Grinding on a lathe with a toolpost grinder

I have a hardened shaft I need remove about .0004 tenths from the OD. It's hardened and ground but a smidgen too big. I plan on doing it on my lathe using a toolpost grinder. Not the best setup, that all I have.

Is there a preferred method as I slowly rotate the chuck..in the direction of the stone rotation or against the stone rotation. I plan on using the powerfeed to slowly traverse the short distance to be ground. Does this sound OK?


Thanks,
Stuart

IMHO you will have difficulty taking .0004 from a diameter with/on a lathe. To come in and take .0004 is a skill talent with an OD grinder. Having the wrong grit or hardness wheel and it is likely to burn the part, or heat swell and be off size. likely you won't run true .0004.
likely will hit at one end and take more off that end.
Note .0004 can often be taken with abrasive paper. auto wet paper takes stock very fast.

First what lathe and what part size and length?
How will you hold the part? (between dead centers is likely the best/only way)
What wheel?
Can you run wet?
Do you have a diamond to dress wheel?
but still determined...and you have Ok answers for above.
1aaaa have the lathe warmed up so spindle does not change
1aaa. be sure the center are grease so thy don't stall or drag.
1aa. you might try to grind a something, perhaps a mandrel to get the feel for grinding.
1a. figure a way to tarp off the machine so coolant and wheel grit can not get into the machine.
1. First indicate the part in rotation, and in length to find less than .0002, both checks..
2. I would mount the grinder in the compound,at slide center with the compound 2 degrees off 90 so almost parallel to your long bed ways, so .100 on your dial would make about .035 in-feed to the part.
3. with having the compound about center on your cross slide.
4. Mount a diamond dresser some how at part center level and dress the wheel.
5. with spindle off bring wheel to rub a .001 feeler (piece of the clear wrap off a cigarette wrap.
6 travel down the part on find the same feel at 2 ends and the center&.
7. take out the feeler to miss the part at the 3 places and the compound in feed to get a wheel rum at 3 places
8. note (write down) the cross and compound number.
9. pull away both and come back to numbers to see the very same feel to the part.
10. very likely the grinder running will not have the same numbers as when parked so expect it to first touch at a different number, this may be .005 different so much care needed.
11 fire up at 200 rpm back way. up at face..this expecting the wheel and sparks going down.
12 Turn on coolant and begin travel back and forth With wheel thought to be .005 away.
13.come to tail end carefully and see if you get a grind.
14. travel end to end to see if all clear.
15, in-feed compound little by little with traveling long until you get the first spark.
16 likely that first spark may be all you need so travel long 4 times and see if it takes a little form entire part.
* Note this is very close work. you need only take .0002 Off the in-feed side as the other side will lose the same amount to make the .0004..likely to close a job for a lathe.
I would seriously advise the abrasive paper route.
Let me think about this over night before rushing to run the part..just in case I change something.
Buck
oh just thought ..put the indicator on the parked grinder to give a one finger push in all directions to se what effect that has and write it down like a map.
so, yes likely i will have more of change something.
Enough of this for now..

Lathe is a 17" Wacheon clone. I'll dial the part, which is about 2" OD X 2" long to zero in a 4 jaw chuck. A 3" white 60 grit wheel running wet at 6800 RPM.

I can't screw this up so it my be the wet paper drill!

Stuart

If the shaft is less than a few inches in diameter, and you're spinning it at under 100RPM, toward or against the stone rotation won't matter too much.

Since you are taking off so little, I strongly recommend that you use an excellent test indicator and generous amounts of time to make sure your shaft is concentric and parallel to the ways before starting up the grinder. Then, if you have a "harker" or a doctor's/mechanic's stethoscope, use that to determine when the stone just barely kisses the shaft. With that little material to remove, you might not see visible sparks.

Here are some pictures of the victim. I have some extremely precise DTI's I can dial this thing is with. I have two of them to do..it's the larger diameter that's a RCH to big for the hole it seats in. These are hardened punch dies, and the shop that made them in Canada is not very responsive to making them fit properly.

Stuart

thats a DOC of .0002" with no chance for a trial cut and you'd need alignment all along the axis of the work to a tenth.... sounds like a big challenge

otoh, an easy for a Q&D lap....just like this. I've even done this after grinding for the finish and relative ease at getting to a tenth

Edit: when I wrote that I didn't envision how short those parts are, makes grinding perhaps a bit easier, maybe lapping a little less so....still...I'd feel I have more control and accuracy with a lap

Thanks...I didn't think of a lap.

Stuart

Mcgyver's advice nailed it.

After I got the parts in hand they measured .0025 heavy which was good. It took more time as the passes were very conservative but I'm creeping up on it. Dialing in the part with a 10th indicator showed me my spindle bearings might need some attention, which I knew.

This experiment impressed upon me the need for quality, tight machinery made specifically for grinding to get a accurate finished product...not to mention knowing ones arse from a hole in the ground on how to do it.

Thanks for the help.

Stuart

Don't most lathes have two slides up top?
If the top is set at 5 degrees what is the infeed per diameter or radius of the grinding wheel to part per thou on the dial? Hello microns.
Why do people not use this when grinding in a lathe?
Bob

CarbideBob said:

Don't most lathes have two slides up top?
If the top is set at 5 degrees what is the infeed per diameter or radius of the grinding wheel to part per thou on the dial? Hello microns.
Why do people not use this when grinding in a lathe?
Bob

Click to expand...

Because the wheel wears with each pass and throws the math down the toilet...you need to re-dress and spark out again. At least with my half assed setup I did.

I'll add that I have DRO's in .0001 scale which, although sketchy, give you a hint of "who's on first"!

Stuart

When I was an apprentice, my master told me if he caught me using a DRO on a lathe I'd be sweeping the ceiling for the next week.

I used to tell lathe guys to use your wife's best towels to cover the machine from abrasive grit..that way you will do it less often.

atomarc said:

Because the wheel wears with each pass and throws the math down the toilet...you need to re-dress and spark out again. At least with my half assed setup I did.

I'll add that I have DRO's in .0001 scale which, although sketchy, give you a hint of "who's on first"!

Stuart

Click to expand...

I've got a little thing I made the clamps on the dovetail and holds a Mitutoyo tenths digital indicator, very handy given you can switch between imperial and metric as well as direction. Still, its a lathe.....engage the power feed and it'll move a tenth. For really fussy work like internal grinding for high end AC bearing fits, I always engage the feed before dialing in the cross feed, and press against it a bit by hand to load the screw up. More or less works but a tenth is bloody small.

Don't most lathes have two slides up top?
If the top is set at 5 degrees what is the infeed per diameter or radius of the grinding wheel to part per thou on the dial? Hello microns.
Why do people not use this when grinding in a lathe?

Click to expand...

It does give a more infeed resolution, you also have its internal clearance that you've now maximized the effect of by having it slewed over. It's hard to work a tenth this way. I get rid of it altogether for fussy stuff and use the T slot thing below which I made and scraped to the crossfeed, with digital indicator for infeed.

For fussy stuff working to tenths on the lathe I always set my cross slide "close but not quite" then give the apron handwheel a little back and forth jiggle to let everything settle out. Then approach final position and another gentle jiggle. This keeps things from moving around when engaging feeds or pressure on the cross screw settles out due to stick/slip etc. I also use a small Compac AGD #1 style .000050" dial indicator on a strong mag base registering against the slide for final tool setting when working to tenths. Readouts are great, but I like insurance for tight work.

Oh, and on this kind of work I prefer to place a stop block in the rear of the compound rest underneath by the handwheel/dial screw and lock it down tight. The less capability for "lost motion" the better.

Some 'post' grinding observations, just for the 'halibut'.

I have little experience doing this, was nervous as the parts were expensive and if I ruined them I would be up the proverbial creek. The first one I tackled came in perfect on finish size but had a little surface chatter on part of it. The little Themac toolpost grinder has accommodations to position the grinding arbor at the proper height for diameter of the item being ground. I failed to do that, and running the lathe FWD and the grinder CCW and with the grinder arbor incorrectly adjusted, made the wheel try to climb up the part as the wheel became dull. Green pea mistake corrected on the second part by speeding up the lathe a tad and running it in REV..opposite to grinding wheel rotation.

The small, white 60 grit wheel gave a beautiful finish to the hardened part, almost as perfect as the original finish, I was impressed. It almost made me feel like I knew what I was doing. Just for fun I did a search for cylindrical grinders and found a bunch of big old guys, full of tooling for under 2K, which I found interesting..not that I would ever need one.

It's amazing how much power it takes to rub a bunch of rocks against a piece of steel.

Stuart

atomarc said:

Because the wheel wears with each pass and throws the math down the toilet...you need to re-dress and spark out again. At least with my half assed setup I did.

I'll add that I have DRO's in .0001 scale which, although sketchy, give you a hint of "who's on first"!

Stuart

Click to expand...

Grinding wheel used in grit and hard makes a difference.
Was this a surprise to you? Grinding is not turning.
Bob

I used the only wheel I had, have no experience doing this, realize turning and grinding are different and am surprised by stuff everyday.

I have a Covel universal tool & cutter grinder that has lots of tooling and wheels but is not powered in the X axis. I felt the power-feed on the lathe would give me much better results.



Stuart

CarbideBob said:

Don't most lathes have two slides up top?
If the top is set at 5 degrees what is the infeed per diameter or radius of the grinding wheel to part per thou on the dial? Hello microns.
Why do people not use this when grinding in a lathe?
Bob

Click to expand...

Because all that s**t MOVES with a great deal less than perfect predictability at these tiny numbers? It's an "engine" lathe. Decent one, but still..

"Backed" abrasive or the lap.

ELSE send it to a GOOD OD grind shop where they do this stuff every day, all day, all year.

Not mentioned - though I trust we ALL "know it", or will twig to it as soon as mentioned - is that EITHER of those DIY methods "float" the abrasive and its support with the workpiece as their source for alignment and reference.

AND NOT .. anything else on the lathe at all.

That takes bed, carriage, and feed perfection or lack-thereof right TF out of the problem zone.

Slow down so abrasive and backer - or lap - aren't excursion-limited by high-cyclic forces vs their inertia?

It even "helps" with spindle bearing or chuck/collet centering imperfections.

So the Wacheon still has a lot of useful life left in it, after all.

QED

"You, too, Stuart!!"

Mcgyver said:

thats a DOC of .0002" with no chance for a trial cut and you'd need alignment all along the axis of the work to a tenth.... sounds like a big challenge

otoh, an easy for a Q&D lap....just like this. I've even done this after grinding for the finish and relative ease at getting to a tenth

Edit: when I wrote that I didn't envision how short those parts are, makes grinding perhaps a bit easier, maybe lapping a little less so....still...I'd feel I have more control and accuracy with a lap

Click to expand...

two question on that lap:

- doesnt the long handle tend to cock the lap?
- how hot does your lap get (mine get too hot to hold at 300 rpm or so)