What's new
What's new

Wire EDM Concentricity

wd8rmo

Plastic
Joined
Jan 16, 2019
Hello,

I'm running a Mitsubishi FA20S. My issue is being able to hold a .0005" tolerance in the concentricity of the part.

My material is very stable (DC-53 with 3 draws). I'm cutting a gear style part out of a 3.000" thick block. I'm having problems holding the concentricity between the O.D. and I.D.

In addition to the part there are 10 smaller I.D. holes being cut out of the ring between the ID and OD. My process is to rough cut everything first starting with ID cuts threw the OD cuts, then skim cut everything the same way.

My inspection shows that I fall out of tolerance for concentricity in the direction of the inner ID holes in the ring. The ID and OD cut sizes are correct, just concentricity is out.

Some suggestions have been to offset my Geometry in the direction of the problem. Feel I shouldn't have to do that, do to, sometimes running different thicknesses and the amount of out-of-tolerance changes.

Just had "X" Ball screw repaired(not new, serviced by ball screw company)("Y" no problems) and a full Preventive Maintenance completed.


Looking for advice; Machine ware - Mechanical check or replace - Software - Parameters - ???


Bruce
 
What machine? I was told some years ago that our Charmilles (290 & 310 models 1996'ish) were only good for +/-.0005" from a service tech. He was FOS IMO because I routinely did dowel holes and such to a couple tenths*, but anyways....

Could be the machine is just not capable?

* It did require (especially on thicker stock) several skim cuts and rotating the hole (entry/exit points essentially) 90deg between skim cuts.

edit: just noticed machine make/model, my bad :o
Don't know anything about them
 
Bad wire alignment would result in a tapered part/hole. OP is not stating that (but maybe has not been checked?)...
 
Good morning All:
Some of you are suspecting wire alignment as the root cause of a concentricity problem between features cut using a single setup for both features.
I don't see how that can be the root cause.

Certainly if the wire is tilted relative to the table, the features will not be correct, but they will all be leaning the same way, so something else must be going on.
One would think that if you tilted the job to make the wire cut features vertical and then surface ground the top and bottom of the block, the bores and outside features would not be round, they'd be elliptical.
But they still would be concentric to one another.

The most obvious culprit to my mind is always material movement; every time I have a problem like this, that's where I look first.

The second place I look is thermal drift...if the filters are clogged, if a pump is failing, if the chiller filters or core are blocked the contents of the work tank will grow as the water warms up and weird position errors will begin to show themselves.

The third place is a mechanical problem in the ballscrews, but typically this manifests as geometry errors within a single feature...round features aren't round anymore etc etc.

I try to diagnose those with simple test cuts, and my favourite is an octagon bore just over 1.2" across the flats in 2" thick piece of hardened D2 because it's easy to interrogate just with gauge blocks, a tenths clock, and a sine plate and it challenges all combinations of axis movement.
If you can make a perfect octagon bore, it isn't the ballscrews or the control.
If you can't, the errors will give you a hint as to what is wrong with the motion control system.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
I agree, there are many different things that can contribute to the concentricity being out. I always start with wire alignment as it's an easy thing to check.

Work holding is up there in the list but there's also wire tension, wire speed, and flushing. Epack settings, part geometry, material, and machine condition all can contribute. I start with wire alignment because it takes 5 minutes lol.
 
Hello,

I'm running a Mitsubishi FA20S. My issue is being able to hold a .0005" tolerance in the concentricity of the part.

My material is very stable (DC-53 with 3 draws). I'm cutting a gear style part out of a 3.000" thick block. I'm having problems holding the concentricity between the O.D. and I.D.

In addition to the part there are 10 smaller I.D. holes being cut out of the ring between the ID and OD. My process is to rough cut everything first starting with ID cuts threw the OD cuts, then skim cut everything the same way.

My inspection shows that I fall out of tolerance for concentricity in the direction of the inner ID holes in the ring. The ID and OD cut sizes are correct, just concentricity is out.

Some suggestions have been to offset my Geometry in the direction of the problem. Feel I shouldn't have to do that, do to, sometimes running different thicknesses and the amount of out-of-tolerance changes.

Just had "X" Ball screw repaired(not new, serviced by ball screw company)("Y" no problems) and a full Preventive Maintenance completed.


Looking for advice; Machine ware - Mechanical check or replace - Software - Parameters - ???


Bruce

Are you dropping any slugs off of the part? It doesn't take much movement to loose .0005 tolerance on concentricity. Maybe switch your program around so you could stop and double check your zero before finishing and verify nothing has moved. If you don't see any taper and your perpendicular to the top and bottom surfaces, plus you know the part isn't moving during machining, then I would be worried the machine itself is not moving properly.
 
Next question:
Are you SURE your rough cuts are completely cleaning up with your skim cuts?
Look for subtle differences in finish as you inspect around the bore.

Also it pays to listen to your skim cuts if you have the patience to hang around in front of the wire that long.
If they sound louder in some places than others, you know the part has moved.
Try a touch off right after your rough cut and see if the origin of the bore is still where it's supposed to be

Check it also for roundness.
Obviously check the outside of the part the same way.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
Hello to all, and Marcus

Thanks for the responses, Wire alignment is always done before this part is started (didn't think to add that).

I run these machines (Mitsubishi FA20S) in an environmental controlled room at 66 degrees. My water temp. does go up, sometimes 1.0 degrees during roughing. But for the most part is stable. I had a concentricity test done and it was determined that the ball screw was out of spec. by .003" and was also effecting my glass scales to be constantly making adjustments, which looked like the machine was vibrating while doing a linear move. Also, had the machines double checked for leveling and any issues with that were corrected too.

New resin bottle and filters were also replaced, just to insure that was not going to be a factor. Full PM. Performed a test cut and no taper and was to size.

All slugs are pulled out and not dropped in on lower head or in the tank.

The roundness of the ID and OD are round and to size. I believe that my flushing is not perfect (lower flushing slightly harder then the upper), I have a single pump for that. Not sure how to adjust them it that would help. I have noticed that after rough the top of the bore looks darker then the hole length of the bore. I cleans up and looks even, after the finial skim.

Thank you all for your input, look forward to see more.


Bruce
 
Hi Bruce:
How did that ballscrew get so far out of spec?
That's a bloody mile in WEDM terms, so did the machine take a crunch at some point in its past?
I find it hard to believe you could wear the screw that much; wire EDM loads are not even remotely like VMC loads and a 0.003" error in a VMC represents a totally clapped out high miler machine.

Was it actually the ballscrew or was it perhaps a thrust collar that got loose?

In any event, it sounds like you've been covering all the simple bases very thoroughly and it's clear you know what you're doing.

To go back to my point in the first of my posts, if you can make a perfectly round bore without issue, then it's not the ballscrews it's got to be something else.

Who's measuring the error, and are YOU convinced it is it actually there?
Seems a flippant question but...you know...stranger things have happened.
If this thing is getting interrogated on the whizz bang CMM...maybe, just maybe it's the CMM that's the problem.
Find a way to center up and spin an out of spec part on a good quality spindle and check it with a tenths clock...don't count on it but you may be surprised!

I have seen it happen more than once, where a part checks good on the fancy gadget but the simple gauge pin won't go in.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
Marcus,

I feel very confident that you and I are on the same page of looking for the issue. I'm checking the part on a Makino 56i and spinning a .0001" indicator around it. Picking up the center ID and checking the concentricity from that point to the OD.

My machine is 12 years old and I cannot speak for the first 6 years of it machining life. I have not had any major collisions or catastrophes in the last 6 years. It's possible the thrust bearings contributed to the out-of-tolerance factor. I had replaced those as well.

At first I thought it might be the ball screws and thrust bearings, but am convinced that it was not.

Thanks for any and all your thought on this.

Leaving for the day, but will keep viewing any responses.


Bruce
 
I'm leaving two glue stops, one on each opposing ends of the OD.

I run this machine, for the most part. 24 hrs. for 6 days a week.

In addition to checking on another machine, this concentricity issue shows up in the part. Don't remember if I mentioned this on why parts are critical on tolerance, they're used in "Fine Blanking". If not familiar with this, it's high tolerance stamping.

This sounds silly, but, I have cut part both Clock Wise and Counter Clock Wise and got the same results.

One of my thoughts is, do to the holes in the inner ring being cut out in proper order, (direction of concentricity shift) may be causing some or all of the movement in the material. Even though the material is as stable as I can make it, it may be moving when the part is out of the Wire EDM environment. I'm open to process change if it makes sense.
 
According to research this material is supposed to have a low movement characteristic. How low is to be determined. If concentricity is a critical component I would cut the bore small and make a fixture to do finishing passes, just to be sure. Shouldn't be hard at all to whip up a quick one.
 
Late to the party, but have you found the source of the issue? If not I would suggest roughing the OD feature first... That's where any potential movement is likely to happen.

Also have you tried cutting a simple ID & OD test part (something easy to check) to verify your machine's positioning accuracy?
 
Concentricity Issues

Late to the party, but have you found the source of the issue? If not I would suggest roughing the OD feature first... That's where any potential movement is likely to happen.

Also have you tried cutting a simple ID & OD test part (something easy to check) to verify your machine's positioning accuracy?


Thank you all. The real issue found was, do to the holes on one half of the part, it threw off the OD concentricity of the part. So, off setting the ID hole by .0003" got it within tolerance.


Thanks.
 








 
Back
Top