Normal eopen-closed mocement for 22J collets

I'm having a very hard time suddenly holding onto 2" diameter 7075 round bar for machining, and a first for us, the bar is walking out instead of heavy cuts pushing it into the collet.
Among the things I'm looking into is the possibility that the collet closer system is not working right.
If I unscrew a collet far enough that it can't squeeze in the collet nose and actuate the closer the collet's total movement is only 1/8", and I'm wondering if that's the normal movement?
The spindle liner is .025" larger than the bar, and I was told that's normal because of bent, damaged or oversized bars. My mention of that goes to the bar "Walking" out of the collet as I've seen carbide endmill shanks walk out with chatter before.
Another factor may be the aggressive high positive inserts I've been using, the theory being that they are pulling the bar as they cut. We've done this job twice before without seeing this problem.

I hardly have any hair left! If anyone could check the pull distance with a 22J collet I'd like to hear how far their machine's spindle nose, collet adapter and collet nose allows an unloaded collet to move in and out. Heck I even grooved the gripping area of the collet trying to stop this problem.
I could mount a chuck, that would hold, but as I said we've never had the problem before.

In my haste last night I forgot to mention the lathe is a Takisawa TW-30 with a hydraulic actuator, ATS collet chuck, Hardinge collets. The drawtube pressure gage is as high as it gets.

The forces on the tool have actually pulled the toolholder away slightly from the seat in the turret, so I'm pretty sure the tool is trying to pull into the work. I suspect that the chip is strong enough to provide ample pull, and with the bar moving ever so slightly in the collet the bar is walking out of the collet.
This AM I may check the drawtube again, without the collet coupler it moves 1 1/2" in and out, but of course movement is limited by the coupler in the collet nose.
I'm still wondering if the hydraulic actuator could be getting less of the pressure shown on the gage somehow. or, as I stated above, the drawtube is not screwed all the way into the actuator, which might limit the last 1/16" of collet pull.

sounds like your coupler isnt screwed in all the way ,or your collet isnt screwed in all the way
dont run a collet on high pressure it ruins them.
if your drawbar tube is backed out from the actuator its needs to be screwed back in.

just for the heck of it is your draw bar sloppy loose up and down side to side play? it really shouldnt have anyplay at all.

Years ago I made sure the tube was screwed in all the way and made a series of punch marks so it would be obvious it was in all the way. Still, even so the bar was pulling out. Oh, and when I made the liner I made it with aluminum collars with an o-ring groove in each of the three collars, so it's pretty snug inside the spindle.
I did some careful machining of the tube/collet coupler and got 3/16" movement, an extra 1/16".
Then I rewrote the roughing section so it was only taking .050" DOC and .010" per revolution.
That seemed to fix it, at the cost of cycle time. At this point I think I'll try increasing the feed until the cycle time comes down.

The weird part was that it's never pulled a bar out before, always when overdoing it it pushed the bars in and I ended up with short parts.
Thanks a lot for the suggestions Delw.
parts

any chance you changed material suppliers?

I had a program/part that I had run half a dozen times and all of a sudden with a new batch of material the blanks were being pulled out of the jaws.
Turns out the the supplier had not given me 6061T6 but a bar of some untraceable chewing gum that was silver. Since I hadn't requested certs, he figured I wouldn't notice.

My only 22J chuck is a Sjogren, so I can't help with pull distance on your closer. However, a Hardinge lever-operated 5C closer with a properly sized collet moves the collet very, very little when throwing the lever over center.

Pursue mjk's suggestion. Your work bar should have essentially no play in the collet before closing the collet. You may have received undersized stock.

lets try this again

Is your part Pushing in or pulling out. you said both.
you also said you cut some grooves in the collet .

take your part out of the collet and make a slug .020-.030 smaller than your bar dia.
put that in the collet and see if its clamping on the part. if it doesnt theres your problem.

now the obvious. your sure you have the correct size collet? not one bigger????? or did you cet and ER(emergency collet) collet over sized by accident.
a bigger collet will only hit in the center of the bar on 3 sides, it wont grip with most the collet.

I usually screw the collet in, to it touchs light the taper when opened, then back it off to the 1st screw lock down hole. can you get it that far?

is your spindle liner keeping your collet from closing all the way? if you make them too long sometimes they will interfere with the collet closing all the way.

can you post a pic of your set up?

Turns out the the supplier had not given me 6061T6 but a bar of some untraceable chewing gum that was silver.

Click to expand...

Recently we made up some expensive 416 piston rods, sent out to heat treat to 40-44 Rc, then to be ground down .010" and Hard chromed, and finish ground. Thus the high price.

The heat treater calls says it won't harden past 30 Rc and it's not magnetic. The invoice said "416 stainless". I offered the customer the parts at half price, a loss for us. Must have certifications from now on.

Is your part Pushing in or pulling out. you said both.

Click to expand...

In the past when really maxing roughing cuts my experience has always been that the barstock pushes back into the collet, but this 2" diameter 7075 pulled out under the cutting pressures. As I said, that was a first for us.
The collets were Hardinge 2" and an emergency 2.008" that I found on the rack from an older job. The first "pull out" crashes were with "2 inch" 7075 that was actual size 2.007", and I surmised that the larger bar diameter made the gripping surfaces tapered enough to allow the barstock to wobble side to side and walk out (The bar was making a racket too). So I tried the E collet that is 2.008" and it pulled out with that too!
In desperation I grooved the gripping surfaces .010" deep every .060" the whole grip length, no joy, still pulled out!
At that point I'd made a few good parts but scrapped several too, and the next bar in the loader (4' bars) measured 2.002" so I went back to the Hardinge 2", again the bar worked out of the collet during the cut and caused a crash, so I grooved that too using a slow RPM and a carbide internal threading insert in a BB, ,010" deep and .050" per groove in this.
BTW, these "crashes" only broke inserts, but at $25 each I'm going backwards.

Again, with the 2" material and 2" collet with gripping grooves the bar walked out 2 out of three times.

"Stop" I said, "must try something else".

The depth of cut was .200 per pass (.400 diameter reduction), two passes (Roughing) at .008 per rev and then at .006 and finally down to .004 and still pulling out.

Our earlier experience with this high shear insert was .200 DOC (.400 diameter reduction) pushing at .035" per revolution in Tellurium copper, a huge production breakthrough. But Tellurium copper makes a brittle chip that looks like gravel bits and sounds like a hailstorm inside the enclosure, but this 7075 has a very strong chip and I wasn't feeding anyways near as hard, and strange as it may sound the explanation I came to was that the insert was trying to pull the bar because of this, for example, the tool holder was actually pulled forward out of the turret seat about .100" (Into the cut).
So the answer I eventually came to was to cut the DOC by 1/4th and push the tool 3 times as hard, the smaller DOC means the chips are breaking up now and it's working.
We've used high shear inserts for 6061 T-6 before, pushing hard, and never saw this phenomenon before, and had someone else told of it I'd probably have doubted it, except it was happening to me.
The bar loader accepts 4' bars, and we still have many 12' bars to cut up and load. I'd like to explore where the change from working to not working is because that would cut the time way down, but probably won't because at .020 feed per revolution and six or so passes the roughing is done pretty quick.
We'll have to measure each bars diameter though and do the bigger diameters last in the larger E collet, though the collet, the collet pressure, and the collet setup is likely not part of the problem if my current theory is correct, that .200" wide chip had too much strength to break up is all. And of course a negative insert would have worked at breaking up the chip but I'd never have pushed one as hard.
parts

partsproduction said:

In the past when really maxing roughing cuts my experience has always been that the barstock pushes back into the collet, but this 2" diameter 7075 pulled out under the cutting pressures. As I said, that was a first for us.
The collets were Hardinge 2" and an emergency 2.008" that I found on the rack from an older job. The first "pull out" crashes were with "2 inch" 7075 that was actual size 2.007", and I surmised that the larger bar diameter made the gripping surfaces tapered enough to allow the barstock to wobble side to side and walk out (The bar was making a racket too). So I tried the E collet that is 2.008" and it pulled out with that too!
In desperation I grooved the gripping surfaces .010" deep every .060" the whole grip length, no joy, still pulled out!
At that point I'd made a few good parts but scrapped several too, and the next bar in the loader (4' bars) measured 2.002" so I went back to the Hardinge 2", again the bar worked out of the collet during the cut and caused a crash, so I grooved that too using a slow RPM and a carbide internal threading insert in a BB, ,010" deep and .050" per groove in this.
BTW, these "crashes" only broke inserts, but at $25 each I'm going backwards.

Again, with the 2" material and 2" collet with gripping grooves the bar walked out 2 out of three times.

"Stop" I said, "must try something else".

The depth of cut was .200 per pass (.400 diameter reduction), two passes (Roughing) at .008 per rev and then at .006 and finally down to .004 and still pulling out.

Our earlier experience with this high shear insert was .200 DOC (.400 diameter reduction) pushing at .035" per revolution in Tellurium copper, a huge production breakthrough. But Tellurium copper makes a brittle chip that looks like gravel bits and sounds like a hailstorm inside the enclosure, but this 7075 has a very strong chip and I wasn't feeding anyways near as hard, and strange as it may sound the explanation I came to was that the insert was trying to pull the bar because of this, for example, the tool holder was actually pulled forward out of the turret seat about .100" (Into the cut).
So the answer I eventually came to was to cut the DOC by 1/4th and push the tool 3 times as hard, the smaller DOC means the chips are breaking up now and it's working.
We've used high shear inserts for 6061 T-6 before, pushing hard, and never saw this phenomenon before, and had someone else told of it I'd probably have doubted it, except it was happening to me.
The bar loader accepts 4' bars, and we still have many 12' bars to cut up and load. I'd like to explore where the change from working to not working is because that would cut the time way down, but probably won't because at .020 feed per revolution and six or so passes the roughing is done pretty quick.
We'll have to measure each bars diameter though and do the bigger diameters last in the larger E collet, though the collet, the collet pressure, and the collet setup is likely not part of the problem if my current theory is correct, that .200" wide chip had too much strength to break up is all. And of course a negative insert would have worked at breaking up the chip but I'd never have pushed one as hard.
parts

Click to expand...

Pulling out is wild, only seen it on running high RPMS with hydraulic barfeeds (which alwasy has pressure on the bar)and lots of playing in the draw tube. or with a cutoff tool insert that broke.
We have turned up to 2.625 with 12 foot bars in collets and not had a problem in the past. max rpm was I think around 3500.

7075 is generally cold finish and nice and uniform through out the whole bar, at least all the ones I have used.
I use high shear ie sandvik alum inserts generally .100-.150 per side makes a nice chip at around .010-.015 feed on smaller than 1.5" stock I keep it at .100 per side my rad on the insert is .015 find it chips better.
.200 and at .035 might be a tad much for a collet. put it should push the bar in not pull it out.
I personally like 7075 better than 6061 in turning I use the same feeds and speeds.

oh yeah you said you had many crashes due to part pulling out. check your centerline. if you turret moved your pushing more than cutting. that could be a problem if your not on center.