Dan Maloney
OK folks, welcome to the first Hack Chat of 2023! I'm Dan, and as usual I'll be moderating today along with Dusan -- I think, anyway. Today we're welcoming Danel Herrington to the chat as we talked about machining with electricity.
But it cuts well open loop
Daniel, are you online yet?
Hi Dan. welcome everyone!
I am here!
Welcome Daniel. Thank you for coming. I've been interested in your company for some time
Hi Dule, Happy belated New Year!
Welcome Daniel
Hi Daniel, welcome aboard. Can you tell us a little about what Voxel does, and about your involvement with ECM?
@Dan Maloney - Thanks, all the best to you too
Sure, happy to. I started Voxel in 2015 and at the time it was just me in a garage building a piece of electrochemical machining equipment from scratch. We have been focused exclusively on electrochemical machining (ECM) and its variants including PECM, STEM, etc. Early on, we were mostly doing R&D funding by federal government research contracts but lately we have started making production parts for people in aerospace, medical, and energy industries.
Although I am most familiar with the electrochemical varieties of metal machining, we use or compete with EDM (electrical discharge machining) all the time.
I love a good garage start-up story...
Is there a reason you were drawn ECM versus EDM?
ECM seemed like it had more untapped potential. It has the ability to be much faster and produce better surfaces than EDM but is not used as widely. I thought that if I could figure out why and improve the technology, maybe ECM could be used more frequently.
Seems like there's a huge established base of EDM technology already. I've only recently heard of ECM myself.
Why do you think it is not as widely used if the surface finish is superior and the process can be faster?
Bit that I've exactly got my finger on the pulse of the industry, of course
*Not that
I know the East has been developing ECM since the cold war, but I rarely hear of it
Like many technologies, the core ideas are old. The first ECM companies started in the 60s. In the US that began with a company called Annocut but there were a number of Russian pioneers in this space as well.
How did you find where the state of the art was, in order to get up to speed?
I've been trying to learn about EDM but there isn't a lot of DIY
But as CNC milling got better and the end mill materials improved, it started to replace some ECM applications. Also, some ECM companies mis-managed the waste stream (dumped into rivers in Ohio...bad idea), so that turned some people off.
How much does ECM differ from EDM?
How does voxel deal with hazardous waste(e.g. Chromium 6)? This is one reason why I switched to EDM. Much more environmentally friendly on the maker-level
So I think that has stymied some innovation in this space. But we are starting to see new interested in ECM processes because metal alloys are continuing to get harder to machine, surface finish requirements are pretty challenging, and ECM can be a cheaper way to make the parts so if you have high part volumes, it is worth exploring.
When I started the business, I read countless research papers. I got access to the NC State University library and just read whatever I could find. You can find some academic papers that describe at least the principles of the technology. But to figure out what really goes on in industry, you have to try yourself or see if you can corner an engineer from an EDM or ECM company.
Chromium? Nasty stuff, that. Is that a byproduct of ECM on stainless steel? Or in chromium used as an electrolyte or something?
I would check Google Scholar to see if there are any microorganisms or enzymes than can reduce chromium +6 into something less hazardous. Back in academia, that was a trend for dealing with difficult chemistry problems
Yes, @Dan Maloney chromium 6 is a result from ECM of stainless(or any chromium containing alloy).
What kinds of projects would you like to do?
CR6 can have it's energy level reduced to CR3 by citric acid, but one would still have to filter it out of the electrolyte using activated carbon or similar.
Wow -- I'd have thought there was so little Cr is stainless that it wouldn't matter much. But then again, it only takes a little hexavalent chromium to ruin your day, so...
In EDM, you create a spark to thermally ablate the material. It blasts the material away and makes micro-scale craters in the process. The spark happens sequentially across the surface of the electrode so if your electrode is twice as big, it takes twice as long (double the surface area for the spark to traverse). In ECM, we are dissolving metal with an anodic process. We are converting metal into metal oxide or hydroxide and liberating it from the surface - basically removing material atom-by-atom. It is a similar mechanism to electropolishing but at a much faster rate with much more current.
Yes, exactly Dan.
or it could be reduced with iron(II) sulfate.
The hexavalent chrome is definitely an issue but there are many known ways to reduce it and the wastewater treatment industry does it all the time.
true Thomas, though citric acid may be easier for the average maker to get
Iron sulfate, sodium metabisulfite, and other methods can reduce hex chrome to trivalent chrome
Daniel, how to you control for cathode corrosion? I know it should be prevented during cutting, but how many cycles can you feasibly get out of a tool?
Hex chrome is a challenge because it stays in solution so you cannot filter it. But if you can reduce it chemically to trivalent chrome, you can precipitate that metal and filter it with standard filtration processes.
Has anyone attempted to precipitate it and remove it via centrifugation? I do not know the solubility of the ECM products verses pH, but that could be handy
When you run the ECM process, you are splitting water at the surface of the cathode (tool). This means that you produce hydrogen gas and hydroxide (OH-). That hydroxide drives the pH to be very basic. So, for one, the cathode needs to withstand a basic environment. Aluminum, for example, will just dissolve.
I've seen that one first hand lol. Poor aluminum tools
But otherwise the bigger risk for the cathode is that you deposit material on the surface (since this is like a reverse electroplating operation). In practice, this doesn't really happen because the chemical conditions are not favorable and you are usually flushing electrolyte through the gap pretty aggressively.
With ECM, do you vary the voltage, to cut at different rates? And what voltages/currents do you use?
One thing that strikes me about ECM (and to some extent sinker EDM) is that making the tool seems like a huge investment in time. So there must be a kind of inverse value proposition, where the cost of tooling has to be spread over a big production run of ECM-produced parts to make it worth it. Is that true?
Stainless steel has always worked for me, and it comes in useful electrode shapes via dispensing syringes
To John's question - once you convert hex chrome to trivalent chrome, you can remove it via centrifuge if the pH is right. There is a company called Tridex (now owned by Glebar) that makes electrochemical grinding machines (ECG) and that is how they remove their waste.
EDM Die tools are not terribly expensive now because graphite blocks have gotten cheaper, and it is very easy to machine
Is the hydroxide produced part of machining aluminum? Is the electrical part important or only a way to split water?
Chromium carbonate is insoluble.
Yes, voltage is a major control variable in ECM. We often run a voltage from 5-40V and the current levels may be 50-100 A/cm2. So this means that you can quickly get into thousands of amps for a hand-sized part.
Thought. Could a tip of a thin wire scanned over the object used as a tool? I am thinking about EDM as alternative to etching circuitboards, and it may even have potential for 3d machining for eg. turning a broken carbide tool bit into a custom shaped threading tap.
Is the voltage and current only mainly to split water, and the actual cutting is chemical action of hydroxide on the metal?
I believe ECM is too difficult to localized for 3D machining if you're not running a 3D die.
Dan - you are correct. ECM (at least today - but we are working on it!) doesn't make sense for low volume work. It just costs too much to design, manufacture, and iterate on the electrode and process. So often we are making parts in the thousands to millions per year.
@Dan Maloney for EDM you use easy to machines material like graphite, brass or copper so even cheap hobby level milling machine can do that.
Daniel, can you use graphite as an electrode material?
I wish we could make our electrodes out of graphite because it would be much easier and cheaper to make them but, we can't.
does graphite have a poor reaction like aluminum?
Why is this?
Most all the electrode chemistry can be done from first principles. So experimenting can refine the calibration, but you do not need to iterate by experiments over a huge parameter space. I used the carbon electrode from batteries when I started.
What would be a common part produced by ECM that would be familiar?
@Arthur Mercer Good question
Thomas - yes, people have made a "CNC mill ECM" machine before but it is very slow. ECM is fast when you can use a tool with lots of surface area. That is why you don't see much ECM single-hole drilling. Instead, that is usually doing with hole-pop EDM, laser, or conventional drilling. But if you need to make a big array of holes, we would machine hundreds of them simultaneously with ECM and that is how we get the speed advantage.
So as far as I understand ECM vs EDM:
+not usage on electrode (could produce massive parts)
-electrode must be made from SS
-problems with chrome
Arthur - if you have bought a Norelco shaver from Philips, all of those shaver caps (the three round parts with slots and holes) are made via PECM. I think they make around 100 million of those a year.
Daniel, why can graphite not be used for an ECM electrode? is it a similar chemical issue to aluminum?
I would like to see EDM get popular for DIY people. EDM can cut stainless steel, which can then be used for making ECM tooling yourself.
What is the "P" in PCM?
PECM
I love the aspect of zero forces on the tool. That means a high accuracy positioning system can be made without requirements for high stiffness that would have to absorb the machining forces without deforming.
@Daniel Herrington what is a maximum volume removal speed you can reach with ECM?
@Cooper graphite is porous and soft, it has its places. You really ought not talk about ECM as one thing, use of electric currents for machining, plating, drilling, facing, chemical treatment -- is old technology, almost older than electricity itself.
Cooper - I believe that graphite is not conductive enough. EDM operates at 100 or 100s of volts and low amperage. We are the inverse with ECM - low voltage and high amperage. So I think to get enough current through a graphite tool, we would have to apply a lot of voltage and the combination of that would be a lot of heat.
@Thomas Shaddack that was my thought too when I first started into ExM. The thought that you could convert a low force motion system, i.e. a 3D printer, was enticing. I've found that I was unable to get sufficient accuracy from general purpose ECM tools however
Arthur - the P in PECM can sometimes mean "pulsed" (thats how we refer to it) and some people say it means "precision". Ultimately, all it really means is that we are getting the electrode very close to the workpiece (around 20um or .001") and that is what enables the level of resolution in PECM.
Thomas I also think that this is a game changer. Whatever cost of tooling it is, you can always find it profitable with enough volume
Daniel - That makes sense, thank you
Slow cheap machines can be clustered. See 3d printing farms. The same with different tool heads could become ECM farms. And revert to printing as needed. Modular high-adaptability architecture.
It is not uncommon for our electrodes to cost thousands or tens of thousands of dollars (once you include all of the development). We can make them to be very intricate because they will not be consumed. Unless the process fails and then we get a spark (a really bad EDM machine) and destroy both the tool and the workpiece. But all of this still works out when you can amortize that cost over thousands or millions of parts.
Conductivity of aluminum is 37.7 MegaSiemens per meter. Tungsten is 17.9 MSpm. Graphite is about -.25 MSpm
https://en.wikipedia.org/wiki/Electrical_resistivity_and_conductivity
Can you electrodeposit metal onto the damaged tool? Or laser-weld a little blob of metal? And then machine it to shape again?
Thomas - yes, maybe. I would not use ECM if you did not have a sophisticated electrolyte management system on hand. It is too dangerous when it comes to poisoning people. Voxel has a closed loop electrolyte system that would prevent this
stainless steel 1,32 Mspm
Daniel, do you expect metal additive to make your tooling costs lower?
We worked with chromate-sulfuric acid when cleaning glass. Cr(VI) is not THAT dangerous.
Thomas - I like the idea of a farm of cheap, but slow, ECM machines.
@Thomas Shaddack A patient man is capable of anything he puts his mind to.
You can repair tools but often it is easier or cheaper just to replace them.
I like the idea of rapidly reconfigurable machine arrays. Additive, subtractive, measuring, change tool head and on you go.
Swarm approach.
Cooper - we have used metal AM pretty extensively for tooling and it is pretty interesting. But it doesn't have good enough surface finish and you can't make features small enough for many of our needs. So that means you have to post-machine the 3D printed part. Sometimes that is worth it, but sometimes it is just cheaper to make it from scratch without 3D printing a near-net shape.
What does the tool/electrode physically looks like on an ECM machine?
https://www.voxelinnovations.com/ has some nice pictures
He only has one video at https://www.youtube.com/@VoxelInnovations/videos
Thought. 3d-print a tool mold from resin. Add a conductive surface. Electrodeposit metal on it. Use that as a tool for ECM.
Electroforming can also replicate silicone imprints. One of its historical uses was counterfeiting of antiquities.
Guru-san - the tool is basically the inverse of the part you are trying to make. But on top of that, you have to build a manifold or nozzle to force flow between the tool and workpiece. It isn't exactly the inverse (due to current field lines, how the electrolyte flows, the vector of the electrode motion) but that is the general idea.
@Thomas Shaddack the main problem with reconfiguring from ECM is the electrolyte. It tends to get everywhere unless you're extremely careful. Electrolyte has ruined my 3D printer's motion system before
Rust everywhere
Yes - the electrolyte is a pain! We have destroyed lots of things with electrolyte.
Discussions
Become a Hackaday.io Member
Create an account to leave a comment. Already have an account? Log In.