PDA

View Full Version : Chasing A Hardened Thread



JoeLee
06-29-2018, 12:15 PM
After going through all the time to thread the mounted points to fit my internal grinding arbor I found out that the arbor doesn't want to seat in the T&C grinders quill.

I've never used this 1/4"-28 internally threaded arbor because I couldn't find any wheels with that size threaded shank. (see my other post)

The the thread on the arbors is 3/8"-24. The one smaller arbor that takes the 5/40 threaded point screws into the quill and seats, it's a little tight just before it seats but it seats and centers OK.

The other arbor tightens up but is centering on the threads and not seating on the taper.

I checked the seating depth of all the arbors using an adapter busing. They all seat exactly the same depth, so it's not the thread bottoming out in the quill. The OD of the threads all measure the same, so it's all in the thread cut.

To further prove this I threaded the regular wheel arbor screw in a die, it went through with no resistance but the other two arbors start in the die but won't go far, a couple turns. That's where the problem is.
So....... it seems like my only option is to chase the threads on the arbor but HSS won't do it. Do they make carbide dies???? I'm not even sure carbide will cut them.

I don't like the idea of a taper tightening up by turning into a matching taper, not a great design in my opinion but no other way to do it.
The threads should be loose enough to allow the taper to seat.

JL..............

https://s26.postimg.cc/k6628en7p/IMG_20180613_142842.jpg (https://postimg.cc/image/k6628en7p/)

bobm4360
06-29-2018, 01:26 PM
Lap it with a brass nut and fine lapping compound. A sharp triangular hard Arkansas stone will also work.

old mart
06-29-2018, 01:31 PM
If the thread fit is best a little loose, why not rotate them slowly, and chase them with a small hand held triangular diamond file?

JoeLee
06-29-2018, 01:48 PM
The thread is not loose..... it's tight. The OD measures right. The thread isn't cut deep enough. A file won't cut it either plus it's a fine thread.

JL...................

CreakyOne
06-29-2018, 02:20 PM
The thread is not loose..... it's tight. The OD measures right. The thread isn't cut deep enough. A file won't cut it either plus it's a fine thread.

JL...................

You mean it's an internal thread and the ID is correct? How truncated do you think the internal crests look? Meaning, how shallow was the thread cut, in thousandths? suppose you take some 1/4-28 bolts and, using an adjustable die, reduce their diameter a few thousandths, just to see what it takes to get a good fit? If not much is needed, you could more easily reduce the external threads you just put on the arbors to get them to fit.

JoeLee
06-29-2018, 02:35 PM
You mean it's an internal thread and the ID is correct? How truncated do you think the internal crests look? Meaning, how shallow was the thread cut, in thousandths? suppose you take some 1/4-28 bolts and, using an adjustable die, reduce their diameter a few thousandths, just to see what it takes to get a good fit? If not much is needed, you could more easily reduce the external threads you just put on the arbors to get them to fit.No, It's the thread on the end of the two arbors pictured.
The OD is correct. The threads are not cut deep enough on these two. Real tight screwing into the quill.

JL................

adatesman
06-29-2018, 03:02 PM
OD doesnít matter; itís the Pitch Diameter that counts. Pull out your thread wires and check if the PD is within spec. It likely isnít.

Yondering
06-29-2018, 04:12 PM
Lap it with a brass nut and fine lapping compound. A sharp triangular hard Arkansas stone will also work.

This. Best way to loosen up a hardened thread is to lap it, and that doesn't take very long.

boslab
06-29-2018, 04:41 PM
Can’t you get it ground?, ive sent threaded components for grinding several times, ok not cheap btw, but it will give a very precise thread, like plug gauges and such, lapping sounds like a solution, diamond in a brass or Ali split nut? Like a split die perhaps
Mark

JoeLee
06-29-2018, 04:43 PM
OD doesn’t matter; it’s the Pitch Diameter that counts. Pull out your thread wires and check if the PD is within spec. It likely isn’t.Your right, the OD doesn't matter as long as it's within spec. The OD on these three arbors are ground so the thread tops have a slight flat to them. They all measure .365.

Here is a better picture. Using the hex die as a go - no go gage, you can see that it threads down the screw of the original wheel arbor with no resistance. It won't thread down the other arbors, I can just start it a couple turns. That tells it all.
I have a full profile insert that I can use if I can set these up in the lathe and line up the thread so it tracks.
I think the carbide will cut it, I can remove material with a file.

This is one instance where you need a loose fitting thread, and extra loose won't hurt. You need to be able to feel when the taper seats, as of now I can't.

I don't like the idea of lapping because you don't know what your lapping or how it's going to turn out. I could end up with sloppy rounded worn looking threads and still no fit.

JL.................

https://s26.postimg.cc/5d8yejuud/IMG_20180629_161907.jpg (https://postimg.cc/image/5d8yejuud/)

old mart
06-29-2018, 05:26 PM
Sorry, you misread my #3 post, I meant that if the fit you require is a little loose, ie, a smaller male thread is required, then put the shank in the lathe as for threading and use one of those tiny triangular diamond files to chase the thread. With back gear and lowest speed, it can be done by hand over short selected areas. Some of the half round ones have tight rads.

Arcane
06-29-2018, 05:33 PM
Your right, the OD doesn't matter as long as it's within spec. The OD on these three arbors are ground so the thread tops have a slight flat to them. They all measure .365.

Here is a better picture. Using the hex die as a go - no go gage, you can see that it threads down the screw of the original wheel arbor with no resistance. It won't thread down the other arbors, I can just start it a couple turns. That tells it all.
I have a full profile insert that I can use if I can set these up in the lathe and line up the thread so it tracks.
I think the carbide will cut it, I can remove material with a file.

This is one instance where you need a loose fitting thread, and extra loose won't hurt. You need to be able to feel when the taper seats, as of now I can't.

I don't like the idea of lapping because you don't know what your lapping or how it's going to turn out. I could end up with sloppy rounded worn looking threads and still no fit.

JL.................

https://s26.postimg.cc/5d8yejuud/IMG_20180629_161907.jpg (https://postimg.cc/image/5d8yejuud/)

I've always heard that if a file would cut the material, HHS would also so if the die you tried is slightly dull, perhaps that's why it doesn't work for you. If that's the case, perhaps a good quality brand new HSS die will bring joy to your life.

JoeLee
06-29-2018, 06:13 PM
Sorry, you misread my #3 post, I meant that if the fit you require is a little loose, ie, a smaller male thread is required, then put the shank in the lathe as for threading and use one of those tiny triangular diamond files to chase the thread. With back gear and lowest speed, it can be done by hand over short selected areas. Some of the half round ones have tight rads.I would give that a try if I had any of those files. I'll have to look into it, but 24 pitch thread is pretty fine. It would have to be a thin file.
I have some small pattern files in various shapes but even the thinnest one won't fit the thread pitch.

I don't think a knife file like this is thin enough to do the job on this thread. It says 1/16" thick. It doesn't say weather that is the thick edge or the thin edge.
https://www.mcmaster.com/#diamond-files/=1di0ml0

JL..............

elf
06-29-2018, 09:50 PM
A lot of threads here are hard to follow :rolleyes:

Rich Carlstedt
06-29-2018, 11:14 PM
Get a Dremel Grinder and mount a cutoff wheel that has been dressed on a pedistal grinder to give you a 60 degree profile.
Then mount it on the lathe and chase the threads.

Rich

CalM
06-29-2018, 11:58 PM
I don't like the idea of lapping because you don't know what your lapping or how it's going to turn out. I could end up with sloppy rounded worn looking threads and still no fit.

JL.................



You don't like the idea of lapping ? But the threads don't fit now.

OK Have it your way.

Paul Alciatore
06-30-2018, 12:28 AM
You say "To further prove this I threaded the regular wheel arbor screw in a die, it went through with no resistance but the other two arbors start in the die but won't go far, a couple turns."

A couple of turns suggests that the thread pitch may not be correct. Did you check that? Instead of using a thread gauge which is relatively short, I would check the questionable one against the good one.

J Tiers
06-30-2018, 12:46 AM
OD doesn’t matter; it’s the Pitch Diameter that counts. Pull out your thread wires and check if the PD is within spec. It likely isn’t.

Gee, I got a lot of heat for saying the same damn thing a few months ago. But,.... let that go....

The OD and minor diameter MATTER A LOT when it comes to a thread fit.

If, as it seems, one part is cut to the right PD and OD, but wrong minor diameter (flat too shallow), then for instance, if it is an external thread, a sharp V thread, or even a strictly correct modern thread will not fit, because it will jam on the too-large minor diameter. Likewise with a wrong (too large a flat on female thread) OD (major diameter), a sharp V or standard thread will jam on the OD.

Basically, the PD has to be right, but so do the other parts, otherwise the thread will not fit "universally" for its fit class.

JoeLee
06-30-2018, 10:29 AM
JT, you are correct. The thread cut is too shallow. Trying to lap it with a loose fitting nut isn't going to change the depth. In fact a regular nut will fit all three arbors with about the same amount of play you would feel in a standard nut and bolt. I'm guessing that some of the error may be in the internal threads. I don't know what KO used to tap the threads inside the quill but I'm guessing it was a regular tap, maybe worn or undersized. Who knows. I lost faith in them years ago.
So to blindly try to open up these threads with various (feel in the dark methods) would only be asking for more headaches.

Chasing the threads out with a good quality full profile insert will get me the results that I need. Using the hex die as ago-no go gage is the key to knowing where to stop.
The arbor has a center drill on the thread end so supporting in my lather won't be an issue. Getting it to track accurately might.

JL..............

lakeside53
06-30-2018, 11:00 AM
I recently chased the threads on 20 5C import collets. Better that than mess up my draw-bar. Yep.. pretty hard (too hard for the purpose in my opinion), but the 16ER insert worked fine.

Picking up the original thread isn't hard: I use a good light with otoscope or usb camera to get the tool dead center in the existing thread. Adjust the lateral position with the compound - be sure to engage the half nuts prior to take up any backlash. And for chasing, don't mess with compound in-feed; go straight in with the carriage.

old mart
06-30-2018, 01:10 PM
Those Mc Master Carr needle files look right, and they are priced at the top end, so should be quality The tapered triangular has a maximum width of 1/8", the point will be a lot smaller. I have a set of Kennedy and some cheapies, they come in very handy when an ordinary file skids off. I also have Vallorbe Swiss files, a couple of the triangular ones have been modified by removing one face to create a safe edge. They are very useful when restoring damaged threads.
If you singlepoint using laydown carbide, it would not be hard to pick up the part formed thread, once the shaft had been set true, even if you turned the lathe spindle by hand.

Peter.
06-30-2018, 02:02 PM
I chased out a hardened thread that someone had managed to mangle somewhat by dressing a normal mini-grinder cutting disc (not the thin discs) to a 60-degree point, making a lash-up mount for the grinder on my lathe tool post and hand-turning the spindle to traverse the grinder along the thread at the correct lead. It worked out remarkably well.

FWIW it was a high-amp watercooled alternator from a Mercedes diesel car. The pulley had failed and left the hub jammed on the shaft. The owner had hacked away at it with a mini grinder then a hammer and chisel to get the hub off but the thread was badly damaged so the new pulley wouldn't even start to screw on.

Yondering
07-01-2018, 01:12 AM
Trying to lap it with a loose fitting nut isn't going to change the depth.

Well, so don't lap it with a "loose fitting" nut. You apparently don't have much experience with lapping threads, but even so you could have had the job done already in the time it took to make the first post here.

Lapping with a correctly fitting nut absolutely will change the pitch diameter, and it's not the loosey-goosey process you seem to think. Lap it as carefully as you want, and check often until the threads fit perfectly if that's what you're after. If you lap it with a fairly snug fitting nut (either brass or steel) and fine valve lapping compound, you'll only remove a thousandth or two from the threads before needing to replenish the lapping compound, so that's a good time to clean it and check the fit.

However, you keep saying you want a loose thread fit, but then you're afraid of imperfect threads from lapping? Which is it?

strokersix
07-01-2018, 05:15 AM
Seems to me you could drill and tap several holes in a block of brass then cut away half to create multiple lapping surfaces. If you make the minor diameter a little small, say 85% thread depth instead of normal 70-75%, you would be able to lap your threads down to the root. And your brass block can have more thread length than a nut for accuracy and stability while lapping. And you don't have to thread your lap on each time, rather just lay it on the threads since it is open.

I have not actually done this. my thoughts.

JoeLee
07-01-2018, 08:01 AM
Well, so don't lap it with a "loose fitting" nut. You apparently don't have much experience with lapping threads, but even so you could have had the job done already in the time it took to make the first post here.

Lapping with a correctly fitting nut absolutely will change the pitch diameter, and it's not the loosey-goosey process you seem to think. Lap it as carefully as you want, and check often until the threads fit perfectly if that's what you're after. If you lap it with a fairly snug fitting nut (either brass or steel) and fine valve lapping compound, you'll only remove a thousandth or two from the threads before needing to replenish the lapping compound, so that's a good time to clean it and check the fit.

However, you keep saying you want a loose thread fit, but then you're afraid of imperfect threads from lapping? Which is it?No, I have never lapped threads before. Never had the need to. I also don't want to use my arbors as the test subject.
I've tried a few nuts on these and they are all a loose fit. If you look at the inside of the nut the threads are flat topped. How are they going to cut the root deeper ??
The root needs to go deeper and doing it with the proper threading insert will do just that while keeping the form correct.

JL..............

CreakyOne
07-01-2018, 10:29 AM
No, I have never lapped threads before. Never had the need to. I also don't want to use my arbors as the test subject.
I've tried a few nuts on these and they are all a loose fit. If you look at the inside of the nut the threads are flat topped. How are they going to cut the root deeper ??
The root needs to go deeper and doing it with the proper threading insert will do just that while keeping the form correct.

JL..............

Not all nuts are made the same, of the same quality or class. You need one with proper dimensions to do the job, which might mean making one (or more) yourself, probably of brass. Lapping works because the abrasive particles are worked into the surface of the lap and they abrade the surface of the work, very slowly. The particles also break down in the process, which is why they need to be replenished at some point. They obviously embed in brass better than in steel.
If your thread was greatly oversize, (10/1000"? -- just a guess), it might be best to start with an oversize nut (but having full threads) so full contact occurs over the entire thread without jamming the parts together. Otherwise, it would require lots of short screwing the nut on/off so only the front edge of the lap is working, gradually reducing the front end of the steel thread.

Since you don't want to do this on your arbor, as a test piece, it might be worth threading another shaft oversize, hardening it, and trying the procedure first on it. The process should come handy some other time, and you could use mild steel and Kasenit for the test to minimize distortion.

I agree with others that holding the arbor in a collet on a lathe and single-pointing it should work well, assuming you can see well enough to get the cutter aligned perfectly.

JoeLee
07-01-2018, 12:02 PM
That's very interesting and one of these days I'll try it, but not with these arbors. It's too risky to be experimenting with these.
Chasing with the insert is the safest way to go.

JL...........

Doozer
07-02-2018, 11:20 AM
Sulfuric acid with eat threads concentrically smaller (more or less)
and also will glass bead blasting. These methods have worked
successfully for me in the past.

--Doozer

CreakyOne
07-02-2018, 12:27 PM
Sulfuric acid with eat threads concentrically smaller (more or less)
and also will glass bead blasting. These methods have worked
successfully for me in the past.

--Doozer

Because acids or abrasive blasting work faster on the peaks than the valleys, they will change the thread form much more than other methods. In addition, sulfuric acid is great at attacking organic materials but not so wrt iron and steel.

Doozer
07-02-2018, 02:31 PM
Ever work on a car battery???

-D

CreakyOne
07-02-2018, 02:38 PM
Ever work on a car battery???

-D

Yes, I have disassembled and even repaired a couple lead-acid batteries, and have used many gallons of sulfuric acid, mostly for other applications though. If you want to dissolve steel, hydrochloric or nitric acid work much better...but not for the purpose you proposed.

Doozer
07-03-2018, 09:05 AM
Individual results may vary,
just stating what worked for me.
I guess everyone is wrong unless they meet your strict guidelines and criteria for a solution.
I am going to go drive my car to the moon. I'll see you when I get there.

-D

BCRider
07-03-2018, 12:20 PM
Diamond needle files will cut for sure. But the issue with the majority of triangular files is that they are actually not three sided but six sided. The edges are not a sharp edge but actually a small flat or radiused edge. I've had good luck using small needle files on coarser threads. But unless the file pretty well buries itself into the V the edge radius of the file means that by the time the right depth is obtained the walls are blown out. Been there and scratched my head about it until I realized what the real issue was. And finding a regular or diamond needle file that is small enough file for 24TPI is just not going to happen.

So what would be needed is a diamond version of a thread repair file which does have sharply pointed, or at least correctly sized flats, across the tops of the teeth forms.

So we're back to chasing with a carbide insert of the correct geometry to obtain a correct depth and root flat. Namely one with a fine enough nose radius on the insert to match or be sharper than the spec calls for. But basically for 24TPI I'd suggest that a dead sharp "V" form with no radius at all is what you're after. Or, in this case, you could likely get away with using a small brazed carbide tool with a sharp end. Pick up the pitch and chase out a couple of thou worth of material.

You could lap it but not like you seem to be thinking of by using any old nut. You'd want to get a proper 3/8-24 tap and make your own. You'd drill and tap into a round brass bar which is drilled slightly undersized so you get a full 100% thread. Then you'd want to split the nut through on one side and cut two more half cuts in towards the threaded hole to form flexing points so that it springs outwards as it threads onto the threads stubs. Or so it can be lightly pressed down onto the threads of the arbors by finger pressure or a light clamp of some sort like a rubber band or hose clamp. You're looking for a light turning drag. And as the arbor thread is lapped away you'll need to let the lap spring inward or adjust the clamp pressure to follow and keep cutting until you get the free spinning fit you're after.

There's very little material removal needed here and the thread lap will tend to cut evenly in a natural manner anyway. So lapping and having the thread move off center would require a big effort on your part. If you just let it float on the thread and just turn the lap on the thread it'll tend to self center.

This would not be a power operation by the way. All hand work.


You are under the impression that the lap will deform at the same time the thread is wearing away. But the beauty of lapping with a SOFTER metal is that the grit embeds into the soft metal and it stays the proper shape. It ends up being the harder material that is abraded away and the lap sees little or no wear. The times I've lapped items there was never anything but a dulling of the surface of the lap due to the pock marking of the grit. Within reason and used with the right touch there would not be any wear on the lap. In this case since the "target" is to deepen and clean the root areas of the threading you'd want to be sure to use a fairly light touch to avoid deforming the lap simply due to pressure on the brass crests of the "V"s in the lapping nut you make up.

But you're not sure about that so mostly I offer this last part on lapping for another day.

CreakyOne
07-03-2018, 02:21 PM
Diamond needle files will cut for sure. But the issue with the majority of triangular files is that they are actually not three sided but six sided. The edges are not a sharp edge but actually a small flat or radiused edge. I've had good luck using small needle files on coarser threads. But unless the file pretty well buries itself into the V the edge radius of the file means that by the time the right depth is obtained the walls are blown out. Been there and scratched my head about it until I realized what the real issue was. And finding a regular or diamond needle file that is small enough file for 24TPI is just not going to happen.

So what would be needed is a diamond version of a thread repair file which does have sharply pointed, or at least correctly sized flats, across the tops of the teeth forms.

So we're back to chasing with a carbide insert of the correct geometry to obtain a correct depth and root flat. Namely one with a fine enough nose radius on the insert to match or be sharper than the spec calls for. But basically for 24TPI I'd suggest that a dead sharp "V" form with no radius at all is what you're after. Or, in this case, you could likely get away with using a small brazed carbide tool with a sharp end. Pick up the pitch and chase out a couple of thou worth of material.

You could lap it but not like you seem to be thinking of by using any old nut. You'd want to get a proper 3/8-24 tap and make your own. You'd drill and tap into a round brass bar which is drilled slightly undersized so you get a full 100% thread. Then you'd want to split the nut through on one side and cut two more half cuts in towards the threaded hole to form flexing points so that it springs outwards as it threads onto the threads stubs. Or so it can be lightly pressed down onto the threads of the arbors by finger pressure or a light clamp of some sort like a rubber band or hose clamp. You're looking for a light turning drag. And as the arbor thread is lapped away you'll need to let the lap spring inward or adjust the clamp pressure to follow and keep cutting until you get the free spinning fit you're after.

There's very little material removal needed here and the thread lap will tend to cut evenly in a natural manner anyway. So lapping and having the thread move off center would require a big effort on your part. If you just let it float on the thread and just turn the lap on the thread it'll tend to self center.

This would not be a power operation by the way. All hand work.


You are under the impression that the lap will deform at the same time the thread is wearing away. But the beauty of lapping with a SOFTER metal is that the grit embeds into the soft metal and it stays the proper shape. It ends up being the harder material that is abraded away and the lap sees little or no wear. The times I've lapped items there was never anything but a dulling of the surface of the lap due to the pock marking of the grit. Within reason and used with the right touch there would not be any wear on the lap. In this case since the "target" is to deepen and clean the root areas of the threading you'd want to be sure to use a fairly light touch to avoid deforming the lap simply due to pressure on the brass crests of the "V"s in the lapping nut you make up.

But you're not sure about that so mostly I offer this last part on lapping for another day.

Excellent description, good detail, particularly re splitting the nut; much better than my short post.

strokersix
07-03-2018, 03:22 PM
I was on the right track with post #24...

BCRider
07-03-2018, 03:30 PM
I was on the right track with post #24...

Yep, you were. The key gain from making the "lapping nut" in the lathe from proper size round bar stock is that it can be split and spring cut to give three segments that press against the threading so they self align to a better degree than the two sided option. Mind you we're splitting hairs here. Mostly I think it would be easier to handle than a block with two pieces that need to be held together and might want to slide around out of alignment unless left with a web along one end of the cut so it springs inward and out but stays in alignment.... which would likely work well too.