[Metrology] Parallel Test Bar For Lathe Setup

petertha

H-M Supporter - Gold Member
H-M Supporter Gold Member
Joined
Jan 15, 2016
Messages
994
I've been on the lookout for a reasonably priced parallel test bar for certain lathe setup operations. I'd be happy with a cylindrical blank say 14" long as long as it had accurate centers ground on both ends. For some reason, 100% cylindrical test bars seem to be kind of rare. But lots with MT# taper ground on one side. OK, I can make that work too as long as the cylindrical section is long enough and I guess the MT will give me some added utility for different setups if I set it in my MT tailstock.

Basically I want to set up custom taper angles on my lathe compound using sine bar & gage blocks, items which I already have. In my mind this seems like a rapid way to get the compound orientated. I've done the DTI traversing across a known MT arbor shank. That works of course, but is also a bit fussy and one is always 'replicating' whatever that happened to be along with any errors as opposed to setting up from base [taper/foot] offset type standard taper dimensions.

So here is what I'm visualizing. Does this procedure make sense to you guys? Something like:
- put parallel test bar between centers, ensure headstock/tailstock centers aligned with various DTI measurements (which lathe should be to begin with anyway)
- set compound to zero protractor scribe line (just rough reference)
- loosen toolholder, set sine bar in tool holder & align so both lobes contact test bar simultaneously, tighten, zero reference established
- loosen compound, swivel rotate & use appropriate gage block stack under one sine bar lobe. Other lobe is contacting test bar, similar to angle setup in a mill vise.
- compound is now set to cut that taper angle & is repeatable. No coupon MT arbors were required. Also any desired angle within stack-up range should be achievable using method.

There seem to be lots of parallel bars like this on ebay claiming 0.0001" runout. Cheaper than I could make one to anything close to this accuracy. Any experience or feedback in this regard and/or my proposed methodology?

http://www.ebay.ca/itm/ETOOLS-3MT-1...=c2a0bfdd3ab34652ad56360073c9da0f&pid=100022&

3-27-2016 0000.jpg
 
That all sounds good, just do not assume that the side of the compound is actually parallel to it's travel axis. You need to test for that, both sides...
 

Tom Lipton from OxTools has a You Tube video on setting up a compound for cutting tapers using a sine bar. He sets the sine bar up along side the compound slide and sweeps with an DTI. The setup starts bout 27 minutes in.

Two assumptions are being made.

1. As Bob points out, the side of the compound rest is assumed to be parallel to the compound ways. It is not on my G0602. However a correction for that error can easily be made.
2. The z axis ways are assumed to be parallel to the spindle axis. We work at making this so with our lathes so we aren't unintentionally cutting tapers and for a properly set up lathe, this should be a good assumption. It can also be corrected for fairly easily if it is not.
 
CraigB1960
- thanks for the solid bar find
- I did see that Edge model too but if I understand that one correctly, you basically can only indicate on the outer rings on either end. The middle part is basically filler. I want to be able to line up the sine bar at any position along the test bar, so the solid bar version would be preferred in my case.

Bob & RJ
- I think I understand what you mean. I'll take some pics of the setup here in a bit. Basically the carriage stays put. The compound is first zeroed against the test bar with the sine bar clamped in the tool holder. It doesn't really matter what the degree setting says, its now the zero datum equivalent. Then the compound is loosened with sine bar still clamped in & gage block spacers put in. The y-axis (cross bed) has to go in & out a bit to both lobes sine bar lobes are contacting tangent.

Actually what I didnt count on was the fiddle factor ensuring sine was contacting properly & then slowly cinching down the compound without moving anything as it tended to want to torque a bit. Mill setups don't really have this issue & gravity is your friend with the sine bar assembly.

The other factor I didn't really pay attention to is when the compound is at a shallow angle (meaning nearly parallel to lathe axis) the turn handle rather interferes with the typical tailstock live/dead center. I'll have to play with this some more, Looks like the movie linked didn't have that, the part is cantilevered outside the chuck. One regret I have in lie is not buying a taper attachment for the lathe when I had the chance. The retrofit is a bit of a bugger & not cheap either. Oh well.
 
Look at Thompson linear shafting, from Enco. Very reasonable and you can check the specs for yourself. I got a 3/4" x 12" for something like ten bucks.
 
The best accuracy will be achieved by straight turning the actual part in place then placing the sine bar on the oversized diameter where the taper will be made and then adjusting the compound angle with an indicator. Then turn the taper as this will remove any tail stock/head stock misalignment because you have already set the taper with the sine bar with this error included, if the ways are worn all bets are off.
 
Back
Top