option. Realistically I cannot get
anything bigger than 6" rotary table on my mill and even that is probably
pushing it. OTOH if one could do things that way it is a $1000 difference in
capital expenditure :-)
BTW can you put a motor drive on a rotary table so it turns at a constant
rate?
[ ... ]
>> Hmm ... another way to hold it fairly concentric (assuming that
>> your stock is truly round and near the desired size) would be a ring
>> which slips over the OD of the workpiece and (with less than 1/2" radial
>> size around the workpiece) and drill and tap four holes at 12:00
>> o'clock, 3:00 o'clock, 6:00 o'clock, and 9:00 o'clock and clamp it to
>> the faceplate with two steel bars going from 12:00 o'clock to 3:00
>> o'clock and from 6:00 o'clock to 9:00 o'clock so they don't interfere
>> with the spindle -- and the ends have to be cut to clear the bed. This,
>> perhaps augmented by the double-sided tape, would hold things firmly in
>> place.
>
> I like that.
[ ... ]
>> Anyway -- with these approaches, you could use higher spindle
>> speeds than you would find safe with just the tape.
>>
>> In any case -- use alcohol or acetone to scrub down the surface
>> of both the workpiece and the faceplate to make sure that they are both
>> free of oil and grease for maximum adhesion of the tape.
>
> Acetone for the faceplate. Alcohol for me...
O.K. But the latter *after* you are done with the machining for
the day. :-)
[ ... ]
>> Think of the special soft jaws available for the Taig/Peatol
>> lathe.
>
> I do. All the time :-)
>
> Those will handle a workpiece significantly larger than the
>> chuck body. For a 5" chuck -- just make large pie jaws to replace the
>> hardened top jaws -- *if* the machine comes with a chuck with two-part
>> jaws.
>
> The problem arises with the Taig also. None of the chucks will hold a 4"
> piece. I do the insides with a boring head on the mill and finish on the
> Taig.
Hmm ... I would have to go down and set up the chuck with the
extruded jaws to be sure -- but remember that you can bore a recess in
the jaws up to within perhaps 1/4" from the ends of the fingers.
But you could *make* pie jaws from solid aluminum plate at need
to have the extra reach needed.
And if the swing of the lathe is the problem, add one or two
riser blocks between the headstock and the bed. Those riser blocks will
stack if you have (or make) two of them.
[ ... ]
>> But in any case, I would not consider 550 RPM to be a good idea.
>>
>> First -- if the tape does let go, it could throw things rather
>> hard.
>>
>> Second, 6.5" diameter means 935 SFM, which strikes me as a bit
>> fast unless you are using something like carbide tooling and a free
>> cutting mild steel like 12L14.
>
> No, no! I did say I shall try with *4"*. Yes, 550 rpm (the lowest the Taig
> will go) is way too fast for that too, but I have managed so far.
Hmm ... replace the motor with a DC motor and a proper speed
controller, or a small three-phase motor and a VFD to run the motor at
perhaps 1/4 speed. (about 140 RPM).
> The only
> way it works for facing is a sharp HSS tool with a small point radius. And a
> very slow and steady hand as one gets to the outside. To do the inside I
> need to bore only about 2" diameter. OTOH there is probably no point doing
> that as the mill method works just fine and it will provide me with only
> limited information about how the big piece would behave.
O.K.
>> Based on the cutting speeds chart for HSS tooling in the
>> Jorgensen steel book:
>>
>> Steel SFM
>> ---------------
>> 1018 125
>> 12L14 300
>> 4140 120 (annealed)
>>
>> Stainless steels below
>> Steel SFM
>> ---------------
>> 304 75
>> 416 160
>>
> Granted. But if I followed that I would be cutting nothing bigger than 1"
> diameters.
Or -- you would swap in motors which could be run slow enough to
get proper speeds.
Hmm .... the current motor has what RPM? 1800 RPM, 3600 RPM?
There are 900 RPM (four pole) motors which will get your first reduction
in speed.
> The frightening thing about it is that to face the piece (1018)
> in question (8" diameter) one should have a lathe with low speed of 60.
> Back to cranking the handle...
Or -- to a proper variable speed motor -- either three phase, or
DC.
Or -- use uncoated carbide inserts to survive the higher speeds.
[ ... ]
>> The right way is probably to buy the largest rotary table that fits on
>> your mill. The disadvantage compared to a gap-bed lathe is that you
>> can't detail the edges as much, you'll be limited to end mill profiles
>> like vee grooves and corner rounders.
>
> I am certainly considering it as a viable option. Realistically I cannot get
> anything bigger than 6" rotary table on my mill and even that is probably
> pushing it. OTOH if one could do things that way it is a $1000 difference in
> capital expenditure :-)
>
> BTW can you put a motor drive on a rotary table so it turns at a constant
> rate?
There is a stepper motor drive for the little Sherline rotary
table, which can be used with a controller/keypad:
<http://www.sherline.com/8730pg.htm>
<http://www.sherline.com/8700inst.htm>
With a CNC milling machine, you could even use it for engraving
the time markings on your dials.
<snip>
>> The problem arises with the Taig also. None of the chucks will hold a 4"
>> piece. I do the insides with a boring head on the mill and finish on the
>> Taig.
>
> Hmm ... I would have to go down and set up the chuck with the
> extruded jaws to be sure -- but remember that you can bore a recess in
> the jaws up to within perhaps 1/4" from the ends of the fingers.
>
> But you could *make* pie jaws from solid aluminum plate at need
> to have the extra reach needed.
The Taig swing is 4-1/8" ...
> And if the swing of the lathe is the problem, add one or two
> riser blocks between the headstock and the bed. Those riser blocks will
> stack if you have (or make) two of them.
I considered that. Those who know better than I told me that the rigidity
declines quite significantly. Machining 4" with the existing setup is a
constant struggle with rigidity as it is.
>
> [ ... ]
> Hmm ... replace the motor with a DC motor and a proper speed
> controller, or a small three-phase motor and a VFD to run the motor at
> perhaps 1/4 speed. (about 140 RPM).
Or for the same money just get a bigger lathe. I went through this some time
ago. The torque at the low speed is a big problem with the Taig. Even as it
is I have to lean on the motor some times to stop it stalling. It is what it
is: A little lathe for little things.
<snip>
> Or -- use uncoated carbide inserts to survive the higher speeds.
I tried those. I thought they were horrible. Got much better results with
properly ground HSS.
This is all done on the un-modified Taig:
http://www.flickr.com/photos/27683124@N07/4170484793/in/set-72157622965185636/
(well, just the step - much milling there also)
http://www.flickr.com/photos/27683124@N07/4177267315/in/set-72157605638700703/
http://www.flickr.com/photos/27683124@N07/3066407095/in/set-72157605638700703/
http://www.flickr.com/photos/27683124@N07/4358722420/
There has been a learning curve :-)
http://www.flickr.com/photos/27683124@N07/3954474060/in/set-72157605638700703/
In any case, I really have no idea what is a reasonable expectation of a
good finish achievable by a lathe of whatever size. This 1' beast was turned
by a professional. The concentric ridges are apparent. Can one ever get rid
of them? Can one make them look uniform without a power-feed? Should one go
back to facing with sandpaper and just do the edges on a lathe/mill? What is
the meaning of life...
I was using a motorized RT for a while, and it seemed to work very well
(hobby, not production speeds & feeds). I used a small Oriental Motor
gearmotor, chain and sprockets and a Phase II 6" horizontal RT.
The RT was turning very slowly, less than 2 RPM, maybe considerably slower.
It's a method I wanted to try, for cutting disks from sheet/plate stock
without center holes.
--
WB
.........
"Michael Koblic" <mko...@gmail.com> wrote in message
news:hlaiu...@news6.newsguy.com...
>
snips
That piece fit on the lathe after sawing it roughly to size. I clamped
the square blank flat on the bandsaw table with tee bolts to cut off
the corners. Then I set the saw upright and followed a scribed circle
as far as the wide blade allows, then opened up the kerf with a chisel
to make clearance the next cut.
> > If the disk can be clamped firmly to the mill table, ...
> I have to digest this...
The upright pivot post is held in the vise, the disk rests on 1-2-3
blocks which raise it above the jaw tops. Tee slot clamp bolts clamp
the disk onto the 1-2-3 blocks, leaving a small space in between to
mill tangents to the OD. It's slower than sawing because of all the
reclamping but doesn't risk cutting in too far by accident, and leaves
only a little metal for the lathe to remove.
> > For even larger circles I set them up on a shaft and round them with
> > an angle grinder, ...Think of an old foot-pedal grindstone frame. You
> > could spin the disk with a sanding drum in a drill.
>
> I think you mentioned this to me before. In fact I believe it was the
> inspiration for this marvel of technology:
> http://www.flickr.com/photos/27683124@N07/sets/72157607743618739/
> Michael Koblic,
The ladder-shaped frame of my sawmill is the model. Both wheels needed
truing and the front one considerable flattening. I used the
crosspieces at the inner ends of the wheels for tool rests, ie to hold
the grinder steady.
You could make a similar frame out of wood by laminating 2x4's with
the ends alternating to form glued finger joints. I'd make two side-by-
side rectangular bays, the second providing a tool rest for faceplate
turning.
The faceplate could be a large pulley faced with plywood. The bearings
on the motorcycle wheels were tight enough to use one as a lathe
faceplate, though the protruding axle interferes. If you did that you
could drive the tire by friction or remove it and run a vee belt on
the rim. I used the sprocket only because I needed to transmit several
horsepower, it required a separate countershaft.
I think an X-Y table would be stiff enough for a tool rest. I acquired
one plus the headstock from a scrapped SB lathe and a brake lathe
tailstock to build a large wheel lathe, but haven't needed to set it
up yet.
jsw
So, it bears further consideration. I guess the next question is "How fast
can you rotate a rotary table?" :-)
I should be able to the very same tomorrow (once I found where the knocking
noise is coming out of the Craftex). BTW, when you got your band saw, did
you change the oil in the gearbox right away or did you trust the Chinese
and left it for a year or so?
>> > If the disk can be clamped firmly to the mill table, ...
>
>> I have to digest this...
>
> The upright pivot post is held in the vise, the disk rests on 1-2-3
> blocks which raise it above the jaw tops. Tee slot clamp bolts clamp
> the disk onto the 1-2-3 blocks, leaving a small space in between to
> mill tangents to the OD. It's slower than sawing because of all the
> reclamping but doesn't risk cutting in too far by accident, and leaves
> only a little metal for the lathe to remove.
Got it.
Another one to digest :-)
I was thinking that if I return to red neck technology I would probably make
the ring spin horizontally this time. But anything will be considered. The
important thing is to spin the work piece quite slowly otherwise the grinder
works quite inefficiently, at least on the face. The RedNeck lathe v5.01
went at less than 100 rpm and could have gone slower.
I found that an X-Y *vise* was capable of holding lathe tools quite firmly
and take large cuts. Here it is with a boring bar holder I made:
http://www.flickr.com/photos/27683124@N07/3715470146/sizes/o/in/set-72157614869315843/
I have to add that the wobble was all in the chuck: The spindle had a runout
of only about a 0.001" :-)
That 4-jaw chuck looks like the one that's supplied with the 9x20 lathes.
The wobble doesn't exist on the lathe spindle because the spindle includes a
raised shoulder feature to register the chuck at a nearly perfectly square
position.
I read quite a few remarks from disappointed new 9x20 lathe owners, wrt the
4-jaw chuck. I found the chuck to be suitable for a lathe of that size and
power.
It's not like a 9x20 is intended to turn a 50 pound workpiece at 1000 RPM.
--
WB
.........
"Michael Koblic" <mko...@gmail.com> wrote in message
news:hld0g...@news4.newsguy.com...
Some RTs are 40:1, others are higher turns per rotation. Worm reductions
typically have slow output speeds although heavy duty models can handle some
serious loads.
If one were inclined, the worm of a RT can be disengaged in many models (or
removed), so the table stem could be driven from the back side or bottom,
depending upon the orienation (of the RT, not the operator).
As long as the RT isn't routinely subjected to the side loads of turning
down diameters, push knurling and other typical lathe operations, a RT may
be suitable (just speculation) for occasional cutting operations similar to
facing cuts on lathes.
If the table's stem support/bearing surfaces wear, the accuracy of the RT is
diminished, so an overhaul may be necessary, possibly requiring boring and a
bushing or sleeve installation.
--
WB
.........
"Michael Koblic" <mko...@gmail.com> wrote in message
news:hlcv8...@news4.newsguy.com...
> I found that an X-Y *vise* was capable of holding lathe tools quite firmly
> and take large cuts. Here it is with a boring bar holder I made:
>
> http://www.flickr.com/photos/27683124@N07/3715470146/sizes/o/in/set-7...
> Michael Koblic,
Very nice!
I traded away my X-Y vise and couldn't remember how far the ways
protruded beyond the jaws, so I didn't mention it. The one you have
looks to be of better quality, I bought mine around 1980 and it may
have been cast from leftover Great Leap Forward backyard iron.
jsw
I bought a second-hand Delta 4x6, so I'm not sure what the gearbox had
in it originally. It was full though black when I changed it, after
reading warnings here. The new oil has stayed fairly clean.
It has adjustable blade guides and the Delta logo on some small parts
so they did add some value over the generic saws. I strengthened the
base and put on larger wheels but otherwise it was reasonably well
made and cuts straight as long as it's on the flat basement floor.
When I use it outdoors on the uneven driveway I have to move it around
until both sides of the handle end of the base touch down at the same
time, and then check that the blade rises perpendicular to the table
with a square. That's OK since the outdoor cuts are structural steel
for welding and don't need to be as accurate. Indoors it cuts square
vertically to within ~0.005" per 1".
jsw.
I did not complain while I was using it. I got it on sale - cannot remember
where.
I dwonloaded two manuals in addtion to the "manual" that came with the saw.
Essentially the same item but differeing advice. Two had me put in "140
weight gear oil", one had me use grease.
Given the state this thing was out of the box I thought I better open up the
gear box and change the oil anyway. The oil was relatively clean. Then I
found that there is no such thing as "140 weight etc." In the end I gave it
a common 80W-90. Worm gears are only GL2 classification and if the oil is
good enough for everything else I am sure it will not kill the saw. In any
case I could not see the saw starting up freely in my cold garage with 140
weight in it.
> It has adjustable blade guides and the Delta logo on some small parts
> so they did add some value over the generic saws. I strengthened the
> base and put on larger wheels but otherwise it was reasonably well
> made and cuts straight as long as it's on the flat basement floor.
> When I use it outdoors on the uneven driveway I have to move it around
> until both sides of the handle end of the base touch down at the same
> time, and then check that the blade rises perpendicular to the table
> with a square. That's OK since the outdoor cuts are structural steel
> for welding and don't need to be as accurate. Indoors it cuts square
> vertically to within ~0.005" per 1".
Nice. Mine is very basic. It does not even lock in the vertical position.
Having said that the first two cuts (one before and one after I tweaked the
guide bearings) were almost identical and within (eyeball!) 0.010" on a 1.25
round mild steel. I cannot see wanting more accuracy than that. In any case
the accuracy will be limited by the vise which will need sorting out in good
time.
--
Michael Koblic,
Campbell River, BC
PS Finally I had a *good* look at your saw-mill. I do not think I understood
it before! Some piece of machinery!
PPS Apropos a comment you made elsewhere, Joss Whedon used to be great but I
was disappointed with the season 2 of the "Dollhouse".
I added extra holes to the fixed vise jaw to move it closer to the
blade for sawing small pieces.
>
> PS Finally I had a *good* look at your saw-mill. I do not think I understood
> it before! Some piece of machinery!
Thanks.
> PPS Apropos a comment you made elsewhere, Joss Whedon used to be great but I
> was disappointed with the season 2 of the "Dollhouse".
He can write and direct a scene that's a gem of perfection, with
dialog and acting worthy of Tennessee Williams, but rarely can he
extend it to the whole episode. He writes for literate theatre/film
buffs who also understand science, and quite a bit of the humor goes
over the heads of a mass audience, like " ..makes Godot seem
punctual". I had to explain to the rest of the internet fans that the
time-stopping experiment on "Angel" was a Bose-Einstein condensate.
For a while Fox here ran Buffy and MASH reruns back to back, making it
easy to see how similar their best moments were, also that MASH was
forgettably weak at times too.
John Locke's current incarnation on "Lost" is straight from Buffy,
where the "First Evil", assumed but never stated to be the devil, can
take the form of a dead person and tries to recruit followers or
subvert opponents by telling them whatever they want to hear.
jsw
Fabricating chuck adapter plates is just one of those things that lathe
users do when required, for flanged or threaded spindles.
A chuck won't come unthreaded from a flange while running the lathe in
reverse, but it happens with threaded spindles once in a while (usually only
once to some operators).
Another collet option ocurred to me (I'm slow).. ER collets could be used if
one were to make an adapter/collet chuck to mount to the spindle flange.
The overall length of the ER adapter/chuck would subtract from the
center-to-center spec of the lathe.
The amount of the extention away from the spindle nose bearing would depend
on the holder design and the length of the collets.
The holder and ER collets would allow long stock to pass thru the headsock.
--
WB
.........
"Michael Koblic" <mko...@gmail.com> wrote in message
news:hkti4...@news3.newsguy.com...
> This here looks a rather attractive machine particularly as it is on sale:
>
> http://busybeetools.ca/cgi-bin/picture10?NTITEM=B2227L
>
> Trying to find out more about it I looked at the manual but it is quite
> useless. One of the questions I had was how is the chuck attached to the
> spindle. The dealer tells me that there is a "flange" but any further
> details have not been forthcoming so far.
>
> Are there any standards that would describe such flange attachment? Are
> there standard sizes? I am concerned about accessories (5C collet
> chuck??). This machine is BTW only superficially similar to the Grizzly
> G0602 which I understand has a 1-3/4 - 8TPI spindle.
Backplates are simply interchangeable flange mounts. Threads aren't
superior, just easier to cut at home than tapers.
> Another collet option ocurred to me (I'm slow).. ER collets could be used if
> one were to make an adapter/collet chuck to mount to the spindle flange....
> WB
http://littlemachineshop.com/products/product_view.php?ProductID=2532&category=-421559299
Or center an ER extension chuck in the 4-jaw.
I wrote "John Locke's current incarnation on "Lost" is straight from
Buffy" and then saw this:
http://www.movieline.com/2010/02/buffy-lost-similarities.php?page=all
It's fair, Whedon lifted plot and character ideas freely and blatantly
from Star Trek, X-Files, Kubrick and Alien(s), which he helped write.
Don't tell the lawyers, I think Buffy was his version of Ellen Ripley,
a character who well deserved extended development in a TV series. He
swapped magic for the tech and demons for space aliens, but used them
in all the same ways, including a self-aware creation like Moriarty,
an ex-demon girl as clueless about humanity as Data, 3 goofy geeks
(clone gunmen), and Quark as the high school principal. "Spike" in the
article above is a vampire Alex from A Clockwork Orange.
jsw
<snip>
> I added extra holes to the fixed vise jaw to move it closer to the
> blade for sawing small pieces.
Also a jack screw might be in order. And maybe drill and tap the table to
cut small flat pieces.
>> PS Finally I had a *good* look at your saw-mill. I do not think I
>> understood
>> it before! Some piece of machinery!
>
> Thanks.
>
>> PPS Apropos a comment you made elsewhere, Joss Whedon used to be great
>> but I
>> was disappointed with the season 2 of the "Dollhouse".
>
> He can write and direct a scene that's a gem of perfection, with
> dialog and acting worthy of Tennessee Williams, but rarely can he
> extend it to the whole episode. He writes for literate theatre/film
> buffs who also understand science, and quite a bit of the humor goes
> over the heads of a mass audience, like " ..makes Godot seem
> punctual". I had to explain to the rest of the internet fans that the
> time-stopping experiment on "Angel" was a Bose-Einstein condensate.
>
> For a while Fox here ran Buffy and MASH reruns back to back, making it
> easy to see how similar their best moments were, also that MASH was
> forgettably weak at times too.
>
> John Locke's current incarnation on "Lost" is straight from Buffy,
> where the "First Evil", assumed but never stated to be the devil, can
> take the form of a dead person and tries to recruit followers or
> subvert opponents by telling them whatever they want to hear.
>
I suffered through the first series of Lost and doubt I shall ever see the
rest of them.
Funny story how I came to watch Buffy: An ex-banger who now teaches
self-defense mentioned he liked the Buffy series. I was intrigued why a
person of his pedigree would like such a program. Turns out he really liked
the English stunt girl who did SMG's work in the middle series (before she
married the stunt coordinator). Indeed she was brilliant. In the process of
watching it I realized that this was an extraordinarily well written and
acted series and became an addict. Angel and Serenity followed. I am sorry
they screwed around with the Serenity.
I hope he rebounds from the "Dollhouse". There is being intelligent and then
there is being self-indulgent. Dushku did not help...
But a chuck like that would be a good project and a handy accessory for a
lathe with a spindle flange.
The LMS chuck looks like it's aluminum, the weight shown is 2.43 lbs
(although ya can never tell what given weighs are, actual part weight or
shipping weight).
The lathe model that Michael referred to earlier, had a MT4 taper and a 1"
bore (5" flange), whereas the 7x mini has a MT3 and a 3/4" bore (with a 3"
flange).
There are MT3 shank to ER chucks, so maybe there are also MT4 to ER, but
they wouldn't have the stock pass-thru capability.
I'm not familiar with ER series collet overall sizes or capacities, but I
was looking at C3 collets a while ago, which can be used with a MT3 spindle
taper by adding a fairly inexpensive (~$35) adapter (and fabricating a
drawtube, or LMS sells drawtubes).
I suppose that if one were inclined, a MT3 to C3 adaper could be made from a
large endmill holder (cut down, bore thru). The same plan could work for
fabricating an ER chuck on a MT taper, maybe.
The 7x mini I bought a while ago, and the 9x20 I've had for several years,
both have MT3 spindle tapers.
I do have a 10-piece set of MT3 collets, but C3s are a self-releasing type,
which I would prefer.
Your perception into details of shows is somewhat astounding to me. I don't
watch many TV shows because there aren't many good ones.
The last series that I really enjoyed was X-Files, and Lost intrigued me at
first, but I haven't watched it since season 2.
The creator of X-Files said his enjoyment of the old series Night Stalker,
in part, led to the creation of X-Files.
--
WB
.........
"Jim Wilkins" <kb1...@gmail.com> wrote in message
news:a34b40ad-5fd1-4555...@x22g2000yqx.googlegroups.com...
Backplates are simply interchangeable flange mounts. Threads aren't
superior, just easier to cut at home than tapers.
http://littlemachineshop.com/products/product_view.php?ProductID=2532&category=-421559299
Pass through is limited to the vendor and strength. Neither one is a pass
through, but I suspect the holder is stronger best.
Martin
> > http://littlemachineshop.com/products/product_view.php?ProductID=2532...
> > Or center an ER extension chuck in the 4-jaw.
> The LMS ER-32 collet chuck at $200 is a fairly expensive accessory for a 7x
> mini-lathe, especially when a collet set is somewhat costly. But it does
> include a wrench.
>
> But a chuck like that would be a good project and a handy accessory for a
> lathe with a spindle flange.
Get good 3 and 4 jaw chucks and use the lathe, collets are nice and
convenient but not necessary.
...
> Your perception into details of shows is somewhat astounding to me.
> WB
Not mine so much, I read reviews partly to analyse and learn the
writer's style. I used to work part-time on film and theatre crews and
thus notice details of acting, directing, sets, props and lighting.
Whedon's DVD commentaries are nearly a master class in the art of film
making.
This is the stunt woman:
http://www.sophiacrawford.com/sophia-crawford-bio.htm
I suffered through a few minutes of Power Rangers to see their
superhuman gymnastic abilities.
Whedon allowed the backstage crew to communicate with fans on the
Internet as long as they didn't discuss plots, and the unaccustomed
fame they received went to a few heads including hers and Jeff
Pruitt's. OTOH the writers were fascinating and soon grabbed the fans'
attention.
http://www.whedon.info/Drew-Goddard-At-The-Bronze-Beta,415.html
He's the grandson of the rocket Goddard and well educated in science,
which he writes into "Lost". "Minions" are his fan base.
jsw
<snip>
> This is the stunt woman:
> http://www.sophiacrawford.com/sophia-crawford-bio.htm
> I suffered through a few minutes of Power Rangers to see their
> superhuman gymnastic abilities.
Great body mechanics.
> Whedon allowed the backstage crew to communicate with fans on the
> Internet as long as they didn't discuss plots, and the unaccustomed
> fame they received went to a few heads including hers and Jeff
> Pruitt's.
That maybe but her replacement was not a patch on her.
BTW is this your IMDb entry?
http://www.imdb.com/name/nm0929273/
Not me, by then I was building expensive toys for the Air Force to
play with. I worked for a student filmmaker in the 60's and a USO and
then community theatre group in the 70's, all part time. I ran the
projector for the classes the filmmaker taught and learned a lot.
We had our own imitation of Warhol's "factory" in the off-campus
hippie village for a while, but I got no nearer to the craziness than
holding the lights during filming. The talented hotties have all the
fun with each other, the crew only gets to watch. All this while
studying for a Chemistry degree. The only course I couldn't pass was
technical writing, which is why I practice here.
In addition to Joss Whedon I'd like to see Brian Fuller and Baz
Luhrmann succeed with a mass audience. You'll never look at Michael
from Lost the same way again after seeing how Luhrmann had him play
Romeo's friend Mercutio IN DRAG. Don't watch it if guns offend you.
The two families were like rival crime syndicates, better armed than
Mad Max, and even Juliet packed a Walther.
jsw
Here he is with the guy who played Romeo:
http://content8.flixster.com/question/57/92/09/5792094_std.jpg
http://blog.morgane.org/blog/images/perso/996RNJ_Harold_Perrineau_Jr_001.jpg
No putting my name under those
I quit film-making in the 60's too :-)
> We had our own imitation of Warhol's "factory" in the off-campus
> hippie village for a while, but I got no nearer to the craziness than
> holding the lights during filming. The talented hotties have all the
> fun with each other, the crew only gets to watch.
Not to worry, it's over-rated...
All this while
> studying for a Chemistry degree. The only course I couldn't pass was
> technical writing, which is why I practice here.
>
> In addition to Joss Whedon I'd like to see Brian Fuller and Baz
> Luhrmann succeed with a mass audience. You'll never look at Michael
> from Lost the same way again after seeing how Luhrmann had him play
> Romeo's friend Mercutio IN DRAG. Don't watch it if guns offend you.
> The two families were like rival crime syndicates, better armed than
> Mad Max, and even Juliet packed a Walther.
I remember a production of R+J in the 60's with leather jackets and knives.
Many people thought it was shocking. The guy who played Romeo then is now
the Czech equivalent of Anthony Hopkins (Jan Triska).
And I will never look at Lost again, period. By the end of season one I was
hoping for all of them to die. Horribly. Particularly the lead.
A bit like the Inglourious Basterds - when you start rooting for the Nazis
you know there is something wrong with the film.
--