Predicting feature size with sinker EDM?

Hello all you EDM gurus,

I am a general noob when it comes to sinkers and EDM in general but I am trying to learn the art. We bought a used CNC sinker for our machine shop (it was cheap) in hopes of using it for the odd repair job or broken tap removal. No one here has EDM experience.

So I have been slowly working my way through the machine manuals (Heidenhain TNC 416 controller I believe) and have been able to figure out how to set up and run the machine without too much issue. But I am having issues figuring out how to nail the finished feature size. Before I go into more detail on that I should mention, I am not trying to hold tenths...if I can get +/- .001" I would be happy.

Say for instance I want to make a .400" square cavity, I choose the e-table that matches my electrode face surface area closest, which tells me I need to make the trode .010" small per side minimum. So I machine the trode to .376" and run the program expecting the cavity to be undersize by .004" (which I could then orbit to size) but the actual size never corresponds to my expectations. So I think I must be missing something.

My assumption is that I should be able to pick an e-table, pick the surface finish desired and run the cycle and the size should come out to the size of my trode + 2x the spark gap. But either I am reading my e-tables wrong or misunderstanding how the process works. Can someone run through how you would sink a simple cavity to size on your machine? I get that flushing has a huge effect on everything but I have a hard time imagining that is want is messing with me. I'm generally seeing variations in the area of .005" to .010" from expectation. So even something like sinking a keyway is not possible at this point.

Thanks in advance!

Levi

Sent from my SM-G965W using Tapatalk

Leviathan said:

My assumption is that I should be able to pick an e-table, pick the surface finish desired and run the cycle and the size should come out to the size of my trode + 2x the spark gap.

Sent from my SM-G965W using Tapatalk

Click to expand...

This is essentially correct, but you also need to account for electrode wear.

Are your sizes consistently undersized from what you are expecting?

To hold .001" you will want at least two electrodes. With your square shaped electrode example above you should be able to easily measure the wear (which should be significant) on the first electrode after roughing the pocket, this may give you some clues.

A blind cavity is also a worst case scenario for flushing, if you can use through flushing even just for a comparison it may be worthwhile.

Undersize typically yes, unless I try to compensate pre-emptively. Then it's over or under depending on what I guessed vs actual.

I did finally get the thru flushing working and that made a massive difference.

Regarding wear I actually have found that the electrode doesn't wear much on the sides when roughing (on the front corners it does for sure), but it does significantly when finishing. Is that normal? Or should I try tweaking the e-table to have a little longer on time?

Thanks for the help, I feel like the more I play with it the more I kind of understand what is going on...but I still feel like I lack enough understanding to make the predictive decisions required to have the results match my expectations consistently.

Levi

Sent from my SM-G965W using Tapatalk

Leviathan said:

Regarding wear I actually have found that the electrode doesn't wear much on the sides when roughing (on the front corners it does for sure), but it does significantly when finishing. Is that normal? Or should I try tweaking the e-table to have a little longer on time?

Click to expand...

Yeah that's normal for an electrode with vertical walls. Since the front face of the electrode is doing most of the work during roughing.

The way it works on our machine, the first part of the cycle uses the highest power setting (in your example equivalent to .010"/side overburn) and burns down in the Z direction with no orbit until it hits the programmed depth. This is the roughing stage. This is the quickest way to burn a cavity, but the higher power setting will leave a rougher surface finish and it would be impossible to have corners any sharper than ~.010" or so.

Once the electrode hits depth, it then drops to a lower power setting and uses the orbit to make up the difference. So for example it might use a power setting equivalent to .005"/side overburn, and orbit .005"/side

This helps crisp out the details in the electrode (the sharp corners in your example) and it's how the finer surface finishes are generated.

Depending on the combination of settings (undersize vs. surface finish vs. electrode material vs. workpiece material etc) the wear can be significant. Finer finish increases the wear.

Fwiw our machine came with charts that help to choose electrode undersizes for a given combination of factors and also to help predict electrode wear.


I would probably avoid tweaking the individual burn settings for now, but I don't know your machine and we do tend to tweak some settings to get the results we want on our machine.

You are adjusting the orbit size in your program correct? Not just picking power settings and assuming that's what holds your size?

Heidenhain is really a low-end EDM control but you should still be able to hold sizes no sweat.