Grinding a groove - need advise on profiling surface grinder wheel with small radius

jj80909 · Aug 7, 2018

I posted a similar question in the CNC forum and it was suggest that I try it here.

I'm looking for some advice on how to machine some small 17-7 PH stainless parts. The part is made out of 0.040" think sheet with a width of 0.375 and the length varies from 1.500" to 4.000". This is a small run of 50 to 100 parts, but will probably be a regular job. The parts can be machined prior or post hardening.

I need to cut a 90° groove in the tip that is basically 0.020" deep, but there is a small radius at the root with a tight tolerance and the surface finish of the groove needs to be an Ra of 0.8 or better. One of the parts has the groove located 45 degrees to the top surface and the other is at 70°.

It was suggested that I use a profile wheel on a surface grinder, but I have limited skills / knowledge when it comes to grinding. I've basically only used the surface grinder to make parts flat. It was also suggested to rough-in the groove with the VMC and finish grind it.

Is this something a standard grinding wheel / fixture could handle or a profiled wheel? How long should I expect the wheel to last between dressing? How does one go about setting this up?

Any suggestions are appreciated.

Toms Wheels · Aug 7, 2018

There a several ways to profile a wheel. What really appear to need is a wheel profiled to grind the Vee and leave a radius. Since it is not that large number of parts a dresser like the J&S could do repeatable profiles.

Although they are out of business the website was still working last I went there couple years ago. The instructions were posted there.
You can set an angle R&L, and a radius. Contact me for details if you get one of these, I will help you through the setup.

There are other types of profile dressers, but I know little about them.

Another way is to

sfriedberg · Aug 7, 2018

There are several relatively easy-to-setup-and-use wheel dressers that will put a tangent-radius-tangent profile on your wheel.

The J&S mentioned above is a J&S Fluidmotion dresser, which holds the diamond in a vertical column and moves it in a horizontal plane. There are also dressers which move the diamond in a vertical plane, which makes them much easier to use with wheel guards installed. These dressers (sorry, don't have a brand+model on the tip of my tongue) somewhat resemble an elongated punch grinder with extra knobs on the back, holding the diamond in a horizontal column.

Sounds like a tangent-radius-tangent dresser is all you need for this job. If you need a more complex profile, there are template-driven reducing pantograph dressers (e.g., DiaForm) where you cut out the desired profile at 10X or 5X the final size and trace it with a stylus.

For production, I'm pretty sure all this has been replaced with CNC grinders with CNC profile dressing. But the manual dressers are still really handy for non-production jobs.

swarfless · Aug 7, 2018

Hi, not what you asked but I wonder if coining the groove would be a better/easier option, cleaning up the resultant bulges by simple surface grinding in bunches .. and/or before parting off the strips. 2c

Milland · Aug 7, 2018

swarfless said:
Hi, not what you asked but I wonder if coining the groove would be a better/easier option, cleaning up the resultant bulges by simple surface grinding in bunches .. and/or before parting off the strips. 2c

Interesting idea, but it would have to be done soft, and it's almost certain that residual stress would warp the part during heat treat.

CarbideBob · Aug 8, 2018

Well certainly rough it somehow.
A bigger problem is the rad size. The 4-6 is not so bad and is more open, the 1-3 will be difficult.

The general rule for a rad is that you want 3 grains inside the radius size. This is sort of the min to actually form a radius.
Yes there are tricks in white wheels to violate this but they don't hold up very long and are fussy to say the least.
Plated wheel people (cbn in this case) won't touch a .001-.003 as the plating thickness buries the grain size needed.
To get an actual controlled .004-.006 I'm in 400-600 grit wheels. .001-.003 I'm into micron graded wheels.

How will you (or more importantly the customer) check it? The 4-6 needs at least 100X optical, the 1-3 300-500X.

On a form cutting tool in HSS or carbide sometimes such rads are sometimes needed.
On a normal part it's often a problem of bad design and easy to draw in a computer.
In the real world you see such prints and the real world parts don't check as good under correct checking.
For this reason when the numbers get down there I'll usually ask to see a sample part, even a used one.
Many times you find what works or is being bought is not what the print says. Hence the sample request.
I know people with 20X comparators saying they make .003 corner rads. :wall:

.........They are not.
On the tight one you are kinda in no-mans land. This is past fine grinding and into art.

I really am :willy_nilly:

when I see such on prints. In the computer and blown up on the screen it seems like no big deal.
Holding 50 parts even though short may turn out to be ................
Both of these parts certainly can be done in a standard surface grinder in decent condition with a good dress and the right wheel.

Bob
(hoping others will disagree and have good or easy ideas as I don't have shit here to help the OP)

athack · Aug 8, 2018

I think if I had to do this job I would mill the slot then use a diamond plated form wheel to finish the groove.

Richard King · Aug 8, 2018

How about using a creep feed crush grinder ? The carbide dresser (s) would be expensive, but the whole part would be. I see another huge and expensive problem would be holding the part down on a plate to grind it. Stainless is (or most) non magnetic, so you would have to make a fixture using mitee-bites or similar products. You would have to machine multiple pockets first to set them in and use a flat head screw and super thin bite.

Kopal(R) Mini Clamps | Mitee-Bite Products LLC.

Form crush dressing of diamond grinding wheels - ScienceDirect

Crush Grinding(Technical Report) | OSTI.GOV

Get a contract before investing any money as it will be super expensive. You may want to check around for a company with the grinders and sub it out. Crush grinders like a Slicer Dicer machine are $$$$.

I'm no expert in this but had to have some parts crush ground several years ago. I have rebuilt Do-All slicer dicers and Magerlee Crush grinders. As I am hitting enter. I wonder if they could be EDM'ed some how? Rich

BSCustoms · Aug 8, 2018

Sort of on topic, what is the ideal way to get the location of the slot after the wheel has been dressed? just cut, check and move?

Paolo_MD · Aug 8, 2018

BSCustoms said:
Sort of on topic, what is the ideal way to get the location of the slot after the wheel has been dressed? just cut, check and move?

Given that the groove is 90°, my ignorant way of doing it would be to build fixtures holding the pieces at 20° and 45° with positive stops for the parts (perhaps, it's worth ganging several parts on the fixture), use a cup wheel and the only things I'd have to adjust is compensating for dressing the side of the wheel (which is no issue if you leave the diamond mounted on the table in line with the edge of the groove).

Paolo

Richard King · Aug 8, 2018

I wonder how long a 90 degree sharp edge would last? Paolo...seems like a clever idea, but I would think consistent set up, dressing on a white stone would be difficult. As Bod mentioned grit size staying sharp on such a 90 deg. edge. Maybe ask this question in EDM forum too?

CarbideBob · Aug 9, 2018

Nominal rad size on the 1st part is .002.
Neither a plated or crush dress wheel can do this and I really, really doubt the ability of a Diaform to do it.
All of these sort of hit a wall at .004 min and that does not stay there very long.

It can be done in the 1200-1600 grit range resin bond wheels but your wheel dresser needs to be top notch solid and true. Stock removal is more like polishing than grinding in these grits.
You would simply dress both sides as straight. The min rad will be there automatically, no radius dress is needed.
Same dress on print number two but depending on wheel used you may hand hit the rad area with a very soft stick oh-so gently in areas. (artsy-fartsy)

The part feature is only 3/8 long, that is a plus.

Another weird thing is that I'd much rather do this grind in carbide or even a CBN strip than HSS or worse yet stainless.
At tiny rads the hard and superhard materials are actually much more user friendly.

To the OP directly: Be sure to let us all know how this all plays out. If your find a secret go fast method you should share.
Bob

cash · Aug 9, 2018

CarbideBob said:
Well certainly rough it somehow.
A bigger problem is the rad size. The 4-6 is not so bad and is more open, the 1-3 will be difficult.

The general rule for a rad is that you want 3 grains inside the radius size. This is sort of the min to actually form a radius.
Yes there are tricks in white wheels to violate this but they don't hold up very long and are fussy to say the least.
Plated wheel people (cbn in this case) won't touch a .001-.003 as the plating thickness buries the grain size needed.
To get an actual controlled .004-.006 I'm in 400-600 grit wheels. .001-.003 I'm into micron graded wheels.

How will you (or more importantly the customer) check it? The 4-6 needs at least 100X optical, the 1-3 300-500X.

On a form cutting tool in HSS or carbide sometimes such rads are sometimes needed.
On a normal part it's often a problem of bad design and easy to draw in a computer.
In the real world you see such prints and the real world parts don't check as good under correct checking.
For this reason when the numbers get down there I'll usually ask to see a sample part, even a used one.
Many times you find what works or is being bought is not what the print says. Hence the sample request.
I know people with 20X comparators saying they make .003 corner rads. .........They are not.
On the tight one you are kinda in no-mans land. This is past fine grinding and into art.

I really am when I see such on prints. In the computer and blown up on the screen it seems like no big deal.
Holding 50 parts even though short may turn out to be ................
Both of these parts certainly can be done in a standard surface grinder in decent condition with a good dress and the right wheel.

Bob
(hoping others will disagree and have good or easy ideas as I don't have shit here to help the OP)

Exactly what I was thinking on all points. Grain size vs radius is your biggest issue.

Richard King · Aug 9, 2018

How would you two hold it down?(Cash and Bob)

Conrad Hoffman · Aug 9, 2018

Seems like an EDM project with really fine wire. Theory says it can be ground, but converting theory to practice will be difficult and time consuming. How much money is somebody willing to pay for these?

uncledenny · Aug 9, 2018

Profile grinder. No worries.

gusmadison · Aug 9, 2018

I realize this whole process has to be planned one step at a time and I'm enjoying the brainstorming. That being said, does the surface finish Ra call out concern anyone as much as me? Seems like were in the territory of needing an optical profilometer maybe?

Those aren't cheap and need their own measurement recipe to be accurate for the individual part being checked? Am I wrong on that?

Just wondering how the OP will QC inspect surface finish, let alone the customer.

Like others on here, for the love of all things right, if you take on this job, PLEASE resurrect this thread and share with us the procedure, trials and tribulations of how it all went down!

Best of luck Sir!
Gus

thermite · Aug 9, 2018

cash said:
Exactly what I was thinking on all points. Grain size vs radius is your biggest issue.

Dunno where the technology stands NOW, but required consistency may lead to some form of "laser" gadgetry?

Right about fifty years since they've been able to "drill" clear into a Diamond to vaporize inclusions, raising the commercial value. Bores left behind - sometimes several per gemstone - were really hard to see, even with a loupe.

I think the "controllability" exists, and there's no tool wear.

AD Design · Aug 11, 2018

We'll leave possible poor design of a sheet metal part aside for the moment. I've done a fair bit of work on the surface grinder and I'd really like to know how the customer intends to verify a .001-.003 radius on the part. It is likely possible to do this consistently but when this drives the part cost through the roof shouldn't the customer be made aware of what they're asking? As Bob has pointed out it's quite easy to design something like this in CAD, the disconnection between on-screen capability and shop floor reality can easily cause an inexperienced designer to ask for absurd geometry/tolerances rather than seek another solution that's more practical. I went from the shop floor to engineering and some of the things like this I've seen on drawings over the years have made me request a private conversation with the engineer/designer. Perhaps this is in order with a revised quote in hand. Would really like to know the outcome for this project.

Robert R · Aug 11, 2018

The part might be produced by a coining operation. This is a cold forging process with closed dies. If the number of parts needed were in the thousands it could be produced by running a long strip through a set of forming rollers and then cut to length. This is a question that PM member Kevin Potter can answer.

Robert

Grinding a groove - need advise on profiling surface grinder wheel with small radius

Aluminum

Attachments

Titanium

Diamond

Cast Iron

Diamond

Diamond

Stainless

Diamond

Cast Iron

Stainless

Diamond

Diamond

Titanium

Diamond

Diamond

Aluminum

Cast Iron

Diamond

Stainless

Hot Rolled

Similar threads