What's new
What's new

thinking of getting a sinker edm

kpotter

Diamond
Joined
Apr 30, 2001
Location
tucson arizona usa
I make hubs and dies for the jewelry industry, I currently carve them by hand and with manual mills and pantographs. We are modernizing our business. I would like to get into using an edm. I now have a haas mini mill that I was thinking I would use to make the electrodes. I currently carve hobs and sink them to make the female dies using a 500 ton hobbing press. I would like to sink female dies and then raise the males in the press rather than carve directly. Whats a decent sinker cost and what brand should I be looking for.
 
Hi Kevin:
How are you planning to create the CAD file you'll need in order to cut the electrodes?
Will you hand carve a master model and have it digitized?
Will you learn Fusion 360 or some other CAD software that has decent freeform rendering?

Next question: do you expect to save time or money this way?
Do you hope to be able to delegate this work to others?
Do you expect to get a better result this way?

I ask because the answers to these questions are not simple, and I'm not sure you're aware just what a big undertaking this will be in time and money.

Moving on to what you actually asked about...the brand of sinker to buy.
Will it be new or used?
Will it be all manual or ZNC or full CNC.
Do you have a budget in mind?
Are you hoping for "push the button" easy or are you willing to fart around getting it to work.
How risk tolerant are you of buying a clunker you cannot fix?

Once you have described a bit more of what you intend, we can offer useful advice.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
I hand carve everything right now, I have been doing it this way for my whole life. I want to keep my company running after I am no longer able to do this work. I have employees that can program and set up cnc mills. I am having a die made right now using an stl file that was done from a scan of one of my carvings, I did it oversize like we would for a pantograph. I am going to have the graphite electrode milled in a cnc. I am going to help a local shop set up and sink the die for me on their edm. Theres is from the 80s they do plastic injection molds. They have never done ornamental stuff. I will get to see what is possible by doing this and hopefully be able to make a better informed decision.
 
OK, Kevin, that tells me a lot already.
If you are hand carving a master and digitizing it, you've just circumvented the need to get proficient enough with a CAD program to make a file that doesn't look like crap.

I believe that to have been a smart decision.
I say this because it's not easy to make real artwork on the idiot box...there are some who can do a credible job of it, but most of what I see looks like shyte.
This is especially true of subjects from nature, like animals or flowers.

Once you get from the artwork to a CAD file the rest is just technology, and what to buy depends on how much hassle you're willing to endure (or willing to subject your employees to) to get through the technological steps.

So the choices for a machine are threefold:
1) You can buy older technology...like pre-CNC.
The pros...it's cheap...five grand gets you a respectable machine.
The cons...it's a crap shoot whether you can make it work.
Once upon a time the premier brands for manual sinkers were Agie, Charmilles and Eltee.
A second tier brand was Hansvedt, and there were a few others that found their way into lots of mold shops back in the 1980's
These were hydraulic ram machines, and their point of vulnerability is the hydraulic servo valve that operates the ram in response to the gap sensing voltage that tells the ram how close to bring the electrode to the workpiece before discharging a spark.
Not replaceable and not fixable so if your candidate machine has or develops a bad valve the machine is scrap.

2) In the early 1990's CNC sinkers took off, in part because they're MUCH easier to use.
The top brands were Agie, Charmilles, Sodick, Mitsubishi, and I've seen prices hovering around 15 to 20 thousand bucks for one of these in operating condition.
Their vulnerability is electronics...flakey computers, flakey floppy drives, flakey circuit boards.
Many of these are fixable only by old codgers who were service techs in the day and remember how to service them and have access to remanufactured or salvaged parts.

3) You can go balls-out and buy a brand new toy for somewhere in the hundred grand plus range.
You will get a phenomenally accurate, easy to use, capable of super finishes, new toy but the cost is pretty breathtaking for occasional use.
Makino, Sodick, Agie/Charmilles, Mitsubishi...all will be more than adequate for you so pick one that has decent local support and training, and you will be more than happy with any one of them.

There are lots of "off-brands" out there...Chmer makes some decent machines, ONA seems to be popular in the Mediterranean, and there are others I can't think of right now.
If there is a service guy right in your neighbourhood, they are fully capable machines, but there's not much of a presence for these machines on the internet, so they can be a challenge to get good at, especially the ones with Chinglish operating manuals.
I have a Chmer wire EDM bought new in 2011, and I still haven't figured out some of the functions.
Typically some dumbass engineer made a totally obscure function, like you have to press two buttons together and then a third one at the full moon while dancing naked under the oak trees to even turn the bloody thing on.
Weird shit like that can drive you nuts, and if you're the only one in North America who has one, not even Google or Youtube can help.
And the manuals are hopeless.

So that's it in a nutshell.

Cheers

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

Oh, while I'm at it, graphite electrodes on the Minimill will beat it to ratshit in short order...that stuff is abrasive as hell and gets EVERYWHERE.
Consider making your electrodes out of tellurium copper instead...they'l work almost as well and your shop won't look like a shithole in a week.
Telco is also easy to engrave if you want to sharpen details you can't easily mill.
MC
 
Oh, while I'm at it, graphite electrodes on the Minimill will beat it to ratshit in short order...that stuff is abrasive as hell and gets EVERYWHERE.
Consider making your electrodes out of tellurium copper instead...they'l work almost as well and your shop won't look like a shithole in a week.
I was going to say, if you are thinking of going this way then look into graphite machines. They are especially made for this and used ones are cheaper than you might think, since they are not super-suitable for regular milling. They usually have really fast spindles, which you be good for you. And they can cut graphite without destroying themselves as fast.
 
Hi Kevin:
You asked:
"I was wondering about copper electrodes. I like that idea much better than graphite. Do they work well."

Yes they do; here is a shot of a set of mold cavities I just finished burning, together with 3 of the electrodes I used. (there are 9 trodes in total and you can see how I split up the burns to make the trodes easier to cut).
If I'd used graphite trodes I might have gotten away with only two trodes per burn but my burden of work would otherwise have been the same.
Letting the CNC mill bounce around on a piece of copper is no great hardship...it only consumes time and a tiny bit of resources, so I don't see it as a big show stopper for what I do.
If I was a proper mold shop running many electrodes per day, I'd invest in what it takes to capitalize on the better performance of graphite (faster burns, lower electrode wear, burr free trode machining), but I build a couple of molds a year so all the smelly black crap up my nose is just not worth it to me.

There isn't much I cannot burn with a combination of Telco copper and copper tungsten.

Cheers

Marcus
www.implant-mechanix.com
www.vancouverwireedm.com
 

Attachments

  • DSCN4974.jpg
    DSCN4974.jpg
    90.7 KB · Views: 249
Hi Kevin:
I know silver tungsten is used sometimes but I've never done it and I don't know why it is desirable.
I do know it's expensive.
My guess is silver is not commonly used because it costs a lot and some of it is consumed.
It may not be a problem when your electrode is jewelry sized but is surely is an issue when it's a car bumper sized electrode.
Were you thinking of casting your electrodes?

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
2) In the early 1990's CNC sinkers took off, in part because they're MUCH easier to use.
The top brands were Agie, Charmilles, Sodick, Mitsubishi, and I've seen prices hovering around 15 to 20 thousand bucks for one of these in operating condition.
Their vulnerability is electronics...flakey computers, flakey floppy drives, flakey circuit boards.
Many of these are fixable only by old codgers who were service techs in the day and remember how to service them and have access to remanufactured or salvaged parts.
How about a 2002 machine ? Just happen to have one for sale...:smoking: 2002 AGIE Mondo Star 20 sinker CNC EDM w/ C Axis and tooling | eBay

One of the cool things about Agie is the phone support is fantastic ! (and free) Couple of guys there that know these Mondo machines inside and out...seriously in all my years of messing with CNC machines they are the best of the lot.
 
Thats exactly what I was hopeing to try, casting electrodes that I printed or carved from an stl file. I have the tooling and electrodes from several companies that have closed and one of them was hand chasing copper sheet and engraving it and then mounting the pieces to an edm and burning dies. Reproducing hand made work with modern machines. This is what got me excited about the process.
 
Hi Kevin:
Something I've always wanted to try but have never had time to get into, is electroplating electrodes using copper.
I believe Bud Guitrau's excellent book on EDM shows an example of a big electrode that was made that way.

I believe the process flow would work something like this:
Make your master by hand carving.
Take a Silicone mold off it.
Coat the silicone mold with silver nitrate to make it conductive.
Electroplate it with copper.
Pour in Cerrobend to make a good thick backing.
Mount your electrode and burn away.

As I say, I've never tried it, but it sure sounds intriguing.

Another way if you don't care about dimensional accuracy so much is to pour Cerrobend into the silicone mold and just copper plate the Cerrobend.

Of course, you could always just investment cast your electrodes in copper instead of silver.
You just have to learn the details of handling copper to get what you want.
Shooting the waxes, investing, burnout and all that should be pretty much the same; maybe you have to get a different kind of investment and wind the sling caster a few more turns to compensate for the different metal density, but I'll bet a few experiments will get you there.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
I like that method you described it is right up my alley. I am going to give it a try. I appreciate you taking the time with these questions.
 
Hi Kevin:
If you want to go down this path there is a fundamental thing you need to give some thought to, and that is how you intend to align your electrodes with your workpiece, especially if you need more than one to complete the burn to your satisfaction.
As you may know, the electrode is partially consumed, the burns that I showed previously in this thread consumed about 0.010" of the first roughing trode, and about 0.002" of the second trode.
They are roughly 0.400" deep and were burned on "no wear" settings.

So if you want decent detail, having more than one trode is pretty much a requirement for any burn that has any depth, and the alignment has to be quite good.
I try to get my electrodes aligned within tenths, and there are a bunch of ways to do that.

First, since I mill them I can make alignment features on the electrodes...simple things like square sides on the trode body of a known size or a precision bore into which I can push an alignment pin.
or dowels to mount subsequent trodes onto a holder.

The second way is to have modular holders that go into a receiver mounted in the mill, so every electrode goes onto a holder and the holder goes into the receiver on the mill.
There is a duplicate receiver mounted on the ram of the EDM machine.
All modern sinkers use a system like this; some popular brands are 3R, or Erowa, or Hirschmann.
They are breathtakingly expensive, but they mean duplicate electrodes will align as accurately as the mill that made them...the receiver system is typically the most accurate part of the whole shebang.

Your work will not likely demand anything remotely like the sort of precision these systems operate to, but you still want the alignment to be as good as possible, and that is not a trivial problem when you have no machined reference surfaces from which to gauge.

As soon as you cast your trodes or electroplate them, you confront this problem and you need to find a solution.
It may be as simple as casting a plaster or epoxy block into which you can set the trodes and mill a rectangular holding tang or drill and ream a couple of dowels in the same location for each trode, but you will have to think of something.
The variability from trode to trode will mean each burn starts out eroding in only a few spots until it seats into the burn, so getting the whole surface to have an even burn finish requires that you burn until you're touching everywhere.
Finish burning is very time consuming compared to roughing, so the better the alignment, the less time it will all take, even if you're not trying for dead nuts accuracy.

So cast your mind to a solution for that problem, whether you make a soldering or epoxying jig to align your trodes to an accurate square post, or whether you make a milling or drilling jig, or whatever.
Be aware this does not have to be very strong; you just have to have a conductive path from holder to trode.
There is almost no mechanical force on the system while it burns...if you can resist the flushing pressure and the force of gravity on the electrode, you don't need much more than that...it's a non-contact process.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
Hi again Kevin:
Further to my babble above, I've been dreaming casually about how I would do it if it were my problem.
Here's my first kick at the can:

I'd mill a rectangular pocket in a chunk of aluminum that's been squared up.
I'd tap a couple of holes on two sides in one corner to allow me to bolt and angle plate onto the outside of the block.

I'd mill up a square 1" x 1" steel stem and surface grind it nice and square.
I'd tap a hole in it somewhere on one side near the foot so I could attach a copper wire strap.

I'd cast my copper trodes with a skinny stub shaft on the back somewhere out on the periphery of the back face that I'd use as a rivet or a positioning post for a solder joint to hold the other end of the conductive strap.

I'd block out one of my cast trodes with an easy-to-remove wax so that the copper is exposed only around the rim and on a small part centered on the face of the business end.
I'd cast a silicone holding fixture using the aluminum block with the pocket in it as a base.
I'd pull the first trode, pick off and boil off the wax, then re-seat it in the silicone mold (blocking out most of the trode makes it MUCH easier to re-seat and also to seat subsequent trodes accurately)
I'd make and bolt an angle plate to the aluminum base that will align the stem over the center of the pocket in the aluminum base.

I'd rivet or soft solder the conductive strap onto the copper trode using the little stem as my rivet or positioning post (this is so I don't have to bother drilling and tapping the trode)
I'd clamp the steel stem into the fixture so it's positioned over the trode with a small gap between the trode and the stem.
I'd fill the gap with epoxy, let it set, and connect the other end of the conductive strap to the stem with a screw.

Once the first trode was used up. I'd pry it off the stem, seat the next trode and stick the stem to the next trode the same way.
I'd go through all the trodes in succession.

To keep the stem from getting beat to hell, I'd consider hardening it before grinding.
To make it easier to get a used-up trode off the stem I'd consider putting some setscrews in the business end of the stem face so they're accessible from the back end.
That way I can jack the used-up trode off with the setscrews so I can re-use the stem easily.
To make the whole thing usable for many shapes I'd make the pocket fairly large and plan to throw or bolt in some aluminum scraps if I want a smaller pocket to save on silicone.
If I was going to get really fancy I'd make a set of nesting frames I could bolt in.

That's it...should work well if done with reasonable care.
Let me know your thoughts.
Cheers

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

Before you go to far down this road you should buy Bud Guitrau's book and
read the sinker section. Good to know what you are getting into.

I think a non-cnc would be just fine for what you are doing. The big advantage
for a cnc is the orbiting which reduces wear on the electrode for certain
jobs. For jewelry artwork, I don't see any advantage to using orbiting.

The advantage to a manual is that you don't have to worry about the cnc
control going south on you. It will be easier to learn to use a manual
machine and to get started. You don't have to worry much about wear since
sinkers don't move around much.

Good luck,

Paul
 
Hi Paul:
I can see your point to a degree, but there are some things that make me believe your approach may be a lot more frustrating for an EDM beginner.
Yes manual machines are mechanically and electronically far simpler...no argument there.

However, one of the great advances in sinker EDM technology in my opinion, is the development of adaptive control.
The big benefit is that you no longer have to know nearly so much about what's gone wrong when a burn stalls; the modern controls adapt on the fly to changing cutting conditions and are much less hair tearingly "black artish" to get to burn successfully.
Not to say it can't be done...of course it can as lots of old codgers like us can attest, but the learning curve can be painful.

The thing many beginners to EDM do not appreciate is that the successful efficient burn requires a whole lot of variables to be set pretty close to correctly in order to make reasonable progress.
Adaptive controls allow you to set up what you want to achieve regarding speed, final overcut and finish and let the machine figure out the best way to get there.
The control will constantly be doodling the pulse frequency, the spark duration, the input energy, the gap voltage and on and on, in order to get you most efficiently to the desired size, burn depth and finish.

In my opinion, no human operator can do that, or even close to that, so a successful burn on a manual machine will take more experience and will always be much slower than a fully adaptive CNC sinker.
I've burned in days, what my buddies with proper sinkers can burn in hours because of that reality, and I've been EDM burning for decades.
Those extra days of fucking around not only tie up the machine and me for longer, they also wear the electrodes more, so I have to make and set up more trodes to get a decent result.

Last, all EDM machines, even dinosaurs are complex devices.
"Simpler" in my world means "I can fix it myself"
With an EDM machine that's barfed its guts...even a golden oldie, I have about zero chance of fixing it without schematics and a whole lot of electronics knowledge I don't have.
Schematics are guarded like the crown jewels by EDM manufacturers, so they're not readily available.
So the admittedly less complex electronics of an '80's Hansvedt are actually no less daunting for most than the control of a '90's Charmilles, but the two are worlds apart in capability.

I'd be trying to score a good deal on a used machine of around 2000 vintage...the one Milacron posted a link to would be a great candidate.
Yeah it's twenty grand, but if you want to do more than occasional simple burning, it'll be SOOO worth it if you can swing the cash.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining
 
Last edited:








 
Back
Top