What's new
What's new

Deep pocket erosion problems

mazo

Plastic
Joined
Jun 2, 2009
Location
Slovenija
I operate Agie Innovation 3 die sinking machine and I have a problem eroding pockets (aprox. 25x55mm) for area inserts (into main insert).
Pockets are blind and 150-200mm deep and must be absolute square and in tolerance +0.01/+0.02mm.
I have problem with machine stopping becuse of short/contact and erosion times are huge (30hours/pocket). I am using Tokai Carbon graphite HK15
and 2 electrodes with undersize -0.15/side and -0.07/side (I want to have sharp edges).
What can I improve and many thanks for your support.
 
Flushing, Flushing, Flushing.


You didn't mention what you do for flushing.. I do similar work often, and whenever I have straight pockets to burn you MUST have flushing on ALL SIDES of the electrode. I would drill 3 holes on each side of the electrode spaced about 1.00" apart. A simple .030" hole would be great. This electrode should look like a sprinkler!

Your contact error is definitely happening due to buildup on the sides. Straight pockets are never fun, but they can be handled accordingly.

Also .006" OB is not enough. I would go .010" / Side OB
 
For those of us with edm experience but not with Agie Innovation 3, could you tell us what kind of sinker this is? CNC? Orbit capability? If this is a non-cnc, no orbit machine you are not likely to hold your sizes, let alone do it quickly.
 
Thank you for your replay.
I did not drill any holes in the electrode and I agreed that flushing is the main issue here. Does it matter where I drill the holes-I would assume more to the bottom of the pocket? How will these influence dimension of the pocket?

Agie Innovation 3 in full CNC EDM machine I can do orbiting, side erosion,.....
I think the machine can do it better, the problem is in front of the machine :-D
 
HI Mazo
What I can suggest is;
First of all, use suction for flushing. It will reduce the electrode wear. But you may need to use four electrodes. Two for roughing, means same size electrodes. just to make depth as much as possible erosion free.
Then use two finish electrodes. Current setting you may know better.
 
Hi Mazo:
I definitely agree with Zaki that suction flushing will be a huge help for keeping all the debris from arcing , shorting and taper cutting the pocket.
However, I'd use a different strategy for flushing.

I'd make a rougher that's substantially undersized (say 0.25 mm per side, and blast it into the job with high power, low frequency settings to minimize electrode wear and hack out most of the material quickly.
Put a single decent sized flushing hole into that trode; minimum 3mm diameter and 2 hole diameters off the center of the trode; use it to suction flush as Zaki recommends.
Put the hole at a 45 degree angle for the first 10 mm or so, then connect it with a vertical hole from the back of the electrode.
This is so you don't end up with a long spike at the bottom of the hole.

Once you've roughed to depth with it, dress off the end, index it 180 degrees and go down again to wipe out the residual spike, (that's why you put the hole in off center...the index will present an undrilled part of the trode face for your second burn). flatten the floor, pick out the corners and rough in the walls to within 25 microns or so.

Use a normal orbit for the roughing burn; if you get too crazy with your orbit, the suction flushing will not be even, and you'll have more problems with debris piling up on the electrode.

Your second trode needs no flushing holes; make it substantially smaller than the pocket and vector it into the corners; don't orbit it.
Direct the flow from a flushing wand down each face of the electrode, so make the trode long enough that you can point the wands where you need to to keep the faces washed clean.
If you jump flush and vector diagonally into each corner, you can pick the corners out easily with no problems at all because you have lots of room around the sides of the burn and you're not going very deep...only a few hundred microns (you're mostly burning SIDEWAYS now, not down in Z so the depth of the pocket no longer matters).

That's it; two electrodes and a fairly simple burning strategy, and you should be able to get your pockets as parallel as your electrodes were ground.
You can easily adjust for size and get them perfectly dimensioned too.

Keep your finishes as coarse as you can get away with, to make it easier to hold size and to minimize arcing, and you're good to go.
Cheers

Marcus
Implant Mechanix – Design & Innovation - home
Vancouver Wire EDM -- Wire EDM Machining
Home
 
Hi Mazo:
I definitely agree with Zaki that suction flushing will be a huge help for keeping all the debris from arcing , shorting and taper cutting the pocket.
However, I'd use a different strategy for flushing.

I'd make a rougher that's substantially undersized (say 0.25 mm per side, and blast it into the job with high power, low frequency settings to minimize electrode wear and hack out most of the material quickly.
Put a single decent sized flushing hole into that trode; minimum 3mm diameter and 2 hole diameters off the center of the trode; use it to suction flush as Zaki recommends.
Put the hole at a 45 degree angle for the first 10 mm or so, then connect it with a vertical hole from the back of the electrode.
This is so you don't end up with a long spike at the bottom of the hole.

Once you've roughed to depth with it, dress off the end, index it 180 degrees and go down again to wipe out the residual spike, (that's why you put the hole in off center...the index will present an undrilled part of the trode face for your second burn). flatten the floor, pick out the corners and rough in the walls to within 25 microns or so.

Use a normal orbit for the roughing burn; if you get too crazy with your orbit, the suction flushing will not be even, and you'll have more problems with debris piling up on the electrode.

Your second trode needs no flushing holes; make it substantially smaller than the pocket and vector it into the corners; don't orbit it.
Direct the flow from a flushing wand down each face of the electrode, so make the trode long enough that you can point the wands where you need to to keep the faces washed clean.
If you jump flush and vector diagonally into each corner, you can pick the corners out easily with no problems at all because you have lots of room around the sides of the burn and you're not going very deep...only a few hundred microns (you're mostly burning SIDEWAYS now, not down in Z so the depth of the pocket no longer matters).

That's it; two electrodes and a fairly simple burning strategy, and you should be able to get your pockets as parallel as your electrodes were ground.
You can easily adjust for size and get them perfectly dimensioned too.

Keep your finishes as coarse as you can get away with, to make it easier to hold size and to minimize arcing, and you're good to go.
Cheers

Marcus
Implant Mechanix – Design & Innovation - home
Vancouver Wire EDM -- Wire EDM Machining
Home


Wow talk about making things overly complicated. Vector the corners? Lol

The only thing I agree with there is about drilling the bottom holes on an angle so they dissolve themselves...

Directing flushing lines to the sides of an electrode 6" deep will do nothing.


If you can get away with it, just be sure to drill holes on the side of the trode on both rougher and finisher. Your contact errors are definitely from side debris.

A simple pocket stone will remove this "tit" on the side when you are done burning.
 
Hi SteveEx30:
What's so complicated about vectoring into the corners?
It's about as simple an operation as you can do; hell even my cheap and dirty Hansvedt Foreman can do it.
You DID read about the part where Mazo wants to get sharp corners didn't you?
Orbiting won't give you that, but vectoring will.

You also did read, I presume, where I recommended jump flushing and a much smaller finishing trode on the vector burns, so your contention that the flushing wands on the sides of the finisher will do nothing isn't true either.
You and I are going to have to agree to disagree on this one, but I've done a LOT of this kind of burning and I've had very good success with this strategy.

Last, if your pocket is 200 mm deep and 25 mm wide, you're going to have an interesting time removing the "tit" accurately, never mind that your pocket walls will look like shyte from all the stone marks.
Maybe it doesn't matter for this job, but it sure as hell matters when it's a mold cavity...how do you deal with that scenario?
Cheers

Marcus
Implant Mechanix – Design & Innovation - home
Vancouver Wire EDM -- Wire EDM Machining
Home
 
Mazo,

How long is your electrode that you are machining with? If you are machining this full depth with a 200mm long electrode, your machining efficiency will decrease the deeper you burn. This is the situation you have described, and is caused by the electrode picking up more surface area and secondary machining on the sidewalls as you machine deeper.

Implex's suggestions for increasing your Roughing Electrode per/side reduction are spot on, and anything you can do to improve flushing and employ different cleaning jump techniques will help improve your efficiency. To achieve deeper machining depths while maintaining your machining speed, you will want to minimize the electrode length during Roughing. This will require you to make some simple home-make tooling (like a metal post extension), but look to make your roughing electrode about 25mm long. This will improve your machining speed by minimizing secondary sidewall discharging, but for finishing, you should use your full pocket length electrode to ensure best size/accuracy/straightness.

-Brian
 
Implex, did YOU read where he referred to this as an insert pocket? NOT a mold cavity?


To deal with that scenario I just make sure the side holes are staggered in location from rougher to finisher trodes. Works just fine.


Good-Luck with your pockets Mazo. Glad you realize the flushing is your biggest issue.
 
Hello All,

After A long weekend by the machine here is what it came out:
1. I have used same undersize for roughing and finishing electrode=0,3mm/side
2. I have made in all electrode in the center for better flushing
3. One pocket I have done with Agie Equimode orbiting
4. One pocket I have done with Sliced star orbiting

The machining time was reduced form 25hours/pocket to 8 hours/pocket + additional 4 needed for dimension tuning.
No significant change on the radius if I change the orbiting 8heck Agie Equimode is really great)

For better understanding how the job looks like I am posting a screens hot from my CAD.
OR-030-ER1.JPG
OR-030-ER2.JPG
 
Man I would not want your job, or maybe I would it just seems like a lot of work+stress and I'll bet those parts are not cheep. You really need to know what you are doing every second every cut. I'll stick to flying around in my VF-2SS with a 1/2 inch end mill at 12KRPM and 130IPM thinking I'm hot stuff, you guys know your shit.

Ron
 
Hello All,

After A long weekend by the machine here is what it came out:
1. I have used same undersize for roughing and finishing electrode=0,3mm/side
2. I have made in all electrode in the center for better flushing
3. One pocket I have done with Agie Equimode orbiting
4. One pocket I have done with Sliced star orbiting

The machining time was reduced form 25hours/pocket to 8 hours/pocket + additional 4 needed for dimension tuning.
No significant change on the radius if I change the orbiting 8heck Agie Equimode is really great)

For better understanding how the job looks like I am posting a screens hot from my CAD.
View attachment 120427
View attachment 120428



Mazo,

Those look like Uni-Lifter pockets, can you confirm? I do those very often myself, we do mainly complex injection molds. Generally 8-10" deep, some straight, and some requiring vector burning.

Fun stuff.
 
I haven't been through and read all the replies, so sorry if it has already been covered.

I would use at least a 1mm gap on the electrode with sliced star. Then your final impulse orbit will be the corner radi (so it won't be 1mm). On the old agie machines you say use a 25mm thick electrode, and there was a setting to spark down say 20mm then obit out, drop down another 20mm and so on until it reached full depth. This could remove problems with 2ndry erosion as you got deeper. I can't remember the name of this stratergy or if it is available on the newer Agie machines
 
SteveEx30,

It is actually not a lifter but we called floating insert-it means that we have a negative form inside these insert and during ejection there insert "floats" for a certain travel and after that the outside is released and the part can be ejected over the negative form.

How are you burning yours-what kind of machine, electrode, flushing, erosion time,…......
 
SteveEx30,

It is actually not a lifter but we called floating insert-it means that we have a negative form inside these insert and during ejection there insert "floats" for a certain travel and after that the outside is released and the part can be ejected over the negative form.

How are you burning yours-what kind of machine, electrode, flushing, erosion time,…......


Ok sounds similar in theory less Uni-Lifter will release the under-cut or negative form as you call it during ejection.

I have adjusted my process over the years to what I have found best. Machine is a Mitsubishi Ex30 or sometimes I use the older Mitsubishi M35K. Mid-Grade graphite, I will cut to -.010" (or .254"mm /Side OB) Flush thru the center as well as the sides of the electrodes (Usually .025" flush holes). Any little dimple left on the side wall is minimal after orbit, and can be stoned away easily. Machine will never contact out and generally burns smooth from start to finish. Ensuring the machine runs ALL weekend unmanned is the highest priority to me.

I have found using Circular (XY) or Spherical (XYZ) orbit produce same end result respect to surface finish and corner rads.

If you are down to 8+4 hours to finish the pocket, I think you are doing good.
 
Further to what I said above, I would also make sure the double timer is on and is a full retract out of the part. I'd also knock the comp and gain up a bit
 








 
Back
Top