What's new
What's new

Looking for information on machine selection for an EDM job

NJM

Plastic
Joined
Mar 30, 2016
Hello,

I have a job right now that will require EDMing a rectangle out of a large box. It is all the way through the part. The material is 316L, the box is 4.9” tall, 7.28” wide, and 13.32” long. The pocket to be cut is 5.118” wide and 11.49” long, corner radius of 0.098”. I need the walls of the pocket to be square to 0.00039” to the bottom of the part and to the opposite wall. They also need to have an 8 micro inch surface finish.

We don't really have any EDM experience so I have been getting quotes out of house but I am not confident that someone can hit the tolerances I need. I also need a quick turn around or we are going to be screwed for the delivery date of the project. So far this project has had several delays due to people not being able to deliver on time or on spec. I am thinking about trying to convince the boss to purchase a machine so we can do the EDMing in house.

I gather that the surface finish is going be difficult to hit but it is only needed in a certain wall area so I could polish locally. My understanding is also that wire EDM is very accurate so the tolerances wouldn't be hard to hit as long as you take your time and do some roughing cuts and clean up cuts to work towards the tolerance. The width tolerance on the pocket is pretty loose, +- 0.005", but it needs to be VERY parrallel to the bottom wall and opposite wall, 0.00039".

Is this something I can do with an affordable used machine like a Fanuc Alpha 1B? Or do I need the accuracy of something like a Mitsubishi SX-20? Or do I need to spend $150K on a brand new wiz bang machine??

What other parameters go into producing a part like this?
 
Hi NJM:
There is an awful lot about the way you've described this job that's ringing alarm bells in my head.
First, I don't have a clear understanding of what you're trying to cut and just which features need to be very accurate so a picture would be a great help.
Second, I don't know how the "box" was made;
Is it fabricated from sheet metal?
Is it machined from a solid block?
Third, it's unclear how accurate the reference surfaces are, that you must maintain parallelism to.
Are they milled? Surface ground? Lapped?

Having said all that, there is a second problem I forsee.
You state you have no experience with EDM, you have a super rush job,that needs to be super accurate, you don't trust the local EDM specialists in your area to do it right , but you are considering buying an "affordable used machine" to do this having never done it before??
I have some difficulty with this...I don't think your plan is going to be successful.

All of what you state as your requirements are achievable assuming the part doesn't move from stress relief when you cut into it.
If you get lucky with that part, or you do your overall job planning to allow yourself some room to adjust the part to meet your tolerances after the pocket is cut, there's nothing else about your needs that's wildly unreasonable.
But you MUST remember, the movement of the part when you chop a big chunk out of it is impossible to predict or to mitigate.
The very hardest thing in wire EDM is to properly align wire cut features to existing geometry; there are constant surprises when you pick up a part, rough cut the features and have to re-adjust your origin and orientation to accommodate part movement during the roughing.

So find somebody willing to take on this project, be clear about what you need, work with them, TRUST THEM (they're not all stupid or lazy or uncaring) and pay them what they demand.
I would not take this on as anything other than a cost-plus job mostly because of the risk, the time pressure, and the screwaround factor.
Don't go out and buy an "affordable machine"...your chances of success are quite low in my opinion because there's far more to it than you think.
Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
www.vancouverwireedm.com
 
Hi NJM:
There is an awful lot about the way you've described this job that's ringing alarm bells in my head.
First, I don't have a clear understanding of what you're trying to cut and just which features need to be very accurate so a picture would be a great help.
Second, I don't know how the "box" was made;
Is it fabricated from sheet metal?
Is it machined from a solid block?
Third, it's unclear how accurate the reference surfaces are, that you must maintain parallelism to.
Are they milled? Surface ground? Lapped?

Having said all that, there is a second problem I forsee.
You state you have no experience with EDM, you have a super rush job,that needs to be super accurate, you don't trust the local EDM specialists in your area to do it right , but you are considering buying an "affordable used machine" to do this having never done it before??
I have some difficulty with this...I don't think your plan is going to be successful.

All of what you state as your requirements are achievable assuming the part doesn't move from stress relief when you cut into it.
If you get lucky with that part, or you do your overall job planning to allow yourself some room to adjust the part to meet your tolerances after the pocket is cut, there's nothing else about your needs that's wildly unreasonable.
But you MUST remember, the movement of the part when you chop a big chunk out of it is impossible to predict or to mitigate.
The very hardest thing in wire EDM is to properly align wire cut features to existing geometry; there are constant surprises when you pick up a part, rough cut the features and have to re-adjust your origin and orientation to accommodate part movement during the roughing.

So find somebody willing to take on this project, be clear about what you need, work with them, TRUST THEM (they're not all stupid or lazy or uncaring) and pay them what they demand.
I would not take this on as anything other than a cost-plus job mostly because of the risk, the time pressure, and the screwaround factor.
Don't go out and buy an "affordable machine"...your chances of success are quite low in my opinion because there's far more to it than you think.
Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
www.vancouverwireedm.com


Thank you for the great input. Unfortunately I cannot post pictures as the project is sensitive, however if you check your work email I actually contacted you early this morning looking for a quote on the project. If we can send you an NDA I will send the prints.

To answer some questions here, the raw stock is a solid block of 316L which will be machined and surface ground to the outside dimensions. The interior would be left solid to be removed via EDM. The reference surfaces are just the top and bottom of the box. They will be surface ground.

You are correct given my lack of experience and the time contstraints it is unlikely I will be able to do it myself without a hitch. This is why we are buying extra stock. Also just to put it out there solving problems like this on a daily basis is what I do. We bought a used surface grinder that allowed us to make a similar part (316L with pockets in the top and side which created stresses) for this project and it is flat to 3 microns. I have no surface grinding training other than 45 minutes on an old manual surface grinder. :). I'd rather learn the hard way and have a machine and a new capability for my business than to take the chance of losing this project because of someone elses screw up.
 
Thank you for the great input. Unfortunately I cannot post pictures as the project is sensitive, however if you check your work email I actually contacted you early this morning looking for a quote on the project. If we can send you an NDA I will send the prints.

To answer some questions here, the raw stock is a solid block of 316L which will be machined and surface ground to the outside dimensions. The interior would be left solid to be removed via EDM. The reference surfaces are just the top and bottom of the box. They will be surface ground.

You are correct given my lack of experience and the time contstraints it is unlikely I will be able to do it myself without a hitch. This is why we are buying extra stock. Also just to put it out there solving problems like this on a daily basis is what I do. We bought a used surface grinder that allowed us to make a similar part (316L with pockets in the top and side which created stresses) for this project and it is flat to 3 microns. I have no surface grinding training other than 45 minutes on an old manual surface grinder. :). I'd rather learn the hard way and have a machine and a new capability for my business than to take the chance of losing this project because of someone elses screw up.

Surface grinding the top and bottom only before EDMing the pocket will not help. What Implex took a great deal of time explaining to u is that as soon as you remove material, the internal stresses will cause part warpage/deformation. As soon as you EDM the pocket, there is no guarantee that ur perfectly surface ground top and bottom with be straight. So what you should be thinking is getting the part as close to net shape as you can before you do any of your right tolerances. Maybe sending it out for stress relieving also (I'm unsure if there is a process for stress relieving 316L).
 
Hi again NJM:
My machine is heavily booked wit a big production contract to the middle of August so unfortunately I'm not your man this time round.
However, there are a lot of really skilled and knowledgeable guys all over the USA, many with decades more wire EDM experience than I have.

Many of them post on PM.
You could choose any one of them and they'll do a brilliant job for you.
Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
www.vancouverwireedm.com
 
I think this is not so much a tough job as it is a fantasy level task. 163" square inches of finished wire cut, and probably double with stress relief cuts. I bet you'd be in the spark for 50-60 hours before it is all said and done. Your 3 micron flatness on the grind eats up one third the tolerance. I'd love to hear how this job was tackled when you get it completed. I'm sure it will be a fascinating journey.
 
Not to pile on the negativity, but I'm not sure you are even at a level where you understand the enormity of what you're asking for.

I think the most important thing to understand is that you need to use every piece of stock you have to deliver as many post heat treat, near-net-shape parts as you possibly can into the hands of the EDM shop who agrees to take this on, and trust them to get you to the finish line. You could probably try to do this on your own for weeks, months, or years, and just not be able to hit the numbers. Even if no other heat treat is called out, these absolutely need to be stress relieved after roughing, because depending on internal geometry, these could move a mile. I've seen parts like this move so much that they had to be started over - the original roughed part didn't leave enough stock to finish grind.

And I really hope for your sake that the parallel and perpendicular tolerances don't apply to the area that needs to be polished. Even the guys who do nothing other than high-end polish work for mold shops, etc., can still struggle to get a surface to clean up sometimes. And if it takes them a tenth or two more than you thought, your part is scrap. I've also seen pitting be a real issue with wire in 316L, making these tolerances and polishing requirements even tougher.

I'm with the other guys - I'd love to hear how this one goes. And if you find a shop that makes it seem easy, send me their contact info :-)
 
I'll chime in to say that of all the materials I've wired, 300 series is the worst for movement after the first pass. You will need that surface grinder for sure! Might as well just leave lots of grind stock before and after wiring.

An SX-20 is a maintenance nightmare...Maybe a FX-10...
 
Hey,
If you want EDM tools of any size or type Connect with us at Maxx Tooling to buy high-quality custom EDM Tooling and mact2.jpghinery solutions at a fraction of the cost charged by other suppliers in USA and Canada.
 








 
Back
Top