What's new
What's new

Magnetic Chuck for thin round shim grinding

jeepinmatt

Aluminum
Joined
Oct 25, 2009
Location
Charlotte, NC
I need a good solution for holding thin round shims on my rotary table grinder (preferably as thin as 0.002"/0.05mm, but a good solution will work for thick or thin). We currently have a Suburban fine pole permanent magnet, which i think has lost its strength, so I want to take this opportunity to get the "right" tool for the job. Shim diameter ranges from roughly 1" to 4", with the majority of the work in the 2-3" range. FWIW, grinder is a Delta LC400 (Delta LC400 rotary table grinder- Surface Grinding Machinery UK )

I've seen the radial pole magnets, but not I'm sure they would offer any more clamping force due to the hollow center of the shims. What says the wisdom of PM?
 
I need a good solution for holding thin round shims on my rotary table grinder (preferably as thin as 0.002"/0.05mm, but a good solution will work for thick or thin). We currently have a Suburban fine pole permanent magnet, which i think has lost its strength, so I want to take this opportunity to get the "right" tool for the job. Shim diameter ranges from roughly 1" to 4", with the majority of the work in the 2-3" range. FWIW, grinder is a Delta LC400 (Delta LC400 rotary table grinder- Surface Grinding Machinery UK )

I've seen the radial pole magnets, but not I'm sure they would offer any more clamping force due to the hollow center of the shims. What says the wisdom of PM?


"We" say you are nuts! :D

Pics worth a thousand words here. Are you grinding a profile or something into the shim stock??
 
Thin parts can be difficult to hold with mag chucks. Before guessing on the Suburban, consider sending it back to them for service and review, contact them to get an estimate for the cost.

For your needs I'd consider a frost chuck, or a custom (very fine grid) vacuum chuck. For a frost chuck, you'd have to confirm grinding didn't raise the temp of the part too high, losing the frost.

But I would also think about keeping a supply of various shim materials in stock to meet the thickness need, then either laser cutting to suit the needed geometry, or profile milling (which would require more attention to burr removal unless you sandwiched the stock between Al sheets and did a clamp-partial profile-reclamp strategy).
 
Going that thin is not going to happen.

We grind HSS all day down to .032", so I am very familiar with thin parts.

Any fine pole chuck, and depending on part size, may be good to around .015" or so, and this is really pushing the limits of magnetism.

I don't even see the possibility to glue parts to a plate.

What is the purpose to grind something so thin?
 
I know 0.002" is, uh, "ambitious" :D But hey, why limit myself if there's a good option out there to accomplish it. Application is preload shims for radial contact ball bearings. Not grinding a profile, just grinding to a precise thickness. We already have a variety of shim thicknesses for each ID/OD combo, ranging from 0.002" to 0.040" or more in various increments, but they want to be measuring and grinding within tenths. I would argue with the boss about temperature, procedure, measuring equipment, operator training, etc, but I'd just get the "this is how they do it in Germany" line :blahblah: I know enough to know that we don't have the equipment or processes to even measure what they want, but if I can find a way to shave a 1/2 a thou off a 10 thou shim without it floating around the magnet, I'll be a hero whether we can accurately measure it or not.
 
Shim BOTH inner & outer races, grind ONE shim to get the required difference starting with 'sensible' thickness shims like the 0.040" you mentioned.
 
I know 0.002" is, uh, "ambitious" :D But hey, why limit myself if there's a good option out there to accomplish it. Application is preload shims for radial contact ball bearings. Not grinding a profile, just grinding to a precise thickness. We already have a variety of shim thicknesses for each ID/OD combo, ranging from 0.002" to 0.040" or more in various increments, but they want to be measuring and grinding within tenths. I would argue with the boss about temperature, procedure, measuring equipment, operator training, etc, but I'd just get the "this is how they do it in Germany" line :blahblah: I know enough to know that we don't have the equipment or processes to even measure what they want, but if I can find a way to shave a 1/2 a thou off a 10 thou shim without it floating around the magnet, I'll be a hero whether we can accurately measure it or not.

With an internal fitted disk to support the ring, you could lap to a precision thickness using a single or double-sided lapping machine. I'd consider this method if you "have to go there". Might need to make duplicates for support around the platen.

Lapping Polishing Honing Grinding Machines Metallographic Products Services | Lapmaster Wolters just one brand, there's others.
 
Going that thin is not going to happen.

We grind HSS all day down to .032", so I am very familiar with thin parts.

Any fine pole chuck, and depending on part size, may be good to around .015" or so, and this is really pushing the limits of magnetism.
Prezactly.

Poles get sore expensive to squeeze together, strongest magnet in the world can only apply force to what is in its field to GRAB.

This is into vacuum - where not leaking is as good as it gets, thickness immaterial, - or into adhesives, where a surface area bonded to is a surface area, not the depth back of it.

Disclosure: Diaphragms. Microphones & reproducers. Tiny ones. In house.
 
Shim BOTH inner & outer races, grind ONE shim to get the required difference starting with 'sensible' thickness shims like the 0.040" you mentioned.
If it were that easy, we would :D Unfortunately many of the shafts only have enough clearance for a few thou, and they are hardened precision shafts so we cannot (easily) modify the shafts.
 
This is into vacuum - where not leaking is as good as it gets, thickness immaterial, - or into adhesives, where a surface area bonded to is a surface area, not the depth back of it.

I agree this would probably be the best approach. It's definitely not something I'm familiar with, but the concept is simple enough. However, I know enough to understand that great execution of a mediocre concept often provides more success than mediocre execution of a great concept.
 
Buy a hand operated 6" wide jeweler's roll mill from PM member Kevin Potter. The roll will allow you to squash the shim to the correct thickness.

Keep in mind that the shim stock was produced on a powered roll mill. The shim will survive a few more passes before work hardening becomes a problem.
One concern with this method is maintaining the roll alignment so that they remain parallel within your .0001" requirement
 
Buy a hand operated 6" wide jeweler's roll mill from PM member Kevin Potter. The roll will allow you to squash the shim to the correct thickness.

Keep in mind that the shim stock was produced on a powered roll mill. The shim will survive a few more passes before work hardening becomes a problem.
Interesting. This is something I am completely unfamiliar with. How good is the parallelism/consistency of thickness and repeatability? Do you have a link to more info? A google search resulted in a bunch of things for sale and not much useful info (and I also don't really know what I am looking at :cool:)
 
A search on the forum using the keywords "roll mill kpotter" will bring up the thread discussing the construction of the roll mill. You will need to email Kevin to find out what accuracy his roll mill can maintain. He may be able to direct your search to another roll design if his mill is not suitable.
 
It is an interesting idea but I have used many jewelers rolling mills and they are definitely not going to hold tenths! Mr. Potter makes very nice tools, but this is way outside their intended tolerance range.

What about grinding a thin section on something larger so you can hang on to it and then cutting the washer shape out of that?
 
Start with the thinest one you can make. Then acid etch them to the thickness you want.

LOST

BINGO!

Uber-accurate final shapes as well as thickness.

"Chemical milling".

Made S. Hattori a wealthy company.

Made Seiko watches so good for the money I once had a certified Rolex watchmaker keep coming to borrow my one - heavy 23 jewel "High beat" tachometer thick as a slice of ONION on yer wrist, PX price, Long Binh all of 25 bucks - to check his electronic master timer.

"This watch has NO RIGHT to be as accurate as it is!"

Indeed it did not. As with MOST Rolex guru's when no customer were around, he wore an Omega. Which could not match that Seiko.

But chemical milling hadn't gotten the "no right to" memo. So it was accurate out of simple ignorance.

Well-proven technology, more than fifty years on, chemical milling has become.

Put it to your service, or hire it from those who have already.

:D
 
OK, warning heresy coming up! If your grinder/chuck combo is good enough for grinding the shims, thicker ones admittedly, can you get approval to grind a face on each bearing race to accommodate shims you DO have? Preferably the unmarked faces of course.
 
We already have a variety of shim thicknesses for each ID/OD combo, ranging from 0.002" to 0.040" or more in various increments, but they want to be measuring and grinding within tenths..

The Delta rotary table grinder has a fine feed increment of 10 microns. It does not have the accuracy that you need for setting bearing preload using shims.


Your work involves assembling angular contact bearing spindles. The procedure includes measuring free motion in the bearing assembly, adding a bearing preload distance, and then making the ring shaped spacer with a
+.0001 to -.0001" tolerance on the thickness and roughly the same tolerance on the out of parallel faces of the ring. You need to be able to produce the custom spacer ring with in a few minutes. Sending the part out for grinding or etching will not allow you to meet the schedule.

One other suggestion:
Use a spray type temporary glue to fasten the ring to a solid surface face plate on a lathe. Mount a roller burnishing tool in the tool post and cold work the ring to the specified thickness. The compound rest can be set to a shallow angle for feeding the roller to provide the necessary accuracy. The carriage will be clamped to the bed. The accuracy of the ring will be limited to the accuracy of the lathe headstock spindle bearings.

The cost of this process is the price of the roller burnishing tool and the custom face plate. There will be no burrs or dust generated.

The roller force is normal to the surface. The tangential force will be a small fraction of this force and will not be sufficient to overcome the friction between the ring and the face plate. The roller will be domed shaped and will be cold working a small area as the face plate rotates.
 
OK, warning heresy coming up! If your grinder/chuck combo is good enough for grinding the shims, thicker ones admittedly, can you get approval to grind a face on each bearing race to accommodate shims you DO have? Preferably the unmarked faces of course.
I do not think this would be acceptable, and I think we would end up creating other problems since the guys doing it are not machinists. If that were acceptable, I think the easier path would just be force engineering to change the design and always plan for a 30 thou plus shim.
 
Your work involves assembling angular contact bearing spindles. The procedure includes measuring free motion in the bearing assembly, adding a bearing preload distance, and then making the ring shaped spacer with a
+.0001 to -.0001" tolerance on the thickness and roughly the same tolerance on the out of parallel faces of the ring. You need to be able to produce the custom spacer ring with in a few minutes. Sending the part out for grinding or etching will not allow you to meet the schedule.
This is a very good summary of the situation.

The Delta rotary table grinder has a fine feed increment of 10 microns. It does not have the accuracy that you need for setting bearing preload using shims.
That is an excellent point. +/- 10 microns (0.0004") is what I have interpreted to be the target number. Like I said in an earlier post, our equipment, process, and measuring tools are not up to snuff, but for now, it is working with what we have. If we can't hold shims still enough to grind, I can't prove how insufficient the results are.

One other suggestion:
Use a spray type temporary glue to fasten the ring to a solid surface face plate on a lathe. Mount a roller burnishing tool in the tool post and cold work the ring to the specified thickness. The compound rest can be set to a shallow angle for feeding the roller to provide the necessary accuracy. The carriage will be clamped to the bed. The accuracy of the ring will be limited to the accuracy of the lathe headstock spindle bearings.

The cost of this process is the price of the roller burnishing tool and the custom face plate. There will be no burrs or dust generated.

The roller force is normal to the surface. The tangential force will be a small fraction of this force and will not be sufficient to overcome the friction between the ring and the face plate. The roller will be domed shaped and will be cold working a small area as the face plate rotates.
This is an interesting thought. I wonder how much effect it would have on the (not critical) ID/OD. I know this is something I can calculate, it's just too early in the morning for such things :D

Having a quick, repeatable, reliable setup would be the key to success. I think anything relating to the 1st point will detract from the 2nd and 3rd, and vice versa.

In my mind, a logical extension of that roller process would be to simply use a press and squish it. It would be easy to calculate force required based on size and material properties, but nearly impossible to maintain flatness and consistent thickness.

The simplest solution would be to buy shims in 5 micron (2 tenth) increments from 0.001"-0.002". That would give us more accuracy, certainly be faster in the end, and eliminate a lot of the human error side of things. The only downside is that we have about 50 different ID/OD combos, so that would be adding another ~250 sizes to our existing inventory of ~500 different shims.

Another thought branching off of that is to have 5 different thicknesses of shim stock, and a punch set for each of the 50 sizes. But I imagine 50 high quality, clean cutting punch/die sets would be pretty pricey, especially for the larger sizes.

:reading::nutter:
 








 
Back
Top