Measuring Threaded Spindle Runout???

[cdhknives]

I don't know if runout is even the correct term. I'm trying to make a 1/2-28 tpi mandrel. I chucked up a piece of 1018 barstock and performed all operations to build the mandrel without loosening the chuck (3 jaw+dead center as required) so at this point everything should be as concentric and aligned as I can make it on my equipment. I actually made several, so I got better and better till I screwed something up (cut threads too deep, for example).

Rev 2 shown (with thread protector and 3 thou undersized threads), rev 3 lengthened center full diameter section for ease in centering with indicator and reduced drive shank to 1/2" for use with collet (eventually).


SO, as the last step I pull back the tailstock so I can part off the nub where the dead center was, and the cutoff blade hangs and stalls the lathe (to belt slippage, not locked in but these belts are grippy). It looked like there was a bit of wobble afterwards in the threads but after finishing the cut with a hacksaw, chucking the mandrel into a 4 jaw (no collets yet) and dialing in the shank I'm not so sure. I need to find a way to measure the threaded shank with my standard dial indicator.

I tried running the indicator over the tops of the threads and noting the highest and lowest values but I have low confidence in this. Given the large (relatively) radius of the nose on my indicator I don't see it reading the 28 TPI thread well.

Should I try lightly gluing a thread wire to the indicator tip? I don't have much budget for specialized measurement solutions. What else will allow me to accurately measure the runout at the threaded shank of this mandrel to check and see if the cutoff blade hanging has bent my mandrel?

 

[francist]

What about placing a 1-2-3 block resting up against the threads and position the indicator off the back of it? In other, a block between the indicator and the threads. If there was any runout, it would slide the 1-2-3 block slightly and that would show on your indicator?

Bit of a loosely-goosey way to do it and would only work for one revolution before you'd need to set the block snug again but it might show you something. Looking forward to someone else with a proper way to do it. Button head indicator i assume you do not have.

-frank

 

[darkzero]

If making a threaded mandrel of some sort where runout is critical, it best to make the threaded mandrel with a register. The register will help maintain repeatability rather than relying on the fit & runout of just the threads.

 

[stupoty]

You could try holding some thin shim stock between the plunger and work piece, the shim would need to be thin enough that it moves easily but hardy enough that it dosn't just snag and bundle up.

Not good for total accuracy but will give you an idea if its got bent for example.

Stuart

 

[janvanruth]

runout in the threads is not that relevant
the piece being held in the mandrel will be kept not by the threads but by the axial register
all threads have play and so does the radial register
similar to the threads in a lathe spindle, the chuck is held against the axial register which provides the fixed point

 

[cdhknives]

My goal is to make a muzzle break for a rifle. The idea is to drill and thread the raw rod stock, cut it off, mount to the mandrel, and shape the outside and face. Therefore the concentricity is critical. All operations after the boring and threading would take place on the mandrel.

Don't lose me with jargon. When you talk about a 'register' are you referring to keeping a flat section ahead of the threads for using with the dial indicator? My latest design kept a longer shoulder behind the threads for that purpose. I had such and was cutting it off (and possibly bent it, hence this thread) as it was too long for the short brake I was trying to make. I was trying to duplicate the thread length I see on rifle barrels. To my recollection the brakes I have looked at do not have a lot of extra volume on their interior...they narrow to just over bore diameter pretty quickly.

Right now I need either a 1 time way to verify that I didn't bend it when my cutoff tool hung up or positive advice to scrap it and move to revision #4 of this little learning experience!

 

[awander]

If you were parting it off when the blade jammed, I don't see how the piece that was being parted off could get bent-I think it would bend just the piece that was clamped in the chuck.

 

[cdhknives]

I had already parted the mandrel off the bar, and was trimming some extra length off in front of the threads with the shank of the mandrel held in the chuck. I actually considered just facing it repeatedly until I got it to length, but needed about 1/2" removed so I went for the quick route.

 

[janvanruth]

now i see

no problem
screw the threadprotector up against the shoulder, the "register"
if it makes contact all around it will be ok