Diagnosing and Repairing a Faulty Inverter: Tips from a Scientist

[Guineafowl]

So 'we' (you) might be getting somewhere here. I should have clarified the transistor arrangement, and should also remind you that the schematic refers to an earlier model, although by your reckoning the 2N2222As would not have worked even on that model.

At Q29 there was an A42 (http://cdn2.boxtec.ch/pub/diverse/A42.pdf) - Max voltage 300V
At Q28 there was a 2N2907A - this appears to be OK.
At Q31 there was an S9013, max voltage 40V
At Q30, another 2N2907A.

I'm thinking...

1. The 2N2907A PNPs are correct.
2. Some eejit put in the S9013, a general purpose NPN, where there should have been an A42.
3. The S9013 has been clamping its own Q34/35 as well as Q38/39, negating the efforts of the correct A42.
4. By replacing the S9013 and A42 with two 2N2222As, I have also been an eejit.

I will try to get some A42s, or failing that, will follow Jim's recommendation (thanks for that, by the way Jim.)

 

[Guineafowl]

Update -

Ordered these:

http://docs-europe.electrocomponents.com/webdocs/0f9b/0900766b80f9b314.pdf

Fairchild KSP42BU

They come in convenient packs of 10. I think they'll be OK - there's a 42 in the name, at least...

 

[Tom.G]

Fairchild KSP42BU

Those should do the job. They appear to be clones of the MPSA42 that was in Q29. The 2N3439 that @jim hardy suggested should also work, they are very similiar, perhaps a little more robust; just be aware that the 2N3439 is in a TO-5 TO-39 package (metal can), not the plastic TO-92 package.

p.s. Since the IRF640 is also penciled in, I wonder what the originally part was. Probably doesn't matter if it works!

 

[jim hardy]

perhaps a little more robust;

I sawed one open. Collector was bonded right to the metal can so it should have better thermal capability than the plastic ones.. Feel of them, if not hot to touch you'll be fine with the little guys. And you already know about EBC vs ECB packages...

Been fun watching you guys. Congratulations on your progress !

 

[Guineafowl]

This is the cause of all the trouble. I replaced the whole set with IRF640N.

It's always the output MOSFETs - couldn't you build a more robust inverter with an IGBT or just a much larger MOSFET with proper heat sinking? Isn't a row of piddly little units bolted to the case rather fragile? Seems to be from my experience - most inverters I've repaired (it's an odd hobby) have failed in this way. Some just need new MOSFETs, others have lost some of the drive circuitry.

Another thing - is there a better way to isolate the driver ICs and other semiconductor devices from a rogue short or overvoltage condition? I suppose surface-mount fuses would fatigue, but what about a MOV or two? SCR crowbar? These mass-produced inverters seem to lack provisions to a) stop the MOSFET from failing and b) stop the fault from propagating upstream. In contrast, even a cheap SMPS wall-wart will sit there chirping with the output dead shorted, and manage not to blow itself to bits.

 

[jim hardy]

Another thing - is there a better way to isolate the driver ICs

some folks use transformers

 

[Asymptotic]

Another thing - is there a better way to isolate the driver ICs and other semiconductor devices from a rogue short or overvoltage condition?

I've never used any, and they are on the expensive side, but Altech offers a line of 'Z trip' circuit breakers supposedly capable of protecting semiconductor power controllers. Would one of these open quickly enough to protect a MOSFET bridge from a load short or severe overload?

 

[Tom.G]

Would one of these open quickly enough to protect a MOSFET bridge from a load short or severe overload?

Doesn't look encouraging. The safe operating area (SOA) of the IRF640 peaks around 30A and 10ms. The breaker curve shows a trip time anywhere from 10ms up to 20 seconds for a 30A load. Ouch!

@Guineafowl Could you verify C73 and C74 values, both on the board and on the schematic? Also inspect them for any signs of failure or prior replacement.
They are bootstrap capacitors used to supply gate drive to the high side MOSFETs. If you have a Capacitance Meter to measure them, check both capacitance and ESR or DF (Dissipation Factor). Do so while the associated MOSFETs are out of the circuit. If they are faulty or the wrong value they could destroy their associated MOSFETs at heavy loads.

I suspect them because the overall circuit is a symmetrical H-bridge and those caps should be the same; but the schematic shows different values (schematic marked-up?).

Were all eight MOSFETs (Q34 thru Q41) the IRF640?

 

[Guineafowl]

Doesn't look encouraging. The safe operating area (SOA) of the IRF640 peaks around 30A and 10ms. The breaker curve shows a trip time anywhere from 10ms up to 20 seconds for a 30A load. Ouch!

@Guineafowl Could you verify C73 and C74 values, both on the board and on the schematic? Also inspect them for any signs of failure or prior replacement.
They are bootstrap capacitors used to supply gate drive to the high side MOSFETs. If you have a Capacitance Meter to measure them, check both capacitance and ESR or DF (Dissipation Factor). Do so while the associated MOSFETs are out of the circuit. If they are faulty or the wrong value they could destroy their associated MOSFETs at heavy loads.

I suspect them because the overall circuit is a symmetrical H-bridge and those caps should be the same; but the schematic shows different values (schematic marked-up?).

Were all eight MOSFETs (Q34 thru Q41) the IRF640?

C73:
10uF 50V. ESR: 1 ohm. Capacitance: 9.8 uF.

C74:
10uF 50V. ESR: 0.8 ohm. Capacitance: 9.9 uF.

Both measured out of circuit.

Yes, all MOSFETs were IRF 640B.

The schematic is a bit of a double-edged sword, isn't it?!

 

[jim hardy]

I've never used any, and they are on the expensive side, but Altech offers a line of 'Z trip' circuit breakers supposedly capable of protecting semiconductor power controllers.

Wow! i never before saw a breaker faster than 10 milliseconds. There's a limit to how fast the mechanical parts can move, and it takes finite time to quench the arc.

Usually one has to protect semiconductors with a current limiting fuse.
Shawmut form 101 Amptrap is an old standby. They're filled with sand to brute force the arc current to zero rather than wait for the next sinewave zero crossing.
I tested 15 amp ones in my plant and found at 10X overload they'd open in usually a millisecond sometimes two, depending on how close to sinewave peak the fault occured.
http://www.ferrazshawmutsales.com/pdfs/A100P.pdf

i couldn't find a recommended fusing I²T for those mosfets, like they provide for big SCR's.

old jim

 

[Guineafowl]

So a possible overcurrent protection would be Jim's Amptraps, but what about overvoltage? Wouldn't an SCR or MOV work well in conjunction with a fuse?

Liquid-cooled MOSFET heat sink?

Now, I have so much confidence in TomG's diagnosis of A42 deficiency, that I've taken on another project:

A Waeco MSP1512 pure sinewave inverter. RRP £900, bought as faulty for £100. Don't worry, it's working for the moment. The blade fuses on the left had blown - 5 standard 40A fuses in parallel. The fuses were soldered in, so needed two soldering irons to remove them as the traces are very hefty, and that dirty great ?inductor was stealing the heat.

It's currently being soak tested at idle to see if any good reason for the blown fuses emerges. It's happily putting out a nice sine wave at 222V rms.

As far as I can find out, this unit has multiple oscillators which, when combined in a Fourier-type way, and when filtered through a large LC, result in the near-perfect sine wave.

 

[jim hardy]

As far as I can find out, this unit has multiple oscillators which, when combined in a Fourier-type way, and when filtered through a large LC, result in the near-perfect sine wave.

wow !

try a search on

magic sine waves don lancaster

old jim

 

[Guineafowl]

wow !

try a search on

magic sine waves don lancaster

old jim

What an odd character! I assume there's a practical reason the system isn't used widely. I would say that if you have a sound engineering idea there's no need to concoct a website that's set out like a pyramid scheme or faith-healer's advert.

Was your "wow" for the brilliance or the idiocy of my deduction?

 

[jim hardy]

Was your "wow" for the brilliance or the idiocy of my deduction?

For your discovery of that inverter.

I once had a Hewlett Packard function generator that synthesized functions that way.

Been years since i looked in on Lancaster's "Magic sinewave" pages and it does look like he's gone a bit off kilter... Sorry ... he used to be imminently practical and authored a lot of "how to" books.

Here's today's version of his old introduction

http://www.tinaja.com/glib/msintro1.pdfGood luck with that inverter ! Teach us how it works ?

old jim

 

[Guineafowl]

For your discovery of that inverter.

I once had a Hewlett Packard function generator that synthesized functions that way.

Been years since i looked in on Lancaster's "Magic sinewave" pages and it does look like he's gone a bit off kilter... Sorry ... he used to be imminently practical and authored a lot of "how to" books.

Here's today's version of his old introduction

http://www.tinaja.com/glib/msintro1.pdfGood luck with that inverter ! Teach us how it works ?

old jim

Yes, the older link is more sober!

My best guess is that the inverter uses that enormous inductor as part of a resonant filter to shape the boxy output into a more pleasing sinewave. Maybe some PWM control to eliminate harmonics? I'll scope the output for you when I next get a chance. I imagine TomG might weigh in at some point.

In the meantime, I'll await the high voltage transistors for the original inverter...

 

[Asymptotic]

Doesn't look encouraging. The safe operating area (SOA) of the IRF640 peaks around 30A and 10ms. The breaker curve shows a trip time anywhere from 10ms up to 20 seconds for a 30A load. Ouch!

Ouch indeed!

Wow! i never before saw a breaker faster than 10 milliseconds. There's a limit to how fast the mechanical parts can move, and it takes finite time to quench the arc.

I remember being surprised as well after finding them in the Altech catalog. Went looking for, but couldn't find an in-depth explanation how they operated, and didn't try any, but I've used plenty of form 101 Amptrap fuses protecting everything from SCR DC drive bridges to resistance heater solid-state contactors.

Just jawing here, but my opinion of fuses versus circuit breakers for semiconductors balances on a razor edge. When a fuse blows, it takes time to get a replacement from stores, lock out the circuit, install the replacement, and re-power the circuit. During that time the semiconductor has cooled significantly, and increases it's chances of survival if the replacement fuse blows. A second fuse failure also prompts the technician to dive deeper into why, if only because there isn't an infinite number of replacements in bench stock.

The same thing that makes a circuit breaker attractive - it's ability to be quickly reset - can also be it's downfall as a protector of semiconductors. It is difficult for a tech to resist the temptation of making multiple reset attempts, the die doesn't have enough time to cool, and usually ends up in a splash of silicon.

 

[jim hardy]

Just jawing here, but my opinion of fuses versus circuit breakers for semiconductors balances on a razor edge. When a fuse blows, it takes time to get a replacement from stores, lock out the circuit, install the replacement, and re-power the circuit. During that time the semiconductor has cooled significantly, and increases it's chances of survival if the replacement fuse blows. A second fuse failure also prompts the technician to dive deeper into why, if only because there isn't an infinite number of replacements in bench stock.

That choice as you say depends on the application.
We once had to co-ordinate the branch circuit breakers in an inverter supplied. load panel with the inverter's own internal overcurrent protection.
In that case we insisted on breakers so that somebody couldn't "just try a bigger fuse". They'd then lose the whole panel when the inverter tried to blow that fuse too and its internal self protection shut it down . ( Loss of that panel trips the plant .) We found some fast enough ten amp breakers for branch circuit protection.
So there's a place for both.

My best guess is that the inverter uses that enormous inductor as part of a resonant filter to shape the boxy output into a more pleasing sinewave.

There exist "Ferroresonant" inverters basically a ferroresonant transformer fed by line frequency switched DC . They have that nice inherent current limiting of ferroresonant transformers and are simple. Only downside is voltage of course is a strong function of frequency , so if they're synched to the grid then during a severe upset they might sag a bit.
You can spot them by the extra winding on the transformer and the precision capacitor connected to it.

 

[Guineafowl]

It works! The KSP42s did the trick. I have no 115V devices (almost all UK appliances are 230V) but I put a clamp meter on the 12V cables and estimated 1A per 10W outut power. The wife's hairdryer, a useful selectable load, plus a hammer drill, put the load up to a good 1200W (120A on the input) which the unit coped with well for a good 10 minutes. I'm calling it a fix.

By some odd convention (wet weather?), some specialist building site power tools are 110V, the intended use of this unit.

TomG, Jim Hardy and others, I'm obliged under British social law to take you all for several pints at the local pub.

I've also dropped an email to the head engineer at Durite pointing out the problems with the schematic, in case some poor sap lacking a TomG or similar tries to troubleshoot one of these.

Cheers, or slàinte as we say here. On to the next project.

 

[jim hardy]

pints at the local pub.

Sounds great but all i did was admire from a distance you folks's work.

Congratulations on success.

One learns a LOT from troubleshooting, eh ?.

 

[Guineafowl]

One does indeed learn a lot from troubleshooting. Thankfully, this new inverter doesn't need too much...

Here's the transistor array. The 12 units on the right receive a 0-12V square wave at 31 kHz. I'm guessing these are the input push-pullers.

Interestingly, the four left-hand transistors receive a gate drive of 0-220V peak in a sinusoidal pattern at 50 Hz. It looks like the sine output is already there before the output is filtered.

A closer look at that daughter board (bottom) - conformal coated on the back to keep out fiddlers like me! Note clock crystal. There's a microcontroller at work here, shaping the gate drive to the output MOSFETs. Maybe Lancaster magic sine waves?!

The output. Note slight wibbliness at zero crossing point.

As I say, troubleshooting is a good teacher, but I'm glad this one doesn't need any (apart from blown fuses). Having to tackle this unit would either teach me a awful lot or turn me to drink.

 

[Tom.G]

TomG, Jim Hardy and others, I'm obliged under British social law to take you all for several pints at the local pub.

Hey folks, how many are up for a group pub crawl across the pond?

 

[jim hardy]

Hey folks, how many are up for a group pub crawl across the pond?

If i can get there !

Anywhere near the "Mousehole Forge" in Sheffield ? I had one of their anvils made about 1823...gave it to a friend who collects them.

I think it's a pub now ,

 

[Guineafowl]

My local, where I grew up - the Farmer's Arms. A 15th century coaching inn which sadly had a fire earlier this year. A common hazard with these traditional thatched roofs. Anyway, pub crawl starts here after the refurbishment...

 

[shjacks45]

Seriously, 500 pounds? Most retailers would be less than $99 US. I think England has Walmarts. I haven't seen a 4093 in a power supply built since 1993. Power supplies (Inverters) is the most likely to fail on tech equipment. If your Inverter is that old, it needs to be replaced as part of prudent maintenance.

 

[Guineafowl]

Seriously, 500 pounds? Most retailers would be less than $99 US. I think England has Walmarts. I haven't seen a 4093 in a power supply built since 1993. Power supplies (Inverters) is the most likely to fail on tech equipment. If your Inverter is that old, it needs to be replaced as part of prudent maintenance.

Holy thread revival!... etc

Inverters vary hugely in quality and price, with the cheaper ones having fairytale power rating and short lifespans. This old unit was repaired, tested, and sold to the highest bidder long ago.