What kind of tip to indicate vitrified aluminum oxide abrasive?

[luke8888]

I have a need to reliably indicate vitrified aluminum oxide (like a surface grinder wheel) and messing with pieces of paper, or shims that rapidly loose thickness, you insert between the indicator tip and the abrasive gets very old very quickly.

So I decided to ask here, what kind of tip is resistant to abrasion by typical vitrified aluminium oxide abrasives? Carbide? Sapphire? Or is diamond the only option? I don't think I ever saw a diamond indicator tip, but diamond brazes well so it should be possible.

 

[winger]

Years ago one of the tool catalogs carried a tip with a roller. Probably Msc or enco.

Dave

 

[Bakafish]

Mitutoyo part #901991 roller wheel tip. The wheel itself on the one I have appears to be POM plastic, so no idea how long it would hold up, but it is likely the right idea. You might think about making your own using a small ceramic raced ball bearing which should give both precision and a harder surface.

 

[luke8888]

Thanks both. I've had a look at those rollers in the catalog. There isn't much info about them.

I already have a couple of sapphire and carbide tips so if anyone knew for sure they would scratch I could at least exclude the possibility I already have what I need. (I'll test on a carbide end mill later).

Let me say this tip is to be used on a stationary or for a slow sweep over an aluminium oxide wheel/block, not under power. The type of aluminium oxide will be white, grey and pink. So if a given material is harder than these that should be enough. Mohs scale says all those materials are 9. (perhaps because ruby and sapphire are chemically the same thing as aluminium oxide). In theory if they are the same hardness they shouldn't scratch themselves, but I doubt that (they grind diamond with diamond dust after all).

In general I'm not too keen on a roller tip mostly because it is unclear what precision I'd loose and they don't say which material it's made of, so potential for abrasion in the long run is still there.

I thought a ceramic ball would be a good idea, but I just checked they're mostly made of silicone nitride which is 8.5 mohs hardness (alox is 9, zirconia is 6.5). So carbide would be a better choice. Some sources claim carbide goes up to 9.5mohs so perhaps this is my best bet with sapphire et al being used before carbide became widely and relatively cheaply available.

 

[Bakafish]

I think anything short of diamond is going to noticeably wear. The question is, are you zeroing out the gauge each time, compensating for wear, or does this need to be a fixed reference over time? Because carbide tips shouldn't be that expensive, I got a bag full of Mitutoyo's for $5 at auction. Just let them wear, and replace then as needed. Otherwise my suggestion was actually a deep groove ceramic race ball bearing so it rolled like a wheel, which will reduce friction and wear to some degree.

 

[michiganbuck]

I once made a slip of .010 hard gauge shim stock to run on a wheel under my tip, so as to ride smoothly...a slip/strip of the nylon cut from a milk bottle would be good.
Rubber band or tape to indicator bar is an option.

To keep wheels true the wheel, spindle end, and the wheel mount should have a line-up mark. This is so gravity and the mount-up mark will make a dressed and taken-off-the-mount run near perfect next time it goes onto another mount/adaptor.

The mount-up also can help to retain the better balance of a balanced wheel.

New bigger wheels come with a mount-up mark so they can run near true at the first mounting.

Amazon.com : roller contact point for dial indicator

 

[luke8888]

I think anything short of diamond is going to noticeably wear. The question is, are you zeroing out the gauge each time, compensating for wear, or does this need to be a fixed reference over time? Because carbide tips shouldn't be that expensive, I got a bag full of Mitutoyo's for $5 at auction. Just let them wear, and replace then as needed. Otherwise my suggestion was actually a deep groove ceramic race ball bearing so it rolled like a wheel, which will reduce friction and wear to some degree.

If the tips are cheap and easy to get it is suffiecient that it doesn't wear noticeably during a single measuring session. So I think if I can find a source of those carbide tips at a good price that's good enough. I think you're right nothing short of diamond will last. I'll look for them.

I do have some carbide tips already (a very sharp point and a flat), but they came as a set with a vintage Czech indicator I'd like to keep as a set so I'd only use them on materials I know are not going to harm them.

I once made a slip of .010 hard gauge shim stock to run on a wheel under my tip, so as to ride smoothly...a slip/strip of the nylon cut from a milk bottle would be good.
Rubber band or tape to indicator bar is an option.

To keep wheels true the wheel, spindle end, and the wheel mount should have a line-up mark. This is so gravity and the mount-up mark will make a dressed and taken-off-the-mount run near perfect next time it goes onto another mount/adaptor.

The mount-up also can help to retain the better balance of a balanced wheel.

New bigger wheels come with a mount-up mark so they can run near true at the first mounting.

Amazon.com : roller contact point for dial indicator

I've done this before when measuring how true a diamond surface grinding wheel is before truing. I used copper foil shaped into a sort of cup and held above a test indicator contact tip with some sticky tape. It was ok, but it got about 2 thou thinner with each rotation of the wheel being measured. So for one time use it's definitely usable. If I had 20 of those wheels to check one after another I'd probably prefer a carbide roller.

 

[jwmelvin]

it got about 2 thou thinner with each rotation of the wheel being measured

Wow, that seems untenable. I wrapped scotch tape around a resin-bonded diamond wheel for indicating it, and it worked well.

 

[luke8888]

Wow, that seems untenable. I wrapped scotch tape around a resin-bonded diamond wheel for indicating it, and it worked well.

That is actually a pretty good idea. I have to try it next time.

It was a 120 grit, 8in diameter diamond wheel and a soft copper tape 0.2mm thick (8 thou). I can't be sure it wasn't some sort of deformation of that foil, but it did seem like it was being ground off. I also had a copper colored line on the surface of the diamond wheel. The pressure wasn't much. Just enough to zero a 0.01mm "Beta" brand test indicator.

 

[michiganbuck]

Name brand AO wheels run very true/close Od to ID. The wheel hole needs to be made a few thousandths larger than the mount/adaptor so it can go on. Indicating diamond wheels is common but indicating AQ wheels is not common because diamond dressing is so much faster and more accurate than indicating in.

Back when I used a slip of .010 shim stock under my indicator tip it lasted for many, many diamond wheels and did not wear through.

The tape idea is good because even simple masking tape will likely run .0005 true to thickness.

*Perhaps explain why indicating AO wheels is needed.

 

[TGTool]

Buck, I suspect there's one difference between how you do it and how he's doing it, if I'm reading between his lines right.

I expect that when you indicate your diamond wheel you bring the indicator up to the shim stock you have against the wheel, note or zero the reading, back off the indicator and hold it while you rotate the wheel 180, put the shim back and check the reading again on the indicator.

I'm wondering if he simply slides the wheel around without backing off the indicator so that he's abrading his shim as the wheel is turned. He might be doing that if he's indicating right against the abrasive so that he gets the average high reading among the alox grains.

 

[michiganbuck]

QT OP (original poster): I used copper foil shaped into a sort of cup and held above a test indicator contact tip with some sticky tape. It was ok, but it got about 2 thou thinner with each rotation of the wheel being measured.
Copper is likely the worst material to consider using.

I have never indicated an AO wheel

 

[luke8888]

I'm wondering if he simply slides the wheel around without backing off the indicator so that he's abrading his shim as the wheel is turned. He might be doing that if he's indicating right against the abrasive so that he gets the average high reading among the alox grains.

Exactly! But it was a diamond, not an alox wheel back then.

QT OP (original poster): I used copper foil shaped into a sort of cup and held above a test indicator contact tip with some sticky tape. It was ok, but it got about 2 thou thinner with each rotation of the wheel being measured.
Copper is likely the worst material to consider using.

I concur, but that's what I had on hand and I managed.

I have never indicated an AO wheel

It isn't about indicating an alox wheel on a grinder. Why would one do it when it can be trued "in situ"?

But I need to measure a number of wheels and blocks when they're off a machine. For QA purpose.

 

[michiganbuck]

QT OP: But I need to measure a number of wheels and blocks when they're off a machine for QA purposes.
Then the tape idea, or perhaps a sealed ceramic bearing, might be the ticket.

Amazon.com

 

[Clive603]

How small an area do you need to indicate on?
As the abrasive surface is rough with (hopefully) a statistically controlled variation in grain point height it seems you need to span several abrasive grains to get a meaningful result that isn't contaminated by grain to grain variation.
If the measuring area is large enough to accommodate a small elephants foot or similar probe end with a significant flat area the actual pressure of the foot / probe on the wheel will be significantly reduced. Possibly by enough to make wear nearly insignificant over decent periods of use. Cosmetic scratching seems inevitable but maybe not wear.

 

[michiganbuck]

Doing extensive wheel OD indicating a simple swing arm made using a carbide-tipped lathe tool bit with a hole drilled in the shank and the indicator resting on the tool bit backside might be a perfect long-lasting checking fixture.

carbide tipped tool bit tool bit - Bing

Intelligent search from Bing makes it easier to quickly find what you’re looking for and rewards you.

www.bing.com

 

[luke8888]

I failed to mention (and I probably should've in the beginning) I'm attempting these measurements to 200nm precision (that's 200 nanometers, or 8 uin - micro inches). I have a microcator and a stand that manages to do that providing multiple measurements are taken and the temperature is stable (gloves), but for reflective metals I just use an optical flat for this. Here we're talking about an abrasive ceramic. I can't make it reflective and bouncing a laser off it(I briefly considered interferometry, but That's too much hassle anyway) doesn't work, because the surface isn't smooth.

The size of the area measured is quite large, 2mm in diameter (80 thou) would be very nice, but it can be stretched even to 8mm if needed(a quarter inch) .

For anyone that thinks it is insane to attempt measurements hoping for such precise results when precision surfaces have to be protected from the object being measured is probably right. This is all in conceptual phase at the moment.

Regarding bearings and rolling elements. The runout would have to be half of my desired precision (so 0.1 micron or 4 micro inches) which isn't happening with my budget...

I'll most likely be using lapped slices of a carbide rod. 3 under my object being measured and one between the object and the tip. I'll find out very quickly if it scratches/abrades.

 

[CarbideBob]

Laser probes do not wear but give rather wacky readings on a wheel like this due to voids in the wheel structure.
Carbide ball the choice for contact measuring. They will wear a flat over time. This is actually good.
Probes or gauges can be fixed in the machine or removable.
Fixed in the machine and the developing flat is always in the right orientation.
On something like a DTI you never hit the flat dead on so your wheel tends to "hone" the edge between the flat and sphere.
Here you see first check one number, second check a differnt number, third check and things settle down but now your zero is now off.

Bigger question is why the need to indicate such a wheel?
It should dress true so I am guessing you are going for size control?

 

[jccaclimber]

When I've needed to laser measure off of thing that don't reflect nicely in the past I've put a ball on them at a known force and then measured the position of the ball (or whatever artifact). You'll end up with a flat on the ball, button, or whatever carbide widget you measure to so you'll need to calibrate out the thickness of that artifact. Since we're in the weeds and you used the word precision, I want to make sure you meant precision vs. repeatability vs. accuracy vs. any number of other things. Not trying to be argumentative, just really trying to understand what your limitations and goals are.

 

[luke8888]

When I've needed to laser measure off of thing that don't reflect nicely in the past I've put a ball on them at a known force and then measured the position of the ball (or whatever artifact).

Actually this is an excellent idea! (with some modifications for my requirements).

You'll end up with a flat on the ball, button, or whatever carbide widget you measure to so you'll need to calibrate out the thickness of that artifact. Since we're in the weeds and you used the word precision, I want to make sure you meant precision vs. repeatability vs. accuracy vs. any number of other things. Not trying to be argumentative, just really trying to understand what your limitations and goals are.

Fair enough. We're talking about precision. I don't have to hit any specific dimension so no high accuracy required. As for repeatability... I need my results to be repeatable within the limits of precision, but what matters is relative measurements across the surface to find out how far out of desired shape (normally flat is the desired shape) it is.

Your post made me think I can use optical methods to measure this. Instead of using my carbide pucks as described previously I can put 2 next to eachother and by putting an optical flat on top I can easily figure out which one sits higher and by what amount. The technique is described in detail in the DoAll book on measurements, but they use it to compare height and parallelism of blocks so I haven't connected the dots until I read your post.

Bigger question is why the need to indicate such a wheel?
It should dress true so I am guessing you are going for size control?

That's not it. These wheels and blocks are not being dressed by me. I don't have time to write a lot about it at the moment.