I'm so very confused.

I have 3/8" shank holders with HSS inserts. I have 1/32 and 1/64
radius inserts.

I was testing the cutting straightness of the lathe with a
1" steel bar over 5". The results I got were 2 thou bigger
at the headstock than at the free end.

I then had a professional in to look at my issue. He said the
bit I was using wouldn't give me proper results. I was using the 1/32,
and he said it needed to be at the biggest 1/64. He ground a
HSS blank with a knife edge. Then he tested using 1" aluminum.
He then aligned the headstock, which he said wasn't that far off.
(On a previous visit he straightened/levelled the bed).
When he got done it was straight within a few thou over 4" or so.

After he left I tried my test again using the 1/64 bit. With the
steel bar it was still bigger at the headstock, but only by .001.
Using the aluminum bar it was off by .0005.

On both bars after cutting toward the headstock, I left the bit
where it was, and reversed the feed toward the tailstock.
More material was cut off. When I measured them, it was only off
a few thou.

I don't know how it could be larger at the headstock, and then
come out straight on the 2nd cut. And what good are indexable inserts
if they don't cut straight?

Once again I'm so confused.
Wayne

Different turning bits may require more or less cutting pressure. The
difference may lessen your problem or make it worse, depending on which bit
you use. Same goes for steel vs. alum. The fact that your lathe turns a
taper is still there, just gets worse depending on the circumstances.

Tony
"Wayne" wrote in message
om...
I'm so very confused.

I have 3/8" shank holders with HSS inserts. I have 1/32 and 1/64
radius inserts.

I was testing the cutting straightness of the lathe with a
1" steel bar over 5". The results I got were 2 thou bigger
at the headstock than at the free end.

I then had a professional in to look at my issue. He said the
bit I was using wouldn't give me proper results. I was using the 1/32,
and he said it needed to be at the biggest 1/64. He ground a
HSS blank with a knife edge. Then he tested using 1" aluminum.
He then aligned the headstock, which he said wasn't that far off.
(On a previous visit he straightened/levelled the bed).
When he got done it was straight within a few thou over 4" or so.

After he left I tried my test again using the 1/64 bit. With the
steel bar it was still bigger at the headstock, but only by .001.
Using the aluminum bar it was off by .0005.

On both bars after cutting toward the headstock, I left the bit
where it was, and reversed the feed toward the tailstock.
More material was cut off. When I measured them, it was only off
a few thou.

I don't know how it could be larger at the headstock, and then
come out straight on the 2nd cut. And what good are indexable inserts
if they don't cut straight?

Once again I'm so confused.
Wayne

(Wayne) wrote in news:51701a28.0407271631.2782c0c9
@posting.google.com:


After he left I tried my test again using the 1/64 bit. With the
steel bar it was still bigger at the headstock, but only by .001.
Using the aluminum bar it was off by .0005.

So if you are referring to being off by 0.001 diameter, it is out of
alignment by 0.0005 over 5"



On both bars after cutting toward the headstock, I left the bit
where it was, and reversed the feed toward the tailstock.
More material was cut off. When I measured them, it was only off
a few thou.


The extra material removal is from the flex in the tool/toolpost/set-up
when taking the inital cut. The information about taper when doing this
is irrelevant and unuseful due to that fact. You would need to be able
to measure the differences in the flex of the tool tip in both directions
in order to obtain any taper information.


I don't know how it could be larger at the headstock, and then
come out straight on the 2nd cut. And what good are indexable inserts
if they don't cut straight?

Once again I'm so confused.
Wayne







--
Anthony

You can't 'idiot proof' anything....every time you try, they just make
better idiots.

Remove sp to reply via email

Anthony wrote in
:

(Wayne) wrote in
news:51701a28.0407271631.2782c0c9 @posting.google.com:


After he left I tried my test again using the 1/64 bit. With the
steel bar it was still bigger at the headstock, but only by .001.
Using the aluminum bar it was off by .0005.

So if you are referring to being off by 0.001 diameter, it is out of
alignment by 0.0005 over 5"



On both bars after cutting toward the headstock, I left the bit
where it was, and reversed the feed toward the tailstock.
More material was cut off. When I measured them, it was only off
a few thou.


The extra material removal is from the flex in the
tool/toolpost/set-up when taking the inital cut. The information
about taper when doing this is irrelevant and unuseful due to that
fact. You would need to be able to measure the differences in the
flex of the tool tip in both directions in order to obtain any taper
information.



Ohh..One other thing:
a piece of 1" bar stock, freely ended 5x Ø out of the chuck *is* going to
deflect under cutting load. You may not have any taper in the machine at
all.



--
Anthony

You can't 'idiot proof' anything....every time you try, they just make
better idiots.

Remove sp to reply via email

Anthony wrote in
:

Anthony wrote in
:

(Wayne) wrote in
news:51701a28.0407271631.2782c0c9 @posting.google.com:


After he left I tried my test again using the 1/64 bit. With the
steel bar it was still bigger at the headstock, but only by .001.
Using the aluminum bar it was off by .0005.

So if you are referring to being off by 0.001 diameter, it is out of
alignment by 0.0005 over 5"



On both bars after cutting toward the headstock, I left the bit
where it was, and reversed the feed toward the tailstock.
More material was cut off. When I measured them, it was only off
a few thou.


The extra material removal is from the flex in the
tool/toolpost/set-up when taking the inital cut. The information
about taper when doing this is irrelevant and unuseful due to that
fact. You would need to be able to measure the differences in the
flex of the tool tip in both directions in order to obtain any taper
information.



Ohh..One other thing:
a piece of 1" bar stock, freely ended 5x Ø out of the chuck *is* going
to deflect under cutting load. You may not have any taper in the
machine at all.




Oops...failed to notice you said *larger* at the headstock. That would
not be material flex. It could be the slide running uphill or downhill
to the spindle though.




--
Anthony

You can't 'idiot proof' anything....every time you try, they just make
better idiots.

Remove sp to reply via email

Consider the possibility that you have a little slop in your headstock. The
larger radius tool will push the stock more out-of-center giving you a taper
which is larger at the headstock.

"Wayne" wrote in message
om...
I'm so very confused.

I have 3/8" shank holders with HSS inserts. I have 1/32 and 1/64
radius inserts.

I was testing the cutting straightness of the lathe with a
1" steel bar over 5". The results I got were 2 thou bigger
at the headstock than at the free end.

I then had a professional in to look at my issue. He said the
bit I was using wouldn't give me proper results. I was using the 1/32,
and he said it needed to be at the biggest 1/64. He ground a
HSS blank with a knife edge. Then he tested using 1" aluminum.
He then aligned the headstock, which he said wasn't that far off.
(On a previous visit he straightened/levelled the bed).
When he got done it was straight within a few thou over 4" or so.

After he left I tried my test again using the 1/64 bit. With the
steel bar it was still bigger at the headstock, but only by .001.
Using the aluminum bar it was off by .0005.

On both bars after cutting toward the headstock, I left the bit
where it was, and reversed the feed toward the tailstock.
More material was cut off. When I measured them, it was only off
a few thou.

I don't know how it could be larger at the headstock, and then
come out straight on the 2nd cut. And what good are indexable inserts
if they don't cut straight?

Once again I'm so confused.
Wayne

On Wed, 28 Jul 2004 01:08:01 GMT, Anthony
wrote:


The extra material removal is from the flex in the tool/toolpost/set-up
when taking the inital cut. The information about taper when doing this
is irrelevant and unuseful due to that fact. You would need to be able
to measure the differences in the flex of the tool tip in both directions
in order to obtain any taper information.

Just a big of trivia..that last final pass is usually called the
"spring pass"

Gunner

"In my humble opinion, the petty carping levied against Bush by
the Democrats proves again, it is better to have your eye plucked
out by an eagle than to be nibbled to death by ducks." - Norman
Liebmann

Anthony wrote:

Oops...failed to notice you said *larger* at the headstock. That would
not be material flex.

Wrong. Work flexes away from the tool resulting in smaller diameter as
you go away from the headstock.

BeamOut ''(30E6,I_tube 1 .5)Beam 0 (5 35)
pos'n pt. couples shear bend. slope def'l
force stress stress

0 -35 -175 -59.42 1783 0 0
5 35 0 0 0 0.0002971 0.0009903

It only takes 35 pounds force to deflect a cantilevered 1" steel bar a
thou at 5 inches from the support.

Ted

Wayne wrote:

I have 3/8" shank holders with HSS inserts. I have 1/32 and 1/64
radius inserts.

You might want to consider positive/negative rake carbide inserts. See
my article at
http://www.metalwebnews.com/howto/toolholder/toolholder.html

Also, see my reply elswhere in this thread re deflection.

Ted

On Wed, 28 Jul 2004 17:42:52 GMT, Ted Edwards wrote:

||Anthony wrote:
||
|| Oops...failed to notice you said *larger* at the headstock. That would
|| not be material flex.
||
||Wrong. Work flexes away from the tool resulting in smaller diameter as
||you go away from the headstock.

Now, as far as being a machinist goes, I'm not. But this doesn't make sense to
me. If the work is solidly supported close to the chuck, then the flex is
minimal and the bit cuts to the intended depth. As the cutter moves away from
the chuck, deflection grows, the cutter cuts less deeply as the work moves away
from the tool. That will result in a larger diameter as it moves away from the
chuck, because less material is being removed.
Am I missing something here?
Texas Parts Guy

Rex B wrote:

from the tool. That will result in a larger diameter as it moves away from the
chuck, because less material is being removed.
Am I missing something here?

Nope. I was missing my second cup of coffee. The deflection
calculation is correct - I just blew the direction.

Ted

Ted Edwards wrote in
:

Anthony wrote:

Oops...failed to notice you said *larger* at the headstock. That
would
not be material flex.

Wrong. Work flexes away from the tool resulting in smaller diameter as
you go away from the headstock.

BeamOut ''(30E6,I_tube 1 .5)Beam 0 (5 35)
pos'n pt. couples shear bend. slope def'l
force stress stress

0 -35 -175 -59.42 1783 0 0
5 35 0 0 0 0.0002971 0.0009903

It only takes 35 pounds force to deflect a cantilevered 1" steel bar a
thou at 5 inches from the support.

Ted




Umm...... Ted, if the part is flexing AWAY from the tool as you go away
from the headstock, it is cutting LESS material off, and therefore the
diameter of the turned part will be larger, not smaller.




--
Anthony

You can't 'idiot proof' anything....every time you try, they just make
better idiots.

Remove sp to reply via email

Wayne wrote:

I'm so very confused.

I have 3/8" shank holders with HSS inserts. I have 1/32 and 1/64
radius inserts.

I was testing the cutting straightness of the lathe with a
1" steel bar over 5". The results I got were 2 thou bigger
at the headstock than at the free end.



This is why I never do these tests with a cutting tool. Instead, I
use a piece of hardened and ground shafting of known straightness
and roundness. I put this in the chuck and wiggle a little until it
turns as true as I can get it. I then read at two distances from the
chuck with a dial test indicator. Lately I've been using a surplus
Federal Maxxum electronic indicator that reads to .00005" although
that last 5 is pretty iffy. I turn the chuck manually and average the
readings. This thing only puts a couple of grams of pressure on
the bar, so the deflection is pretty small (but it is still there, of
course).
You bypass all the errors caused by heating, built-up edge, cutting
tool wear, surface roughness changes, etc. that you would also have
when actually cutting a bar. Also, you can use the carriage handle
to rock the carriage left and right, and see if the reading changes.
This can detect wear in the bedways or carriage.

Before doing this, I level the bed with the best level available, for
front-back level, at both the headstock and tailstock end. You want to
eliminate as much twist in the bed as you can, as this can foul up
all other measurements.

Jon

"Jon Elson" wrote in message
...


Wayne wrote:

I'm so very confused.

I have 3/8" shank holders with HSS inserts. I have 1/32 and 1/64
radius inserts.

I was testing the cutting straightness of the lathe with a
1" steel bar over 5". The results I got were 2 thou bigger
at the headstock than at the free end.



This is why I never do these tests with a cutting tool. Instead, I
use a piece of hardened and ground shafting of known straightness
and roundness. I put this in the chuck and wiggle a little until it
turns as true as I can get it. I then read at two distances from the
chuck with a dial test indicator. Lately I've been using a surplus
Federal Maxxum electronic indicator that reads to .00005" although
that last 5 is pretty iffy. I turn the chuck manually and average the
readings. This thing only puts a couple of grams of pressure on
the bar, so the deflection is pretty small (but it is still there, of
course).
You bypass all the errors caused by heating, built-up edge, cutting
tool wear, surface roughness changes, etc. that you would also have
when actually cutting a bar. Also, you can use the carriage handle
to rock the carriage left and right, and see if the reading changes.
This can detect wear in the bedways or carriage.

Before doing this, I level the bed with the best level available, for
front-back level, at both the headstock and tailstock end. You want to
eliminate as much twist in the bed as you can, as this can foul up
all other measurements.

Jon

i don't know, I usually do them with a cutting tool. I am only interested
in how the lathe cuts, after all...If I am trying for a thou or less over
some distance, I sneak up on it, learning how my bit is cutting at the depth
and feed I will use for the finish cuts, and if I have to I tweak the lathe
bed to get what I need. Obviously this will differ between OD cutting and
boring. Thankfully I very rarely need to bore or cut over more than an inch
to that level of accuracy. My lathe can do a couple of inches to less than
a thou no problem.

Brian

Brian,........How do you "tweak the lathe bed" in this particular case ?
regards.....Jan ter Doest.


"Brian" wrote in message
...
"Jon Elson" wrote in message
...


Wayne wrote:

I'm so very confused.

I have 3/8" shank holders with HSS inserts. I have 1/32 and 1/64
radius inserts.

I was testing the cutting straightness of the lathe with a
1" steel bar over 5". The results I got were 2 thou bigger
at the headstock than at the free end.



This is why I never do these tests with a cutting tool. Instead, I
use a piece of hardened and ground shafting of known straightness
and roundness. I put this in the chuck and wiggle a little until it
turns as true as I can get it. I then read at two distances from the
chuck with a dial test indicator. Lately I've been using a surplus
Federal Maxxum electronic indicator that reads to .00005" although
that last 5 is pretty iffy. I turn the chuck manually and average the
readings. This thing only puts a couple of grams of pressure on
the bar, so the deflection is pretty small (but it is still there, of
course).
You bypass all the errors caused by heating, built-up edge, cutting
tool wear, surface roughness changes, etc. that you would also have
when actually cutting a bar. Also, you can use the carriage handle
to rock the carriage left and right, and see if the reading changes.
This can detect wear in the bedways or carriage.

Before doing this, I level the bed with the best level available, for
front-back level, at both the headstock and tailstock end. You want to
eliminate as much twist in the bed as you can, as this can foul up
all other measurements.

Jon

i don't know, I usually do them with a cutting tool. I am only interested
in how the lathe cuts, after all...If I am trying for a thou or less over
some distance, I sneak up on it, learning how my bit is cutting at the
depth
and feed I will use for the finish cuts, and if I have to I tweak the
lathe
bed to get what I need. Obviously this will differ between OD cutting and
boring. Thankfully I very rarely need to bore or cut over more than an
inch
to that level of accuracy. My lathe can do a couple of inches to less
than
a thou no problem.

Brian

Tweak the lathe bed - for example if I am cutting big on the OD on the chuck
end of the work-piece, I raise the rear tailstock leg a bit. Intoduce a
tiny bit of twist to the bed, so that the bit is a tiny fraction farther
away from the work at the tail-stock end of the piece. I use aluminium
wedges to level the lathe, I just tap one. I've only done this once or
twice at most. To me a lathe isn't really designed to cut to a half-thou
over five inches, or bore to similar accuracy, that's in the realm of
grinding and honing.

Brian


"j. ter doest" wrote in message
l...
Brian,........How do you "tweak the lathe bed" in this particular case ?
regards.....Jan ter Doest.


"Brian" wrote in message
...
"Jon Elson" wrote in message
...


Wayne wrote:

I'm so very confused.

I have 3/8" shank holders with HSS inserts. I have 1/32 and 1/64
radius inserts.

I was testing the cutting straightness of the lathe with a
1" steel bar over 5". The results I got were 2 thou bigger
at the headstock than at the free end.



This is why I never do these tests with a cutting tool. Instead, I
use a piece of hardened and ground shafting of known straightness
and roundness. I put this in the chuck and wiggle a little until it
turns as true as I can get it. I then read at two distances from the
chuck with a dial test indicator. Lately I've been using a surplus
Federal Maxxum electronic indicator that reads to .00005" although
that last 5 is pretty iffy. I turn the chuck manually and average the
readings. This thing only puts a couple of grams of pressure on
the bar, so the deflection is pretty small (but it is still there, of
course).
You bypass all the errors caused by heating, built-up edge, cutting
tool wear, surface roughness changes, etc. that you would also have
when actually cutting a bar. Also, you can use the carriage handle
to rock the carriage left and right, and see if the reading changes.
This can detect wear in the bedways or carriage.

Before doing this, I level the bed with the best level available, for
front-back level, at both the headstock and tailstock end. You want
to
eliminate as much twist in the bed as you can, as this can foul up
all other measurements.

Jon

i don't know, I usually do them with a cutting tool. I am only
interested
in how the lathe cuts, after all...If I am trying for a thou or less
over
some distance, I sneak up on it, learning how my bit is cutting at the
depth
and feed I will use for the finish cuts, and if I have to I tweak the
lathe
bed to get what I need. Obviously this will differ between OD cutting
and
boring. Thankfully I very rarely need to bore or cut over more than an
inch
to that level of accuracy. My lathe can do a couple of inches to less
than
a thou no problem.

Brian

Anthony wrote:

Umm...... Ted, if the part is flexing AWAY from the tool as you go away
from the headstock, it is cutting LESS material off, and therefore the
diameter of the turned part will be larger, not smaller.

Right. I posted a correction.

Ted

Ahh, the infamous test bar.
I've searched, but can find nowhere to get one.
I'm getting the impression you have to make your own.
But since I can't make straight cuts yet, I wouldn't
know how to do it.

This is why I never do these tests with a cutting tool. Instead, I
use a piece of hardened and ground shafting of known straightness
and roundness. I put this in the chuck and wiggle a little until it
turns as true as I can get it. I then read at two distances from the
chuck with a dial test indicator. Lately I've been using a surplus
Federal Maxxum electronic indicator that reads to .00005" although
that last 5 is pretty iffy. I turn the chuck manually and average the
readings. This thing only puts a couple of grams of pressure on
the bar, so the deflection is pretty small (but it is still there, of
course).
You bypass all the errors caused by heating, built-up edge, cutting
tool wear, surface roughness changes, etc. that you would also have
when actually cutting a bar. Also, you can use the carriage handle
to rock the carriage left and right, and see if the reading changes.
This can detect wear in the bedways or carriage.

Before doing this, I level the bed with the best level available, for
front-back level, at both the headstock and tailstock end. You want to
eliminate as much twist in the bed as you can, as this can foul up
all other measurements.

Jon

Gunner wrote in message . ..
On Wed, 28 Jul 2004 01:08:01 GMT, Anthony
wrote:


The extra material removal is from the flex in the tool/toolpost/set-up
when taking the inital cut. The information about taper when doing this
is irrelevant and unuseful due to that fact. You would need to be able
to measure the differences in the flex of the tool tip in both directions
in order to obtain any taper information.

Just a big of trivia..that last final pass is usually called the
"spring pass"

Gunner

Now I know the name for it. I've seen that happen on my mill/drill.
I kinda thought a much sturdier lathe taking only .002 diameter wouldn't
do that.

In article 51701a28.0408011333.7e68d461
@posting.google.com, says...
Ahh, the infamous test bar.
I've searched, but can find nowhere to get one.
I'm getting the impression you have to make your own.
But since I can't make straight cuts yet, I wouldn't
know how to do it.

You can buy a piece of 1" X 12" Thomson shaft from McMaster
for about $16, for example. It's guaranteed to be straight
within .001"/ft, but I think it's likely to be better in
this large a diameter.

Ned Simmons

On 1 Aug 2004 14:37:13 -0700, (Wayne) wrote:

Gunner wrote in message . ..
On Wed, 28 Jul 2004 01:08:01 GMT, Anthony
wrote:


The extra material removal is from the flex in the tool/toolpost/set-up
when taking the inital cut. The information about taper when doing this
is irrelevant and unuseful due to that fact. You would need to be able
to measure the differences in the flex of the tool tip in both directions
in order to obtain any taper information.

Just a big of trivia..that last final pass is usually called the
"spring pass"

Gunner

Now I know the name for it. I've seen that happen on my mill/drill.
I kinda thought a much sturdier lathe taking only .002 diameter wouldn't
do that.

Even my stiff as an anvil Hardinge HLV-H will, depending on the
material and diameter, take a spring pass.

A good way to check, is to make your cut normally, right to left, then
reverse your feed and go back to the right without changing the
cutting depth. You will be surprised on how much and how close to the
headstock it will shave off on the return.

Gunner

"At the core of liberalism is the spoiled child -
miserable, as all spoiled children are, unsatisfied,
demanding, ill-disciplined, despotic and useless.
Liberalism is a philosphy of sniveling brats." -- P.J. O'Rourke

As your experience "grows", you'll find .002 inch on the dia.
is more in the realm of grinding.

With lathe work, more consistent results are obtained with
a heavier cut. Suggest you use .010 on dia. as a finishing pass.

Why? You are putting more of a force on the whole setup,
which takes up the slack in bearings, etc.

It also puts the actual "point" of the cutting tool below the surface
of the material as opposed to just skating across the existing surface.

Spring passes are fine when necessary, but in most cases should try
to avoid as a common operation.


"Wayne" wrote in message
om...
Gunner wrote in message
. ..
On Wed, 28 Jul 2004 01:08:01 GMT, Anthony
wrote:


The extra material removal is from the flex in the tool/toolpost/set-up
when taking the inital cut. The information about taper when doing
this
is irrelevant and unuseful due to that fact. You would need to be able
to measure the differences in the flex of the tool tip in both
directions
in order to obtain any taper information.

Just a big of trivia..that last final pass is usually called the
"spring pass"

Gunner

Now I know the name for it. I've seen that happen on my mill/drill.
I kinda thought a much sturdier lathe taking only .002 diameter wouldn't
do that.