STATIPOWER HF Welding supply

[Skogsgurra]

This one is old - even by my standards. From 1975, I think.

It is an 80 kW 10 kHz inverter for induction welding.

The problem: Fuses blow when the inverter is started.

We have checked gating pulses, all there, right shape, right amplitude (yes, checked gate current, too).
We have checked thyristors. All thyristors seem to be in good working order. We didn't check with full rated voltage. But was OK at 500 V.
No GND fault. No shorts.
The soft start circuit seems to have been tampered with. The schematics are not very detailed. We do not know how it is supposed to work. Could be a simple current limiting resistor that is shorted by a DC contactor.

Anyone having any experience with this make? Similar problem? Please?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[CharlieGill]

No suggestions, but I don't think 1975 is all that old

Charlie

Charlie Gill
--

http://www.calibrator.com--

 

[Skogsgurra]

Agree. I have seen older equipment. But, nonetheless rather dated.

Anyone that has been in touch with the IPE induction welders lately? Does the company exist? Other name?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

You saw this?

http://www.inductoheat.com/statipower-sp16-induction-power-supply.php

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

Thanks, Keith!

Yes, that's the one. Compare the front panel from year 1975 to the one on the home page abt 30 years later. A very slight modification - but not much. Obviously a very successful design.
Still stuck with this beast. There was a problem with the soft start - it didn't work at all. Connection lost. Fixed it and reduced starting surge from around 700 A pk to ten percent of that value. Thought everything OK. But no.

Still blows fuses when gating pulses applied. Next thing is to remove 16 thyristors that are hidden behind copper tubing (cooling water AND electrical connection) to do a thorough test.

I will contact Inductoheat to get some inputs from them.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

Yes you should contact them. They have a parts supply section. And list the service staff. You could run the question past them as they have probably heard it all before and can confirm/negate your theories and also supply parts. They may have suggestions about what else you'd want to replace at the same time.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

OK. It worked - in a way.

Contacted inductoheat and got an answer after a couple of hours saying that we should talk to the office in UK. That's the good part of it.

Talked to the UK office. All that knew anything about equipment this old had either died or retired. So, no help.

Lots of copper tubing and hoses for water cooling. Deep under all this, we found thyristors and triggering boards. Took six hours to get to them. Not exactly built for easy service.

Checked everything, found a faulty triggering transformer. Replaced. Another six hours of re-connecting pipes and hoses. Test run.

No success. Water system clogged. Not enough water flow. Obviously happened something when removing/fitting pipes and hoses. The guys worked almost 24 h to fix it. Test run. Worked.

For four days. I did active duty for around four hours. The rest was plumber's work and hunt for fuses.

Lesson learned: Stay away from that kind of equipment - or make sure you get paid also for idling.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Skogsgurra]

Follow up.

Machine failed again after 1200 cycles. New visit on site.

There are four problems: No one seems to know anything about this equipment any more, the thyristors are "secret brand" and cannot be found anywhere (remember, they were made before 1976), there are only one fuse left (three needed to run the machine), the machine is like a cobra - it "bites" immediately when anything goes wrong. Not a very nice situation when fuses are hard to find. Also, this machine produces components for large roller bearings. No components - no bearings. It is not allowed to stand still.

This time, we tried to make the machine run with low voltage. Used a 27 V power supply from an Ericsson telecom station and low amp fuses (35 A). Ran several cycles when, suddenly, there was an intermittent fault (see attached picture).

Took everything apart again, including the large (ca 300x200 mm) control board (no diagrams). Retightened every single connection. Resoldered every single solder joint. (I didn't have to do it myself - thanks)

Found three 200 A semiconductor fuses in a distant steel mill and had them sent by taxi.

Started again last Wednesday and machine is still running. Several thousand cycles so far. But, you never know. I think that we can say after a month or two if it is a sucess or not.

BTW: Anyone got Ferraz 200 A semiconductor fuses in stock? Type is V300122 but we can use any fuse with same rating. We have already rebuilt so we can use either cartridge or stud types.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[electricuwe]

Hi Gunnar,

Use of OEM labeling on the power semiconductors allows the supplier to make tempting offers on new equipment and to earn his money on the spare parts. Keeping schematics secret is usefull in supporting this strategy.

As the inverter is still running at low voltage, thyristors are still o,k., but continueing troubleshooting is risky with no spares available.

As in this frequency range fast thyristors have already been replaced by IGBTs at least 15 years ago most power semiconductor companies have discontinued such parts already. Even if either the OEM or others made a lifetime buy at discontinuation, it is quite unlikely that there are components left on the market.

There are some smaller companies that specialised in fast thyristors and that serve the market longer that the large suppliers, Westcode beeing the best known. So try to compare the thyristor package in use in your inverter with Westcode products and estimate if devices in comparable package might work in this circuit. Beside voltage and current, turn-off time is the most important rating in this case.

Keeping such old equipment in operation as a vital part of a production line without having sound stock of spare parts and detailed knowledge (at least schematics) is very risky.

Had to help customers in similar situation in the past and sometimes the only solution was to completly refurbish with different parts or to supply totally new equipment. But suggest your customer to insist on a list of spare parts with semiconductor type designations, that allow to source directly from the semiconductor supplier, and of courese, schematics.

When comparing quotes, if you can afford this approach in regard of time, always compare quotes including some stock of spares.

 

[ScottyUK]

Gunnar,

RS Components in the UK have a fairly good range of semiconductor fuses ex-stock, certainly there are several 200A ratings in different styles. Delivery is usually next day to the UK, might take a little longer to Sweden.

Agree with Uwe that Westcode are a good bet for inverter-grade thyristors, Powerex is also a likely source although I have never managed to get a fast delivery out of them.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Skogsgurra]

Thanks to both of you!

Our customer has slowly realized what a pinch he is in. He will discuss new equipment with a supplier tomorrow, Monday.

But delivery time is said to be around six months - at best.

We finally found a diagram for the control board. The customer signed an NDA before he got it. This was long time ago and it had obviously been kept under lock somewhere ever since.

We are contemplating to offer a rebuild where we replace the existing 16 thyristors (yes 16, two in series and parallel for each leg in the H bridge) with modern modular components. Or change to IGBTs - which will require latched gating and gate drivers. But seems a lot safer than what they have now. And spare parts available.

It is interesting to see these old dinosaurs again. My first inverter was a Siemens motor drive for a nylon fiber factory back around 1970. It used auxiliary thyristors for turn-off and consumed loads (yes, literally) of fuses. In a flash-back, I saw the guy that came with a wheel-barrow full of fuses one afternoon. Enough to keep the factory running for another week. Eventually, that factory closed. We never got it running properly.

Technology has come a long way in 30+ years. And still, GTOs in the USD 10k/each price range are changed at a rate of two or three every month in a nearby paper mill. Supplier tells customer to buy new equipment. History repeats.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

Since it's limping along maybe you can take the time to find a used newer machine..

Perhaps the factory knows of one.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

As the French say: Limping - Le mot juste!

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Pud]

Skogs,

If you do ever need stuff from RS, we have an account with them - happy to order for you and send them over.

Harry

http://www.tynevalleyplastics.co.uk

 

[Skogsgurra]

That's nice Harry! I will surely remember.

We are now waiting for decisions to be made by the customer. Rebuild (IGBTs) or trying to get new thyristors or installing entirely new equipment is what it is about. Might take some time before we have the decision. In the meantime, it looks like "the child was saved from drowning - no fence needed".

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...