Induction Heater problem 2

[itsmoked]

I've got to go look at an industrial induction heater tomorrow.

What I know:
It's an EASYHEAT 7590LI solid state induction heater.
It's less than 2 years old.

It worked great for heating a tube up for a subsequent insertion of some piece into it for shrink fitting.
I'd do ~150 pieces a day typically.
One day it started not heating as well, taking longer and longer to heat the workpiece. No material/process changes.
They sent it to the manufacturer to check. Manufacturer said there were some minor issues, certainly no show stoppers, and that they took care of them. Returned it with a clean bill of health.


Still not working.

That's all I know so far. Besides the usual suspects of bad power, damaged cables, etc., is there anything typical of induction machines that's problematic?

Skogs I know you've worked on one.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[VEBill]

Interesting update on the controller. I've had cases where the reported evidence 'proved' that the fault could only exist in an alternate dimension of space (excluding all possible causes in This Universe). I've learned to mentally append the word "maybe" to all such reported evidence.

Where does the operating frequency (150 to 400 kHz) originate? Being a communications guy, I had assumed that there would be a low power, adjustable (controlled) oscillator that is then amplified to 9kw. An alternative architecture is a big silly free running 9kw oscillator. The key difference between the two is that in the former (low power oscillator under some frequency control), the unexplained frequency shift is more likely to be an effect (red herring) rather than a cause. In the latter (free running power oscillator), the frequency shift might be a direct part of the failure mechanism.

Is there a schematic of the "RF" section?

Does it use a bank of transistors in parallel to make the 9kw? If you have power transistors that are heat-sinked and soldered into a PCB, then there might be stress failures of the solder joints at the PCB (thermal expansion of the Al heat sink yanking on the devices' solder joints).

In my experience, about half of all such failures are visible under close inspection (the other half being internal to a component). It can be worth a close inspection of the likely areas (hot, stressed) under a high power microscope.

 

[VEBill]

Another idea: if you haven't already, monitor the internal power supply voltage(s). If they drop coincident with when the RF output drops, then the root cause must be in the PS. If the voltages ever so slightly tick up with the reduced output, that's normal and expected IR drop delta. If they're steady, that's good too.

 

[ScottyUK]

VE1BLL,

If these are ferrous workpieces then the magnetic properties change as the workpiece reaches the Curie temperature and the load coil resonant frequency changes. Most commercial heaters have the ability to track the resonant point to maximise power transfer into the workpiece.

The actual choice of frequency is based on the desired penetration depth - for localised surface heating for (e.g.) to harden slideways the frequency is high to limit the skin depth, where for bulk heating the frequency is relatively low to uniformly heat the workpiece.

 

[mikekilroy]

I saw it has serial interface... what can that tell you? perhaps you can watch some pesky internal variables real time and see which one(s) change to see where in the circuit the change occurs to help narrow down if it is in the osc, power amp, feedback, or load?

 

[btrueblood]

Scotty,

Smoked said they were aluminum parts. I'd agree that you'd want the freq. fixed once you've figured out the heating cycle you want for the particular part.

Smoked, you mentioned getting some more parts to test. I'd think you/they could come up with a) about 3 or 4 scrap or salvaged parts that can create reasonable facsimiles of the production parts, and b) several large ice chests filled with ice and maybe some water to drop the hot parts into to cool them down before running the heat cycle again, and c) some tasty beverages to help fill the excess volume in said ice chests and to keep the test operator(s) occupied whilst waiting for the phenomena to recurr. I hereby volunteer to assist you with the tasks of 1) specifying the beverages, 2) maintaing the ice chest with the beverages at the proper temperature (to be determined by repeated random sampling of said beverages), and 3) by pushing the little button that makes Skog's test-o-scope go "ping".

 

[itsmoked]

mikekilroy; If you look further... You will see the serial comm is "optional". They didn't check that box. Otherwise that's a great idea.

btrueblood; Excellent idea. Also exactly what I recommenced. We made about 20 rejects so there are lots available. I described the cold water bath and was told, "I'm uncomfortable with water near the machine". It would work absolutely fine with no issues and not even potential issues. Heck, the actual jig is even plastic! But I didn't feel like pressing the issue at the time.

Last night I tore my LCL**, so one thing I know.. I won't be wondering around any factories in the immediate future.

** Lateral Collateral Ligament

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Aw Keith. The military never worries about collateral damage, why should you? grin

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[btrueblood]

"I'm uncomfortable with water near the machine".

And you of course didn't remind him that the coil is filled with water (gasp), under pressure (shudder). But that (and the blown LCL) is why you have the trusty assistant - to fetch cold objects from the bath, and carry hot ones back to the bath. Put the bath around the corner, down the hall, out the door, around the parking lot, etc. Or even in the pub across the street.

 

[waross]

We haven't heard from you in this thread for awhile Keith. I hope that you are healing well.
Yours
Bill

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Skogsgurra]

Getting worried, too.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[itsmoked]

I'm here guys. Thanks for checking. Nothing has transpired on this project. Left them with the assignment to jig up so the problem can be brought on in a reasonable time.

No admission to this achievement yet. Just as well, I'm still pretty chair bound, haven't left the house since.. Ice, Ice, Ice, Ice, Ice, my 'walking' orders.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Let me guess; Playing sports with the kids. grin.
I had a little temporary pain last fall. I went riding with my small son to help a friend move some cattle. My horse stopped abruptly.
As my body continued forward, the saddle pushed my legs out sideways. Then inertia bent me over at the hips and left me with my arms wrapped around the horses neck. I have a little arthritis that usually doesn't cause any pain, but in that position.. OUCH. But the pain soon left and my son and I have some wonderful memories to share for years to come. (ps. I'll probably do it again.)
I did decline to go skiing with him this year and just sat and watched.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[MikeHalloran]

((( I was preparing the car for the trip South in October, forgot that I had installed a hitch ball, and banged my knee on the ball support. I spent an extra month with the knee elevated and immobilized and painful while trying to get a decent image and a diagnosis, which an MRI eventually revealed was more or less equivalent to a spall on the convex surface of the tibia. By November I could drive, and walk, slowly, with a cane. Now I don't need the cane everywhere, but i have to go up and down stairs one leg at a time. Up with the good, down with the bad. Forgetting brings immediate pain from the bad knee. ... which is still healing, slowly. )))

Part of my first job, 45 years ago, involved taking care of axle scanners, which induction heat and then quench an axle shaft while it's rotating. The cycle took maybe 30 seconds to 'scan' the shaft with a hollow copper coil, two-ish turns, maybe 3/8" diameter, thick walled, brazed to large copper terminations. New coils were fairly rigid, but after a couple thousand parts, i.e. every few days, the copper would get sort of microcracked so they didn't conduct properly, and the locomotive size M-G sets on the mezzanine would complain and the case depth (which was checked hourly) would go out of spec, at which time we'd have to call the scanner tech to install new coils.

I.e., when you start banging out a lot of parts, induction coils become a consumable.

In between coil changes, they'd still have to be removed and cleaned and flushed of stuff like boiler scale that appeared on both internal and external surfaces.




Mike Halloran
Pembroke Pines, FL, USA

 

[itsmoked]

AH man you guys give me pain just thinking out your injuries. Ouch.

I was on a 3 mile hike with the dog. It got dark on us. I stopped at an oft missed turn to study it. The path I was standing on collapsed down a steep hill under my left foot. Not desiring to fall a hundred or so feet I immediately pitched with everything I had forward. Everything went onto my right knee cap, even skinning it. It hurt but after 5 minutes dwindled off. I trekked the mile and a half back to truck without issue.

Couple of days later I got up to walk dogOwitz after dinner and my knee hurt enough I found myself limping a bit on the evening mile. Next day my LCL was a little sore. That night I was really limping right out the gate. I cut the jaunt short I subsequently went to my office.

Coming out late the big security gate was closed. I limped over to open it. It's a big ten foot wide chain-link vehicle gate with slats thru it making it really heavy. It's so heavy it has a set of outrigger wheels on the opening end. Sometimes while looking at the latch you run one of those wheels into your foot.

Tonight I looked at the inside wheel, where I was, and thought what if I push that wheel with my foot? I did it with my right limping foot. The gate blew open! 'Wow that works really well'. Must be the effective gear ratio or something. I walked thru then had to reverse the process. I immediately realized I couldn't use my right leg to reverse the process because 1) I had to make the gate come towards me now instead of a shove-away. The outside wheel was now next to my left foot so I stood on my right limping leg and tried to "foot" the gate towards me with my left foot. This put all my weight on my bad knee. And kicking back towards oneself is.. cumbersome. I promptly lost my balance with the twisting/footing motion and to recover it, my bad leg had to provide the restorative torque. There was a tearing sound in my head and a lighting bolt of pain out of (what I now know) was my LCL. It brought tears to my eyes. Crossing the road 30' to my car took me five minutes. I could weight bear on it at all the next day. Luckily my son was just off his crutches (Lyme) so I was able to commandeer them.

Mike: Thanks for that info. I am pretty sure the problem is in the "head" where the matching transformer and the coils connect or the coils. I'm not sure the coils are water cooled. The coil wire is really small and wrapped with some burnt orange colored tape. They sent the head back for testing but not the coil. Definitely something to check.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MagicSmoker]

Sorry for the late contribution to this thread. Just wanted to toss in my experience with a similar problem of a wander resonant frequency on an induction brazing system (for attaching carbide tips to saw blades). It turned out to be gradual failure of the film capacitors that are used to form the resonant tank with the work coil. The proximate cause was (painstakingly) traced to a failure of the joint between the terminals and the "schoopage", likely from thermal cycling. In short, whenever there is a problem with an induction heating system I suspect the tank capacitors first.

BTW - solved that particular problem by replacing the film capacitors with ones designed for GTO snubber use. AFAIK, that induction heater is still in service some 10 years later.

 

[itsmoked]

Very good info! Tell me that cap would be in the work head instead of the controller?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MagicSmoker]

itsmoked: yep, the tank capacitors were inside the work head. Another important factor was that there were several capacitors in parallel and the one closest to the induction coil had the most internal damage - obviously because the circulating current between the L and C wasn't dividing equally into each capacitor. I don't remember the exact value of the capacitance, but it was somewhere in the 1-2uF range and was easily accommodated by a single GTO snubber capacitor. I also recall that the resonant frequency of the system changed enough to technically take it out of "spec", but the but the controller/driver had no problem with it.

 

[itsmoked]

Thanks again Magic!

You mentioned "gradual failure". How exactly did the symptoms manifest?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MagicSmoker]

I wasn't really given an exact sequence of the progression of the symptoms from the client, just a complaint that the brazing process was taking longer. What I concluded based on the internal damage to the capacitors was that the ESR of the capacitors increased as the connection between the end spray metallization ("schoopage") deteriorated, leading to more heating and more schoopage deterioration. Not precisely the same symptoms as what you've reported here, but I figured it was worth mentioning.