What's new
What's new

Punch Lapping Machine

Zach201

Plastic
Joined
Mar 18, 2020
Location
Middle TN
Hi, guys. I am a grinder in a powdered metal tool and die shop. A few times a year we have to build some kind of O.D. splined form punch. So, our CNC mill man will rough the spline form on the mill, then heat treat, then it comes to me for finishing. Depending on the form, mostly on the base radii, I may either CNC jig grind the O.D., or dress the tooth gap form onto an aluminum oxide wheel and set up the punch on a 4th axis fixture on a manual surface grinder. Sometimes I may have a gage plate to test for die fit, sometimes I may have the actual die. But what usually ends up happening is, no matter how meticulous I am with measurements, the punch will bind in the die somewhere I know not where, and I end up LAPPING the punch into the die itself. This cause me varying amounts of grief, depending on the size, (mostly weight), of the punch. I have done this several different ways, the last couple big ones I have done, I have actually set the die up in our big lathe chuck, and engineered some way to attatch the punch to the tool post, using the lathe the hold things steady while moving the Z in and out and in and out and in and out until it fits. Very short version, it's a GREAT BIG PAIN. I'm guessing a lot of folks would just EDM sink the punch with a wire cut electrode. Any other powdered metal guys out there without a good EDM man do things like this? I have about 7.5 yrs experience in powdered metal stuff. First, in a production factory running CNC lathes in secondaries, now as a tool and die man for the past 4.5 yrs. So, not an old hand at this. Is there any such thing as a purpose built machine that I could attatch both punch and die to that would do this for me automatic-like? I'll take whatever advice y'all may have, Thanks.
 
No help here but it sounds like some pretty fancy work.
Guess I would say with lapping it in have a line up mark so all stays on the same mark at running.
Old hat this so likely you already have this under wraps.
 
Thanks for the reply Buck. I just signed up on Practical Machinist. Started a small shop in my garage and figured I'd better tap into the wealth of knowledge on this site. I have some limited experience with PLC's, built a can crusher a few years ago that ran on one, here. Was thinking I might be able to design a machine to do the lapping. Just didn't want to invest time, energy, and money, into inventing something that already existed.
 
How large/small are your clearances? Yeah I would guess that lapping can be a big PITA sometimes. Standard spline/base radii shapes? Are you checking the die for print accuracy from the WEDM before trying to make the punch? How's your measuring equipment/methods?

On a personal note, where(approximately) are you located?
 
When you lap from one side, are you at risk on inducing a taper in the punch and die set? Are the particles being compacted big enough that they couldn't migrate into the clearance when the punch is at the top of the die, and start binding as the die is inserted?

What I'm getting at is that when you make your punches, perhaps they should be ~1" longer than needed, so you could cut off the extra bit and use it as the lap itself, allowing you to stroke longer or shorter or withing certain boundaries of the die to prevent taper from showing up.
 
What material? What clearance? Have you considered leaving the punch soft and shearing it into the die, then hardening?
 
Alright, I'm going to try to get everybody with this one reply. Typical clearance between punch and die will run .001 to .0025 or so per side. Measuring equipment . . . eh . . . We have a manual CMM, I've never used it, not sure if it would help. I could check mic under wire and minor I.D. dim's on die with pins and joe blocks or bore gauges. But, with the clearances already noted, cutting my punch to print size should be sufficient. Lapping from one side of die is fine, as most of the punches will be dedicated upper or lower punches. Not sure what sift of powder is used on the various parts. Making punches longer and cutting off is no good, because I'm not actually trying to make the die bigger, I'm just making the punch fit it. On Taper, a lot of the dies will have taper added to the top on purpose to assist in ejecting the compacted green part. Die insert mat's usually CPM 10v, D2, Carbide. Punch mat's A2, CPM 3v or 9v. Shearing: Punch major O.D.'s might be 5 or 6 inches, and spline form, that's a lot of linear inches of CPM to break off.
Basically guys, I think what I'm doing up to the point of lapping is sound, although I know I'm not perfect, and always trying to improve, so fine with the ideas. Like I said, judging from tooling I've seen from other companies, most folks nowadays with a couple EDM's would probably just sink it and be done with it. I'm just wondering if there was a machine that did what I am doing automatically.
 
Welp, that seems to cover it. Just make a version of your can crusher with a variable stroke crank and a VFD drive to a gear motor, along with some linear ways and clamping structure to hold and guide. Throw in an aquarium pump and some abrasive slurry and you're set.
 
Shearing: Punch major O.D.'s might be 5 or 6 inches, and spline form, that's a lot of linear inches of CPM to break off.

Its a lot of form. But you are getting it very close on the grinder. You are probably only a few tenths to a few thou in a few places. I wasn't considering shearing the whole profile, although I've seen it done.
 
But, with the clearances already noted, cutting my punch to print size should be sufficient.

-Yes it should be sufficient but apparently it's NOT sufficient if it doesn't fit. Is it possible that there's something as simple as a mismatch in the radii? Have you checked for clearance with a die light on the surface plate? A comparison of the two drawings and full profile analysis with a CAD program would be the first step I'd take after checking clearance with a die light, this is SOP for ANY new punch/die made for clearance and alignment in mounted location as well. Takes too much time? How much time is being wasted by lapping? Not knocking your solution at all. My solution would be to not have to do that more than once by finding what is wrong or not according to print. If both drawings show clearance all around the profile then somebody isn't producing what the print calls for. It may be a lot of area but .001-.0025 PS should be a drop-in. If no-fit has happened more than 3 times I'd also request the WEDM to cut a template of the punch out of some metal plate for verification of geometry. Fitment/dimensions can be checked and verified in the template. How you measure the punch is how you'll measure the template. Solutions to problems call for stepping outside what is already being done if you want results. If this is too much trouble then you're welcome to continue lapping all you want. Not intending to sound harsh or critical, hope this helps.
 
-Yes it should be sufficient but apparently it's NOT sufficient if it doesn't fit. Is it possible that there's something as simple as a mismatch in the radii? Have you checked for clearance with a die light on the surface plate? A comparison of the two drawings and full profile analysis with a CAD program would be the first step I'd take after checking clearance with a die light, this is SOP for ANY new punch/die made for clearance and alignment in mounted location as well. Takes too much time? How much time is being wasted by lapping? Not knocking your solution at all. My solution would be to not have to do that more than once by finding what is wrong or not according to print. If both drawings show clearance all around the profile then somebody isn't producing what the print calls for. It may be a lot of area but .001-.0025 PS should be a drop-in. If no-fit has happened more than 3 times I'd also request the WEDM to cut a template of the punch out of some metal plate for verification of geometry. Fitment/dimensions can be checked and verified in the template. How you measure the punch is how you'll measure the template. Solutions to problems call for stepping outside what is already being done if you want results. If this is too much trouble then you're welcome to continue lapping all you want. Not intending to sound harsh or critical, hope this helps.

AD - I could go through each of your thoughts individually, but the short way is - I appreciate and agree with everything you've said. I don't take it as critical at all. I think ideally, our client could require every spline form die to be wire cut with a .250 piece of flat stock clamped to the top, so that there is a gage made at the same time as the die, and to the exact dim's. Then when they sent out the punch job, they could send the gage plate along with it. That would solve a lot of problems on my end. When we make the dies, we do usually cut gages, for future punch builds. As I said in my first post, I usually end up lapping, not always, sometimes I do have punches that fit right it. If I'm lapping one, it's because that seems to be the best way to finish the part. Example: Last one I did, base radius was about .081, I cut it with a .157 CBN wheel. Took 2.5 hrs to make one round on the jig grinder. Once it was off the machine, figured with set-up and cycle time, it would be easier to just lap it than to set it back up. Lapping times may range from a couple hours to a day or so (working at it in spells, not constantly), depending on where the problem is and how much material is still there. Also, as far as doing this more than once, I've probably only made more than one of the same punch a few times, and I do learn the problem areas of them. Mostly, I see stuff I've never done before, though, so it's always an adventure.
I also wanted to add, that when talking about fitment, in these situations, the punch will START into the die. If I can't get it to go in a half-inch or so, I know something is still way off, and I don't even attempt lapping a part like that. But if it will start in at least that far, I know I'm in lapping territory, because it can't be too far off.
 
Zach- Glad to hear you're not taking this as a personal attack, never was meant as one. I will say that I may have misunderstood all the dynamics of the situation. If the punch is way smaller than I imagined (you did mention weight), that makes mapping more difficult. Not being proficient with the CMM doesn't help, I'm not either with most of them. Not repeating the job makes every one a NEW ADVENTURE (not in a good way either). I'd agree that a 2nd set up in the jig grinder is very time consuming, set up always is. Your choice to rub the problem areas out does have it's own merits, especially if the boss walks by and wonders what's taking so long ("Here, see for yourself"). I will comment on the punch starting into the section but meeting interference. Sounds to me like there's taper to the die land from the WEDM. I'm not familiar enough with why this is so (too fast a feedrate, no finish skim cut, poor burn settings) on a WEDM, perhaps some of the other members can comment. A quick check with a DTI on the die section can reveal bell mouth or reverse taper. If radii are the trouble then it can still show by comparative (not quantitative) values with a DTI. If this is true then your company isn't getting what was paid for. If you're ok with it then that's the end of anything I'm willing to suggest. Lapping may not create the correct clearance required but that's different topic. I would like to know in a PM who's doing your wire work so I don't use them or at least verify any work from them. Ganbatte.
 








 
Back
Top