Forming a radius on a diamond grinding wheel

[M. Roberts]

Hello good people, TGIF. Does anyone have advice, knowledge, or experience with dressing a corner radius on a diamond grinding wheel? Details on the wheel: 12" dia. x 2" wide, 100g vitrified bond wheel, the desired corner radius is 0.04". Details on the part and material: part is composed of a SiC based ceramic composite. The part itself is flat; 13.5" diameter x 0.625" thick, with a 0.5" wide x 0.125" deep step around the circumference...the 0.040" radius is on the inside of the step. The thinking is that we could purchase a new Okamoto rotary (the reason for my previous post), grind the top of the part, true the radius (more on that), and kiss the step. The grinder that I am looking at has a single point diamond dresser built located under and behind the wheel, and has the ability to be programmed to perform a true/dress cycle automatically...not sure if it can do a radius though. Great, right? Now the issues...with diamond wheels, by far the best way to true them (that I have found) is by using a truing spindle equipped with a diamond wheel. What I really would like to do is to remove the single point, and install the truing spindle in its place, and when a dressing/truing cycle is called out, the truing spindle kicks on, and the machine does its thing...seem to crazy of an idea? Add some of your own if you like...Thanks, Mark

 

[eKretz]

I would expect truing a diamond wheel with a diamond to dull the diamond abrasive particles in the grinding wheel, but I haven't tried that personally. I'll defer to @CarbideBob or @michiganbuck on this one. I am interested to hear the response.

 

[M. Roberts]

Thanks eKretz. I had a conversation with reps from Okamoto, and I don't think they completely understood the need to re-form the radius on the edge of the wheel. I would think that someone has come across this before. Thanks again for the reply. Mark

 

[CarbideBob]

Okay,
How did that radius get there in the first place or was it never there and we are making a new one?
Assume this is a resin bond wheel that we can put a corner on?
How are you checking this rad and can you cut parts oversize to check what you have?
They make special machines for this. Your wheel maker can dress/true it but guessing that not a good option.
This is one swing machine but I think it ... 0h..https://rushmachinery.com/products/wheel-dressers/
Some people who need "teardrop" shapes for fluting actually send it back to the maker for re-truing.
How close do you need to be true to form? We do this very often by hand... like every day but only .001/.003 true to form
Would never try this it on a API threader tool.

Brake or motorized truing device. Normal stones go away like crazy. Comping for it a challenge doing radius.
Plated wheel/(stones) die fast.
Crush dressing uses a diamond plated roll with your form in it. They only work with very soft bonds which give poop for life.
Same with wheels you can use a Diaform on.

This material of a guess needs lots of wheel sticking when running. That pucks up your rad if not done right.
Tell me more..... and looking for @michiganbuck input.
Bob

 

[michiganbuck]

I have never had a dedicated radius dresser for a diamond wheel. At one shop we would down-grind carbide parts(when down-grinding was appropriate ) and set a wheel aside when a corner would wear to standard radius size. I have a handheld roller dresser that has a carborundum wheel and a very hard diamond wheel that I use to hand-dress a diamond wheel corner to a radius, to be checked with a radius gauge or looking with a comparator. A close part spec on an inside corner radius is a foolhardy design.
Rather than buying machines a plated wheel might be the choice. They are expensive and not that long-lasting but may be better than buying special machines. Often you grind the form/radius with a slow feed rate to conserve that kind of wheel.
At one shop we had a Cincinnati couture and template follower machine/grinder, that with a very hard diamond wheel would likely do a great job of grinding a radius to a less hard diamond wheel but we never tried doing that. Doubt one could find that rare machine today.
*Sending wheel out to a shop that has the dresser machine that Carbode Bob mentioned would be the berries... or if buying enough wheel the wheel manufacturer may put on the wheel corner radius.
SiC-based ceramic composite would grind wet so easily that a wheel would last a good long time, so the plated diamond wheel might be a good choice.
You would need to indicate it very close to get a decent finish, but that is easy.
Plated wheel with two corners having your radius might be good.
With electro chemical grinding wheel last a coon's age but tricky to set up on an Okamoto rotary because you need creep feed.

 

[michiganbuck]

The thinking is that we could purchase a new Okamoto rotary. Can you get the part diameter and step concentricity to spec with that machine?

 

[eKretz]

Bob, he mentioned in the OP it was a vitrified diamond wheel. I've never used one of those, but have heard good things.

 

[CarbideBob]

Bob, he mentioned in the OP it was a vitrified diamond wheel. I've never used one of those, but have heard good things.

My bad not reading that.
Same,same.
Vit wheels tend to have a more open structure. We use then in 800, 1200, 2000 grits mostly for PCD and CBN grinding.
400 and below we use resin bond wheels.
Needing to hold an internal rad at +/- .001 over lots of parts and plated wheel but those do not live real long or grind fast.
I avoid plated when ever possible. Mostly more for form work and we rough out all we can first.
A 6 inch by 1/2 wide plated from wheel costs about $400 and only has one layer of diamond.
If this material is like the SiC fiber in ceramic like the Greenleaf whisker stuff it will eat plated diamond wheels for lunch.
This whisker stuff is also very bad on your lungs so very wet and vented. Much worse than carbide.

Often these rads are for stress relief so they do not need to be perfect and you can put them on by hand using a old dressing stone.
Sort of art but actually easy once you do it a few times and a high school kid can master it in a day or two.

It actually sounds easy to me and not scary but do not know the details.
We do this in carbide, SiN, CBN, PCD all the time. Like Buck once we have a 1A1 at .0625, .040, .080 we label and keep it aside to make life easier.
Get up over .060 on a new wheel and we will dress a series of flats, then hand stick/stone it in. Once there it only needs gentle touch up and a dress of both sides.

Moly stick on a rad dresser sort of works but the stick keeps getting smaller and curved.
A round end PCD insert in like a VDB125-250 sort of works for profiling like you would do on a lathe for touch up.
We do this sometimes but the dresser insert is one and toss. Zippo cost here as we have a few scrap guys laying around with ends not perfect. Again probably not an option.

Every shop, every job is different....... Not much help and 2 cents worth of guessing here.
I feel worthless... Wish I had a straightforward do this, make lots of parts.

 

[M. Roberts]

Good morning all, and thanks for the input. The block tolerance on the 0.040" inside radius isn't tight; +/-0.01". These parts are rough machined in the "green/unfired state", including the radius, and after the firing process, the part is taken to the final dimension and tolerances. We used to use resin bonded wheels, but they do not hold up as well, ths the 100g VIT wheels. The desire is to use a single wheel to do the entire part, and not have to swap them out; regardless, at some point the radius on the wheel would need to be touched up.
MB; Yes, a new Okamoto Rotary would do the entire 1st side of this part (and hopefully the radius) nicely.
In an attempt to save my sanity, I changed jobs within my company; from a machining engineer in R&D, to a machine/equipment engineer...although the boss is easier to deal with, and I have already made a lot of eye-opening suggestions, it is not without the pitfalls of dealing with peoples attitudes when the shortcomings of their decisions are brought to light....I try to be politically correct about it, and try to lead them to the conclusion that "there is a better way", but that is not often understood. Sorry to go astray; I should save this for the "Psychological Help Section". Regards, to all that has commented, Mark

 

[texastoolmaster]

Hello good people, TGIF. Does anyone have advice, knowledge, or experience with dressing a corner radius on a diamond grinding wheel? Details on the wheel: 12" dia. x 2" wide, 100g vitrified bond wheel, the desired corner radius is 0.04". Details on the part and material: part is composed of a SiC based ceramic composite. The part itself is flat; 13.5" diameter x 0.625" thick, with a 0.5" wide x 0.125" deep step around the circumference...the 0.040" radius is on the inside of the step. The thinking is that we could purchase a new Okamoto rotary (the reason for my previous post), grind the top of the part, true the radius (more on that), and kiss the step. The grinder that I am looking at has a single point diamond dresser built located under and behind the wheel, and has the ability to be programmed to perform a true/dress cycle automatically...not sure if it can do a radius though. Great, right? Now the issues...with diamond wheels, by far the best way to true them (that I have found) is by using a truing spindle equipped with a diamond wheel. What I really would like to do is to remove the single point, and install the truing spindle in its place, and when a dressing/truing cycle is called out, the truing spindle kicks on, and the machine does its thing...seem to crazy of an idea? Add some of your own if you like...Thanks, Mark

The tool I always have used is called a brake truing device. It has centrifugal brakes inside the drum opposite the wheel. The wheels are usually around an 80 grit(dark bluish grey). I have put radii on diamond wheels for jobs before, but they were usually smaller(6-8") and putting the radius on is relatively easy. Have you spoken with a wheel manufacturer about just getting it on a new wheel for the machine. It won't add a lot of cost, especially such a small radius. When you mount the wheel it will probably run pretty close so truing dressing should be relatively quick. I however, have done it with really hard bond fine grit A/O wheels that I keep around after they are used up, break them into chunks and you can hand dress the edge by stroking the wheel around on the corner until it breaks down. Takes a little getting used to and be really careful not to injure yourself. I always find a comfortable way to sit or even lay so that easy access is achieved and comfortable. Start at a 45° and get that corner to start breaking down, just nice even strokes with the wheel and a radius gage to rough check with. Thin cardboard can allow you to check the radius on a comparator, just push up into the wheel on the corner until you get enough to check. As for what works best? I have used 36 grit wheels with a soft bond all the way up to 150 grit wheels(the Norton orange beasts) Just drop it on the floor and break it up into 3-4 chunks easy to handle.

 

[ZEPMaine]

Sorry to revive an old thread but I happened to be searching for input on a vitrified diamond wheel to improve our current process where we dress a .050" corner radius on a 150 grit resin bond diamond wheel.

We use a 1" wide 20" diameter wheel in a cnc cylindrical grinder. We mount a rotary dresser with a diamond disc so that the spindle is vertical and we dress across the diameter and the around the radius.

We grind a 1" +/-.010" radius in a WC coating. We use that .050" wheel radius to interpolate the 1" radius. We can grind probably about 10 parts without having to redress the radius. The problem is, as others have suggested, it will give you the right form but it is dull AF. After dressing the form on the radius then we dress the diameter traditionally with a SiC wheel and we stick the hell out of the radius. It works well enough, but I'm interested in using a vitrified bond diamond to make this process easier. This is also done on a aerospace part that requires compliance with AMS2449. That specification doesn't explicitly prohibit dressing resin bonded diamond with a diamond disc, but I've run into someone recently who warned of potential compliance issues dressing that way with this specification.

 

[Milling man]

This is also done on a aerospace part that requires compliance with AMS2449. That specification doesn't explicitly prohibit dressing resin bonded diamond with a diamond disc, but I've run into someone recently who warned of potential compliance issues dressing that way with this specification.

Do you know what this requirement is related to? What could go wrong?

 

[ZEPMaine]

Do you know what this requirement is related to? What could go wrong?

It's an aerospace material supplier spec that covers the grinding of tungsten carbide coatings. After thorough review and thinking on it for some time I really don't think we will have any issues. The bigger thing I just haven't wanted to miss anything.

These parts are cursed. The manufacturing takes over 2 years per part currently and the vendor that we are required to use for carbide coating is incompetent. For about a year they were trying to blame their carbide failures on our grind process that had remained unchanged since 2014. But we keep finding voids, inclusions, and delaminations in their coatings. These issues require stripping and recoating. It keeps getting worse as well. Every single piece they send now requires rework. It's like a terrible Christmas present we grind to unwrap and hope there are no issues when we get to finish size. Planning on having to rework every piece on a soon to be flight safety part is ok right?

 

[Milling man]

and hope there are no issues when we get to finish size.

Have you thought about using some kind of non-destructive testing method such as X-ray or ultrasound? This task reminded me of a story I told a colleague about machining large parts for rockets. The plant receives an aluminum forging with a diameter of 150 inches, scans it using some method (I don’t remember exactly which one) and then moves the scanning result for a long time, combining it with the 3D model so that the “bad” places in the workpiece coincide with the part material being removed. Sometimes the forging goes back to the manufacturer

 

[ZEPMaine]

There really isn't any benefit to us inspecting the coating prior to grind because the company doing the coating doesn't have the means to remove the coating on their own. So even if we found something we would have to grind the coating to about .0005" radial thickness and send it back for chemical stripping.
They can't chemically strip the full coating thickness because the time in the tank would be risky to the 2 other coatings on the part and the other WC bands that are conforming. They also do not have a grinder large enough to strip grind these parts themselves.

 

[Milling man]

We use that .050" wheel radius to interpolate the 1" radius.

I have a question - why do you approximate the radius on the part with such a small radius on the grinding wheel? It seems to me that using a larger radius on the grinding wheel will increase the time from dressing to dressing.

 

[michiganbuck]

I have a spinning wheel dresser one a cracker jack mini , and the other I forgot the name. They were made mostly for AO wheels, but do work on diamond wheels.
Using a simple radius gauge and eyeball one can easily get a decent radius to +- .002"

I know this is true because I had to check my hand generated radiuses on the shadow graph.

 

[ZEPMaine]

I have a question - why do you approximate the radius on the part with such a small radius on the grinding wheel? It seems to me that using a larger radius on the grinding wheel will increase the time from dressing to dressing.

The wheel they bought was a type 1 straight wheel with 1/4 thick abrasive material. I think they went with .050" radius for wheel life, but I'm uncertain. I'm in the process of quoting vitrified diamond wheels now that should be easier to dress this way. If I want to grind all the carbide with a single wheel I'll be stuck with a .075" radius max but if I go with 2 different wheels I can easily get a stock wheel with a .250" radius. Currently this radius takes about 3.5 hours to grind and the wheel change takes about an hour so long as it runs out well enough where it doesn't require dressing. Parts are 5 setups with 4 different wheels and 3 different coatings. 10 hours setup and 40 hours run per piece. It makes inefficiencies easier to ignore on low volume parts. I wouldn't be surprised if there was no testing of alternate parameters when it was initially set up either.

 

[Milling man]

10 hours setup and 40 hours run per piece.

Hehe, I'm even a little jealous of such a complex and interesting job

 

[michiganbuck]

Sometimes it can be good to buy a plated wheel, that is a wheel made with the desired form. They don’t have as much abrasive thickness but are very true to desired form/shape.
Sidley Diamond wheels is one outfit.
For a radius one would order the radius on both corners of the wheel.