ER32 M3 Collet chuck excessive runout.

[Stockyj]

To skim the chuck taper my first choice would be a very sharp HSS-Co boring tool with generous relief.
If runout is .04mm (=.0015'') then theoretically it only takes a .00075'' cut to true it.

it might be a good idea, while in the same set up, to pick up the external threads and skim those too. For this preferably I would turn the lathe spindle manually with a crank

Before removing the ER chuck from the lathe put an index mark on it so you can put it back always in the same position in the spindle taper.

You cannot cut the Hardened steel with a HSS tool that is definite.

 

[Stockyj]

You have a lot of interfaces in your set up - the spindle Morse taper, the chuck Morse taper, the chuck's internal taper, the collet external taper, the nut and the test rod. Add to that the need for adequate torque on the nut before you can check run out on the rod. This is collectively called stacking tolerances but each of these can contribute some/most/all of the run out and you need to nail down exactly which it is or what each one contributes in order to figure out what to do about it. I would not grind/bore anything until you do this. As Tozguy said, make a diagnosis first.

I can tell you that a cheap collet can add 0.0007" of run out all by itself. A cheap Chinese nut can double that. This all matters if you are doing second operations work and need precision. For first operations work, it matters very little.

All was tight all was clean all was repeatable, I have asked for a return thats a long shot if no return I will grind it useless as is.

 

[Mitch Alsup]

Any reason you cannot chuck up an end mill in a collet and check for run out on a surface gauge,
and then use that data to mill the faces true and square ?

 

[mikey]

All was tight all was clean all was repeatable ...

And yet something is off. But as you say, return it. Maybe the next one will be dead on accurate.

 

[mksj]

How do you know it is the ER taper is off vs. the mt end, have you tried to clock the chuck in the MT taper at say 90 degree increments to see if the TIR changes? The additional run-out with the pin could be because of skewing, poor collets or as you say poorly ground. I would also check the +/- readings along a bar out to say 4" to determine skew vs. off center. Problem is if you recut/regrind the taper the nut will still clamp off center then. Not sure to what level these are hardened, put when I make my ER chucks I use a set-tru design and cut the ER taper and nut threads on the lathe so it all runs true.

 

[Stockyj]

The latest is I received a full refund very happy with that. Today I made a holder for my Air grinder will try and grind it true tomorrow will report on the outcome.

 

[Lo-Fi]

I had much more success with a quality carbide tool fixing the taper on my (15 thou runout) Bridgeport spindle than with the die grinder. The grinder tended to simply "follow" the runout no matter how gentle I thought I was being.

While not particularly hard, the Bridgeport spindle destroyed HSS very quickly. Carbide left a gorgeous finish. I'd expect the same from a collet holder.

 

[Stockyj]

Thanks for your reply I had thought of using a carbide I do have a very good Boring bar and quality tip. did suspect that the grinder will only follow the runout as the cut is so fine. With a 0.04mm runout, how much cut do you recommend a light or go under the runnout.

 

[higgite]

Before I’d grind on it, I’d do what I mentioned in post #9 and mksj suggested in #15, that is, test it at different clock positions in the spindle. But, I would take it a step further. Not only check TIR at each position, but also mark the high and low spots to see if they follow the repositioning or stay at the same spindle position. This will tell you for sure whether the problem is with the chuck or your spindle bore. The spindle bore can be dead nuts zero runout at the mouth but still be skewed. Just a suggestion. At best, it’ll verify that your spindle is okay. At worst, it’ll indicate that that the spindle is the culprit. Either way, I’ll sleep better at night just knowing.

Tom

 

[mksj]

I would setup the angle with a sensitive test indicator, use a boring bar, choke up on it so it is as short as possible to minimize flex. You maay also want to ink up the taper and mount up a collet with a rod in it to see if there are high spots or the taper is off. With the boring bar make sure the tip of the insert is at the center-line of the spindle. Ideally for shallow cuts in steel I use a sharp ground edge insert with a positive edge geometry, something like a CCGT 32.51 type which is typically used in aluminum. A molded insert or negative rake will not cut the metal in a shallow cut, wit a low feed. I then use a very fine sandpaper to polish the taper and remove any ridges.