Metallurgy? Wire EDM? Black Magic? Issues with HSS wear and looking for answers

So, we have been making some cutting tools made of HSS, lately. Up to now, I have tried some pre hardened HSS as well as using some "donor" tools that were already working tools. Both choices result with using WEDM to cut the tool I want, from the "blank".

All examples were tested before AND after WEDMing. They were even tested after sharpening. Bottom line is that they are all verified to Rockwell C scale 61 - 65.

The issue I am experiencing is that the cutters are not showing any life. AT ALL. I can get about 1 - 2 passes or uses from them and then they need resharpening. Teeth appear to be crumbling. When they do not crumble, they round over. Much like one would expect to see from typical, normal alloy steels running against something harder. Except, that the situation is decidedly reverse. The work piece is in the low Thirties to high Twenties in Rockwell C scale hardness, and the cutters ( as previously mentioned ) are in the 60's. They come from several sources, so it isn't the origin of the material.

Frankly, it is acting almost exactly how we used to expect Carbide to act after WEDMing. This only makes it more confounding and frustrating. Interestingly, once they are re-sharpened to a point much deeper, they perform more adequately but still dull fairly rapidly, considering. We're getting only about 7 - 10 applications before needing more extensive re-sharpening. ( They should last at least twice as long, and need 1/4 or less sharpening. )

So, I am wondering if there is something I am ignorant of with regard to WEDMing hardened HSS, whether it be process related, or metallurgical, or simply some manner of black magic. It just isn't making sense.

Ironically, the carbide ones are performing wonderfully!

I do not really know but it sounds like the EDM has annealed the HSS for some small distance. When you get it sharpened deep enough you get to the non annealed part so it works fine from then on. I thought it was not uncommon to wet grind hardened metal to keep it cool.
Bill D.

With both sinker and wire EDM you get micro cracking at the cut edge. It’s never been a problem with the bits I’ve made on either of my machines (sinker and wire), but I’m only an enthusiastic amateur so not doing vast amounts.

I understand super critical parts should have the surface etched away to remove the micro cracks which of course act as stress risers.

Hi ZK,
It's likely there's some localized thermal or chemical damage being done by the EDM cutting. If you have options to try using as low power settings as possible during the cut, and/or a finer wire, see if those make a difference. A change in wire (from, say, bare to coated) might help too.

If you really want to investigate things, take a sample steel, make a straight cut across it so you wind up with a flat surface on a sample perhaps 3/8" to 1/2" on a side, then "pot" the sample in a good epoxy with that flat face down inside a sample dish. Then,using low-stress abrasive methods (hand lapping on increasingly fine silicon carbide paper with water) take the minimal amount of material to polish the surface.

At that point you can etch the surface with some metallurgical acids to expose grain structure or mechanical faults that can be analysed for deviation from a similar sample prepared from a known-good section of steel. A microscope meant for metallurgical work with a good light source will be important for this step.

Metallurgical Microscope Information

Another aspect is microhardness Vs bulk hardness - the methods used for checking shallow depths of hardness are different than a "standard" Rockwell or Brinell hardness test.

Microhardness Testing, Microhardness Test | Laboratory Testing Inc.

It's probably best to take roughly prepared samples to a good metallurgical lab and let them do the full analysis. While not cheap, it would give you valuable data with confidence in the results.

As mentioned, it’s probably burning too fast, making the recast layer too thick. You don’t necessarily need to lower the power settings, it’s unfortunately not that simple. Try changing the on/off times or find a parameter with more off than on.

Could also be you’re not getting enough fluid into the burn. If it’s submerged, still put a nozzle facing the workpiece. If not, might try some adjusting.

A snapshot of your parameters and a picture of what the surface looks like after the burn would help; I’m just guessing here.

I appreciate the responses. I remain open to learning anything possible from this, but have serious reservations about the concerns expressed. These were all 3 and 4 cut strategies, using low power. While I am keenly aware of recast, I’m willing to bet hard money that a recast layer is not the issue, here. Simply put, after one roughing and three skim passes, there is basically no recast layer to be problematic. As well, they were still lightly glass beaded, at low pressure, and following that, they were then sharpened. In short, I’m not new to WEDM. Only new to this specific application, and with a fair amount of experience doing similar work. Just not this specific application.

I AM willing to buy the micro fracturing, but will want to verify it so that if it proves to be the case, we can take steps to work with or around it. Just seems odd that quality HSS would perform as poorly. These are the concerns that are typically attributed to wiring carbide. Not HSS. As for the heat concern, I am again seriously dubious. Even when sharpening, temperature of the tool in general never exceeded 110°F and was monitored with a laser thermometer, aimed at point of contact.

Zahnrad Kopf said:

As for the heat concern, I am again seriously dubious. Even when sharpening, temperature of the tool in general never exceeded 110°F and was monitored with a laser thermometer, aimed at point of contact.

Click to expand...

Laser thermometers (cheap-ish) are not great for monitoring metals temperatures. Broad field of view (so it's easy to pick up readings out of the zone of interest) and emissivity matching when dealing with a metal are issues that are better dealt with a proper IR camera like a Flir. A fast reacting contact thermometer would be a good choice, but there's still a big delta from the actual grind zone to where the measurement will be.

My experience was the same as Camscan's, did not do too many but never had a problem. Some were used in a screw machine others in a manual turret lathe.
My guess is your problem has to do with speed or power settings. For a test compare a simple grind on one end and the same EDM profile on the other end of the same blank.

I know next to nothing about edm, but do have a chemistry/chemical engineering background. Could you be depleting the surface of the tool with some sort of electro-chemical process? There are electrons flowing. You might be "deplating" a critical trace material(sort of like the loss of carbon due to improper heat treating techniques).

When I was in industry, with a gigantic industrial laboratory, I would have done elemental analysis via SIMS on the surface, and analyzed the electrolyte for ionic content, but that's probably not realistic in your situation.
Can you change the dielectric to alter the chemistry or reverse the polarity you are using?

When dealing with EDM, I'd lean towards "black magic" every time

Just a shot in the dark, but perhaps the EDM process is affecting the interface between the various metal carbides (non-conductive?) and the alloy matrix? The ebb and flow of electrical current near the spark erosion zone could have some impact on the microstructure interior to where the actual material is removed.

That said, my first job was at a machine builder for micro-hole EDM machines. Largely poked holes ranging from .005~.011 on otherwise finished, already hardened diesel injectors, and never had any issues with rapid wear/deterioration at the EDM zone. DI water dielectric, pure tungsten electrodes, .2µs on 5µs off typically, slow and steady cutting pace (like 1 min/hole for a .008" diameter .100" deep thru hole). In case you care about a concrete comparison of experience.

Milland said:

Laser thermometers (cheap-ish) are not great for monitoring metals temperatures. Broad field of view (so it's easy to pick up readings out of the zone of interest) and emissivity matching when dealing with a metal are issues that are better dealt with a proper IR camera like a Flir. A fast reacting contact thermometer would be a good choice, but there's still a big delta from the actual grind zone to where the measurement will be.

Click to expand...

Yes, and no... It's a _halfway_ decent model, and accurate enough that I use it semi-frequently for verifying some heat treat/tempering concerns.

That said, it's not the only measure that was employed. I've been making and grinding tooling for over 30 years now, and being able to keep my finger on the rotating tool while its being ground is just as valid in verifying that a "Red Hot Steel" ( HSS ) did not ever come even close to being annealed or heat checked during the grinding ( sharpening ) process. Combined with a cold air venturi blowing upon the tool during the grind, I just do not give much credence to the concept that the tool ever became softened by heat.

wheels17 said:

I know next to nothing about edm, but do have a chemistry/chemical engineering background. Could you be depleting the surface of the tool with some sort of electro-chemical process? There are electrons flowing. You might be "deplating" a critical trace material(sort of like the loss of carbon due to improper heat treating techniques).

<snippage >

Can you change the dielectric to alter the chemistry or reverse the polarity you are using?

Click to expand...

BoxcarPete said:

When dealing with EDM, I'd lean towards "black magic" every time

Just a shot in the dark, but perhaps the EDM process is affecting the interface between the various metal carbides (non-conductive?) and the alloy matrix? The ebb and flow of electrical current near the spark erosion zone could have some impact on the microstructure interior to where the actual material is removed.

Click to expand...

This is exactly what my gut is screaming. I really do think it is something along these lines. I do have the option to reverse polarity and may try it. In all honesty, though... It's a small pain in the buttockal unit to do and if that becomes a requirement I may simply default to making all the cutters from Carbide blanks simply so I don't have to deal with it. I was thinking that the HSS blanks would be a little more economical to accomplish once we get to a 3" - 3-1/2" diameter of size due to the Carbide likely having less lateral strength at typical thicknesses. However, I've given it some thought prior to this and am fine with brazing the Carbide Blank to a Steel "backer" to prevent strength issues. Just not a big deal for us.

As for testing the tool from one side to the other, that would prove a bit difficult, as the tool essentially acts like a planing/shaping/broaching tool for cutting and given the relief and specific form involved, it would not be of any real use at the other end and reliefs would be inverted, preventing any real useful testing. These are Gear Shaper Cutters.

One thing I may do is change how I have been designing them, thus far. Heretofore, I've been designing the initial cutting edge profile to be of the optimal Form and size for the job at hand. However, I may dabble with the idea of designing them purposely oversize, such that grinding for sharpening takes enough of the result away to leave the tool at that optimal Form size AFTER the initial grind.

Thanks all.

Hi there Zahnrad Kopf:
Are you interested in trying a simple empirical test?
How about if each of us cuts an identical form tool on a single stick of HSS; you put your form on one end, send it to me, and I'll put the identical form on the other end and send it back to you.
That way you can determine if it's your machine that's the source of the problem or something else.

If we make something simple like a 60 degree vee threading tool, you can run threads in the lathe until it craps out and then flip the stick and do it again to compare.
You could also cut it in half, and grind the profile on one end to compare ground to wire cut.
I'm game if you're game.

You could also do it with someone closer to you if you want to avoid the cost and hassle of shipping across the line and across the country.
Let me know if that idea is attractive to you...I must confess I'm as puzzled as you are about this, and I doubt that theorizing about it is going to get us very far.

On a related note, but still speculating wildly, I wonder if powder metallurgy processed HSS performs the same after wire cutting as conventionally processed HSS?
I've never checked.
I wonder also if cobalt HSS, especially PM cobalt HSS suffers from cobalt leaching like tungsten carbide does...I can't see how because the chemistry and microstructure is completely different but if cobalt generally dissolves in the wire EDM environment then...?
I did read that cobalt carbide along with other intermediate transition metal carbides are soluble in acids and in water but to what extent I haven't a clue.

Cheers

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

implmex said:

Hi there Zahnrad Kopf:
Are you interested in trying a simple empirical test?
How about if each of us cuts an identical form tool on a single stick of HSS; you put your form on one end, send it to me, and I'll put the identical form on the other end and send it back to you.....snip

Cheers

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

Click to expand...

Now THAT is professionalism.

We make stamping tooling for very small stuff. So take this with a grain of salt.

Have you tried changing materials? I would be very interested in seeing how A CPM alloy would perform.

We usually wire out our die sections (very small inserts) to leave around .003" per side to grind/adjust the insert for fitment and cutting clearance. We used to use HSS for this and got tired of the wear and the Likelihood of the material we were cutting sticking to the HSS. This was really bad when we cut nickel. As soon as we switched to CPM 15V the sticktion went away and as an added bonus, the edge retention was three to four times what the HSS was.

Just brainstorming here, but it would be very easy to try. Maybe even pop for a cryo H.T. to boot.

implmex said:

On a related note, but still speculating wildly, I wonder if powder metallurgy processed HSS performs the same after wire cutting as conventionally processed HSS?

Click to expand...

Sonuvab|tch.

That's it.

This has been bugging the living shite out of me the whole time, for two reasons. The first, because I just could not understand why HSS would be affected this way. The second, because it was just nagging unrelentingly in the back of my head that I had seen this before but could not place it. I kept comparing it to how Carbide used to act when wired.

When I read your pondering, it struck me like a lightning bolt. I've seen this before, in POWDERED METALS. Extensively. As you will likely recall from our conversations, I've done a fair amount of work prototyping gears and other parts with Powdered Metals. Many different grades and types. This is almost identical to the types of failures one would see when running torture and endurance tests upon those PM parts.

FURTHER... This was common in PM when it was WEDM'd and other measures not taken to combat it!

The bottom line is that the WEDM process essentially removes some of the binders at the surface of the PM material. ( as well as weakens some yet further from the surface via the electrolysis encountered ) When this is done upon Green PM, it is of less concern, since they are to be heat treated and processed further, later. HOWEVER, when the material is already processed ( pressed and heat treated ), the damage is done and is either accepted or accounted for by further refinement of the exposed surface in much the same way we dealt with Carbides.

I'd bet hard money that this is what is occurring here, and feel even more confident because the donor cutters I started with are CPM HSS. ( !!! )

I'm looking for some 3-1/2" Diameter M2, M4, or M42 right now so that I can prep it, harden it, and then make a test cutter with it for comparison. I do feel rather confident that we've solved the riddle, though.

Thanks, Marcus.

Zahnrad Kopf said:

Bottom line is that they are all verified to Rockwell C scale 61 - 65.

Click to expand...

Too soft, I think, ZK. It's been too long but I am almost sure shaper cutters are harder than that. What I do remember is ITW was quenching cutters from less than 50* away from the melting point. And then some are nitrided, which would make them 70 or better on the surface. Even roller bearing life is shit under 68 Rc.

EmanuelGoldstein said:

Too soft, I think, ZK. It's been too long but I am almost sure shaper cutters are harder than that. What I do remember is ITW was quenching cutters from less than 50* away from the melting point. And then some are nitrided, which would make them 70 or better on the surface. Even roller bearing life is shit under 68 Rc.

Click to expand...

Hi EG,

Those ARE OEM shaper cutters. However, see my last post... They are all CPM HSS. It makes much more sense, knowing that.