TIR drift of Set-Tru Chuck

[PeterT]

I have a Bison Set-Tru 5C collet chuck purchased new, no damage or crashes or abuse of any kind. The Set-Tru system has a mounting plate whereby you can dial in concentricity using radial fine pitch screws integrated in the adapter plate, kind of like a 4 jaw independent chuck principle. I haven't used it in a while but I know it was within a tenth when I set it. I have a mark indicating the (D1-4) pins so they engage to the same spots each time. When I mounted a part I knew something was fishy, it was off by 0.0035". I removed the collet & measured the 5C seat itself, exact same result there. In other words runout was not due to the part or the collet. All the screws felt reasonably tight including the front face screws. Not massive gronk but reasonably tight. I reset concentricity using DTI, again within a tenth. About a week later I checked again, its off by 0.001". The lathe spindle is accurate so no issues there. Same test method & apparatus & shop temp etc.

This got me suspicious. I checked my Set-Tru 6J chuck, same deal. Out by 0.002". But here I'll allow dumb points. My front screws were not super tight. Maybe I just need more Wheaties. So I dialed that one in again & will check back to see if the shop goblins mess around with that one too. Is it common for these to drift over time for some reason? In all honesty I have become complacent & just assumed they would stay put.

 

[RobinHood]

That is a very interesting observation.

On the 5C chuck, was it removed and remounted to the spindle during the timeframe where it “drifted” out of concentricity? We know the 6J was removed.

If it was taken off and put back on the spindle, my bet is on the backing plate taper engaging too heavily before it is drawn completely onto the flange - thus distorting the plate (or, the taper may be too loose). No matter how carefully you mount the chuck, it is near impossible to put the exact same force on the same cams in the same sequence from one time to the next.

Might have to look at your backing plates.

 

[PeterT]

In both cases, the entire chuck (meaning integral D1-4 Set-Tru backplate + chuck) was removed & stored on a shelf.
I also measured the the chuck face nose with DTI, it was 0.0000" both before & after readjustment, so the runout is entirely in the faceplate plane.
I also checked my spindle nose, its still my typical 0.0002" depending where on the nose taper.
Funny thing, my regular Bison 3J (non SetTru) reads the exact same runout every mount/dismount.

I'm going to have a closer look at the adjuster screws. They adjust opposite direction to how one would thing a set screw acts, but same general principle as 4J chuck - back off slightly, tighten the opposite, balance them to same torque round & round. The front screws have to be snug just to allow displacement & then tightened when runout done. I'm just doing that with an allen key for leverage but maybe it needs more. I'll have to pull the instructions & look at torque. My regular Bison 4J chick jaws were gronked but good.

After watching the Renzetti movie where he shows deflection by torqueing screws. Maybe this is real life & the back plate is stress relieving slightly with no spindle nose support? I just always 'assumed' it should stay put. It just seemed excessive to my expectations.

 

[RobinHood]

After watching the Renzetti movie where he shows deflection by torqueing screws. Maybe this is real life & the back plate is stress relieving slightly with no spindle nose support? I just always 'assumed' it should stay put. It just seemed excessive to my expectations.

That could be a good explanation. Something is going on with the backing plate.

 

[Dabbler]

I understand that you use the adjusting screws to get to zero on lightly tightened securing bolts. then you tighten, check, tighten, check again, etc until the tightening bolts are at the manufacturer's specified torque. you then back off the adjusting screws until they are all at the same, light torque, just enough to lock them. - oh, and then check again, of course!

perhaps a pair of adjusting screws were very tight at storage and it distorted the chuck over time?

 

[PeterT]

My adjusting screws are definitely decently tight. I don't have a torque wrench. I may have to read the instructions :/

 

[RobinHood]

I second what John is pointing out: once TIR is set, the adjusting screws should no longer be tight - if I understand the principle correctly. The setting should be held by the “locking” bolts.

 

[PeterT]

hmm, nothing after step-12 tightening the front bolts

instructions bison set tru chuck - Google Search

www.google.com

weekend warrior, same deal (assuming unlike me, he read instructions haha)

is he hinting at the same issue?
The guys are right on with the hold down screws, just loosen them a bit, the adjusting screws develop more than enough force to nudge the chuck around on the back plate. It is an art to set one of these and takes a bit of practice as to tightening one side or slightly loosening the other. A word of advise...... don't over tighten any one side as it may start to deform the body of the chuck. Many times I find that I have to gradually tighten down the mounting screws as I play with the adjusting screws to obtain the desired setting.

https://bbs.homeshopmachinist.net/forum/general/66467-bison-set-tru-chuck#post1374290

and finally from Bison. They just say 'equally tightened'

https://bison-chuck.com/wp-content/uploads/2021/06/Manual-3200_3500-1001-01.pdf

 

[Susquatch]

Is it common for these to drift over time for some reason? In all honesty I have become complacent & just assumed they would stay put.

I've had a Bison C5 Collet Chuck on a D1-5 spindle for at least a decade. It was perfect when I got it and still is. I've never touched it. I guess I'm complacent too.

If I ever did, I firmly believe in torque wrenches. I think I have over a dozen - 3 adjustable Wiha drivers, 6 clickers, 4 electronic, 2 rod pointers. It might sound excessive to some but they each have their place and they are all used. I would not be without them.

If you have none, sounds to me like you need at least 3. Two clickers (1/4 and 3/8) and a Wiha driver (10 - 50 in lbs) for small screws. This sounds like a great excuse to get the 3/8 clicker. Don't forget to set it to zero for storage or it will relax just like your chuck did......

 

[PeterT]

Yes, I'm overdue for torque wrenches.
Chuck body distortion acting across the spindle cavity would not be a good thing, although I don't rule it out that affect because the adjusting screws act perpendicularly. They didn't feel any different mounting to spindle. Maybe more torque on one screw vs. the opposing allows it to relieve itself over time. I'll try backing them off & setting slightly looser until the new tool shows up

 

[Susquatch]

Yes, I'm overdue for torque wrenches.
Chuck body distortion acting across the spindle cavity would not be a good thing, although I don't rule it out that affect because the adjusting screws act perpendicularly. They didn't feel any different mounting to spindle. Maybe more torque on one screw vs. the opposing allows it to relieve itself over time. I'll try backing them off & setting slightly looser until the new tool shows up

You can spend a small fortune on torque wrenches, but I don't recommend that (all except for the Wiha's which I do recommend). Basic Neiko or Tekton Clickers on Amazon are just fine. They are not perfect but still INFINITELY better than your unaided gutntite hands. They also come in VERY handy hard plastic storage boxes.

 

[trevj]

I would only expect repeatability at a single size of work-piece, and an accurately sized work-piece, at that!

In line with a four jaw chuck, every part has to be considered different, and dialed in individually.

About all I would expect from my Set-Tru PB chuck, is that it 'should' be closer than the un-named three jaw I have that I don't expect much from, other than that all three jaws will contact the work.

 

[PeterT]

I see what you are saying but my question is more about the chuck body deviating from the spindle backplate over time, subsequent to having been originally set dead nuts. Using my Set-Tru 5c collet chuck as an example, I'm indicating off the chuck's collet seat itself, no collet or part is involved for this test. ie. if after being indicated to say a tenth, then it sits on a shelf for 3 months & is subsequently 'different', my question is why. I agree, jaws & scroll probably introduce a little bit more variation but not really what I'm asking.

 

[Chipper5783]

The points above are all good. One area to check is the fit of the back plate onto the spindle nose.

The back plate is supposed to mate to the short taper AND the flat vertical face of the spindle nose. That is not easy. If the taper lands first, by even a small amount (say 0.0005”), then it will prop the chuck slightly. One clue is if the chuck is stuck when you remove it (i.e. Needs a bump to get free).

When I was truing my 3J, I took a small amount of material out of the taper bore of the back plate (stoning and abrasive cloth - no cutting tools, it is a taper fit - only half a tenth will move the taper quite a ways). This approach is written up on various posts (see hobby machinist board).

My new 4J Bison does not sit completely down. If I need/want the jaw faces or chuck face running true then I dial the face while tightening the cams (heck it is a 4J - depending on the part being held I’ll dial radial and face, or I’m going to machine the surfaces and that makes them true.

Suggestion for you - install and remove your chuck a few times (with a test pin in the collet). Does it stay the same each time? Does it matter the order you snug the cams? Does it matter if you tighten the cams hard on the first round, or do several rounds?

Let us know how you make out.

 

[Dabbler]

One clue is if the chuck is stuck when you remove it (i.e. Needs a bump to get free).

@Chipper5783 all D1-x chucks must lock onto the taper and REQUIRES the 'bump' to get the chuck off. If you do not lock tightly onto the taper, your fitment is completely wrong. It is the taper that transmits the power to the chuck - not the pins. the pins are only there to draw the chuck to the axial flat, and keep it there.

 

[Susquatch]

@Chipper5783 all D1-x chucks must lock onto the taper and REQUIRES the 'bump' to get the chuck off. If you do not lock tightly onto the taper, your fitment is completely wrong. It is the taper that transmits the power to the chuck - not the pins. the pins are only there to draw the chuck to the axial flat, and keep it there.

@Chipper5783 , @Dabbler beat me to it. The mallet Rapp is a requirement. If it doesn't take a Rapp, it isn't fitting correctly.

I have found that a hard plastic mallet works wwwaaayyy better than soft plastic, brass, or rubber. Something about the vibration just works better.

But your point about it potentially moving on the mount is still a good one. I agree with doing the tests you suggest. If everything is working correctly, it shouldn't move between successive mounts and dismounts.

 

[PeterT]

Yes the SetTru chucks seems to fit the spindle nose just fine. Bison is decent quality, at least for my pay check. But it always pays to never assume. Yes, I am aware of the D1-x type 2 simultaneous surface contact theme. My other non-SetTru chuck repeats the same TIR & my spindle is decently accurate ~0.0002" so all those numbers go round.

This is why I'm wondering out loud if the issue is the chuck being OFF the machine for some period & its somehow stress relieving itself as a function of the radial set screws. Maybe if the chuck stayed resident on the machine it wouldn't drift (as much) over time? The set screws are pretty meaty, like M8 if I recall & I'm staying within reasonable limits of my hand tightening on a typical Allen key, no added gronk leverage. But maybe this is what I'm doin wrong - reasonably tight is desired, excessive tight is bad?

This video isn't quite the same issue but clearly shows the extent that things can move even decently thick steel & how he remediated. Don't get me wrong, I'm not opposed to adjustment tweaking if required & maybe that is the unwritten little secret. I always assumed once runout was set, it should stay very close to that for an extended period. I'll monitor.

 

[Tom Kitta]

@Chipper5783 all D1-x chucks must lock onto the taper and REQUIRES the 'bump' to get the chuck off. If you do not lock tightly onto the taper, your fitment is completely wrong. It is the taper that transmits the power to the chuck - not the pins. the pins are only there to draw the chuck to the axial flat, and keep it there.

Yeah some of my home made back plates stick like if they were glued on. But I am not super consistent. Some were not as good as others.

 

[Tom Kitta]

Chuck removal and placement again should not change the TIR too much - at least in even the worst back plate I made (and I bloody made a video of that one!) I am not seeing much movement - in fact my test is to dismount, turn and place few times checking TIR.

My best home made plate is well under 0.001 TIR. On my cheap Chinese 5C the issue is not home made back plate but the grinding of 5C collet taper which is touch off.

For set true I would definitely check if anything got loose with the st true part.

My expectation for set true is it holds accuracy of best collet or better - i.e. the collet is the limiting factor not the chuck. I.e. once set to specific bar stock it should repeat within say 0.00025 TIR or so. Regular BISON should be somewhere around say 0.0005 or at least under 0.001. I think unground Chinese 150 CAD 5C with cheap Chinese 5C collets is around 0.0015 or so. I expect it to drop to under 0.001 with grounding and better collet.

I measure run-out close to the chuck - if you measure far you may get more of it. This is important for some chuck types - cheap Chinese 4 jaw when not ground have tendency to have set run-out of under 0.001 1" from the chuck while it is a whopping 0.005 say 3" out and 0.01 say 6" out. You can see part wobble.

 

[PeterT]

Like I mentioned, for the Bison 5C ST I'm measuring the collet SEAT, no collet or reference pin involved.
The ST 6-jaw on a carbide reference bar was a bad test because I noticed my front bolts (the axial chuck retention bolts) were insufficiently tight right from the get-go, so all bets off there. Not sure why, maybe a brain fart on my part. I've reset it too & will confirm.