repairing broken gallien kreuger - help!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi, finally getting round to repairing my Gallien_Kreuger 700rb MkII - it's a ~500W solid state bass head. A friend of mine was playing through it years ago and it stopped working - no noises, no smoke, no drama - just turned off. I opened it up and found it was blowing fuses but never had the time or patience to get my hands dirty - until now

After testing that the circuit worked fine with the secondary disconnected, I diode tested the output devices - which measured up ok. The I went looking at the power supply bits and noticed the potting compound on one of the bridge rectifiers was all cracked, so i measured it and sure eough it had blown - but why? So I replaced that, turned it on - fuse stays intact, I connect a dummy load - still good, I put a signal through it but just as im trying to get a scope to the right settings I notice some a kind of faint hot smell and the fuse goes again.

this time its one of the output devices, the bridge rectifier still measures up fine... I remove the heatsink and tke the output device out and turn it on again, it turns on fine and doesn't blow fuses but some of the transistors in the drive circuit are getting really hot...

so, my question is, are these devices gonna get hot if I remove one of the output devices, or is this an indication that something else is wrong? Should I replace the output device before testing further or try and troubleshoot why these transistors are getting hot? I have attached a picture of the power amp schematic, the bridge rectifier I took out is ringed with green, the output BJT is pink, the two that are getting hot are brown.

I'm fairly good with electronics but I have very little experience with amplifiers, so go gently, but not too gently :)
 

Attachments

  • gallien-krueger_700rb-ii_sch (1).jpg
    gallien-krueger_700rb-ii_sch (1).jpg
    807.5 KB · Views: 199
Bad insulator under the output transistors can overload drivers & PS too. Your meter diode check doesn't pretend to produce enough voltage to really stress dielectric.
Transistors can be stressed by a load 1/4 phone plug slightly out of the hole, or dirt on the fan or heat sink. Following failed transistors can stress drivers or predrivers or op amp.
Rectifier blown could have been a sign of load transistor bad. A fuse afterwards is good, the new PTCR resettable ones are tiny, cheap and work up to 3 amps and 72 volts. You turn the power off to reset them. You can drill holes through the board and cut a land or remove a rectifier wire to install one.
Forthermore, if an output transistor went, I'd do a Iceo test on each one. They can be marginally okay on diode check @ 2 v but blow up under full voltage - I exploded a couplethat passed diode test before I discoverd about power line limiters - light bulb box, or for >1000 w amps, a room heater element in series with the AC mains going into the amp.
You do an Iceo test with a medium voltage power supply, a current limiting resistor, and the milliamps scale of your DVM. Warning voltage over 24 across the heart can stop it. Current through rings or other jewelry can burn your flesh to charcoal. No rings, one hand at a time when voltage is on. Use clip leads.
You put say a 30 V copier/fax/printer wall supply from the charity resale shop, in series with say a 22k or 47 k resistor, then to the C of the transistor under test. E of the transistor goes to the other lead of the power supply. + to C for PNP, - to C for NPN. If the Iceo is less than on the datasheet, okay. More, it is zorked.
You can run the amp for test with only one pair of output transistors but don't go very loud or use very low resistance load.
I use mica heat sink insulator kits from newark, multicomp brand. The 1" diamond transistors are TO3, the plastic ones with a tab are TO220 if small or TO247 if big. The kits come with new little ferrules to put through the board around the transistor leads if you drop one under the table. Also newish heat sink compound is a good idea. Warning this compound is poisonous, no fingers in mouth nose or eyes until washing up.
Replacement output transistors should all be the same part number age and preferable same batch. Vce matching is useful eliminate current hogging by the high gain ones.I can't see your emitter resistor values, but if less than 0.5 ohm matching becomes more important. I use a 12 v car battery charger with parallel ripple smoothing capacitor for this test. A 10 ohm resistor from supply to the collector for TO3 or TO247, a 100 ohm resistor to base, measure the Vce at this gain of 10. Write voltage on case with sharpie. If you have a few extras, you can put the outlier voltages in the parts bin. In my case of 24 transistors 4 were a little higher gain (.01 v lower Vce) than the rest so those became the oddballs and I installed 20 output transistors with .13 & .14 Vce.
Best of luck.
 
Last edited:
well, i just though id replace ALL the output devices to be sure...

and the two transistors ringed in brown in the picture i posted still get very hot just a few seconds after turning on. They're usually under the heat sink but im don't think this is normal, all the same.

so, what might be wrong aside form the output transistors to cause this? none of the other transistors in the drive circuit get hot
 
Did you replace the insulators under Q25,26,27, 22, 23,24 at the time you replaced the transistor? If translucent mica, that stuff gets old. Plus new heat sink grease if it is mica. If it is that grey opaque silicon wafer, not so important to replace it every time.
If C22 or C14 was leaky, that would cause Q14 or Q11 respectively to heat up. There is a voltage marked under the red circle. Make sure the voltages are as specified. Also C10 or C16 leaky could disturb that stack. And there are a bunch of zener diodes around there, those are high failure parts, too. Make sure the difference between the two ends of the diode are what it says on the print.
To avoid blowing up your transistors you may need to set up clip leads for your tests, then flip the power on for a couple of seconds then back off again. VOM read faster than DVM, you may not have time for a DVM average over 3-4 seconds to make a valid reading before things blow up.
Another possiblity is bad solder joints. Make sure the component leads at the end of each line in that stack between Q11 and Q14, are actually connected. Plus if D6, D7, D1 or D12 regular diodes are blown, that could mess things up too. Make sure regular diodes are either 0.7 v forwards, or a negative voltage backwards from one end to the other.
 
Last edited:
OK so I'm back after a long(!) break trying to get this amp up and running...

if anyone is still here, I have tried to find the problem with the amp but no joy

in reply to Indianajo, the insulation is facing up into the heatsink so that when I take the sink off (as I have, to work on the amp) the insulation is removed and i am looking down onto the metal backs of the transistors, so I guess that's not the problem. I tested C14 and C22 with a DVM - seemed to be fine. All voltages listed on the schematic are pretty much the same except for the collector resistors for the hot transistors - these start off OK but rise over time. Forward biased zener diodes are fine but I just realised i didn't check the zener voltages so I will check those...

it seems funny to me that the same driver transistor on BOTH the push and pull side is getting hot - seems to suggest to me that what they are feeding into is the problem. Is there a way to disconnect these from the rest of the amp to check they are OK? If i can in the problem down to the drive part or the power amp part that might help...
 
I fixed a Woodson amp that kept tripping the thermal cutoff. These solid state power amps are susceptible to thermal runaway. One of the diodes in th efront end between the two sides is usually mounted to the heatsink. It automatically adjusts the bias between them based on the temperature. If that isn't working it will slowly heat up hotter and hotter and may or may not fail. The Woodson amp had a cutoff bi-metal switch that shut off the AC power when the main heatsink got to hot. I traced it to a small diode mounted in a springy heatsink mounted to the main heatsink. The small clamp on the diode was loose so it wasn't sensing the temp. I got some thermal epoxy and it fixed the problem.

These circuits have high power transistors direct coupled to the pos voltage bus, and another set connected to the neg bus. Then they are connected together. The from end biasing circuit is designed to just barely turn both sides on to minimize crossover distortion. If they are turned on too hard, you have a direct path from the pos bus to the neg bus and nothing to limit the current.

I can't read the part numbers, but the biasing circuit is just to the right of the differential pair on the left side. Are any of those parts mounted to the heatsink?
 
CapnDenny1 - That would make a lot of sense, I'm testing this without the heatsink, two of those transistors that are normally under the heatsink keep getting hot and directly between them is the one of the transistors from the section that you mention

In the meantime I got desperate and started to replace all the transistors under the 'sink I had not yet replaced, before realising that some of them are obsolete. The replacements parts I have for these are as follows -


Q11: 2SC3502 -> KSA1381
Q14: 2SA1380 -> KSC3502
Q3,Q7: 2SA1837 -> MJE15035
Q13,Q16,Q17: 2SC4793 -> MJE15034

can anyone give a thumbs up to those?
 
Hi, unfortunately I cut the old ones out in desperation... in future I hope to be less impetuous

I checked the emitter voltages on the VAS the other day and they looked OK but I will check again and see if they are in 5mV. Which emitter-resistor voltages will R28 affect? R33 / R26?

I have very little experience with doing this, I have been turning the amp on for a couple of seconds before it overheats to make measurements, the problem really is that those transistors are overheating but I now understand this may just be normal behaviour without the heatsink
 
ok, i am back to fixing this amp again. hoping to finish it this time!

i put the amp back together and found that something was overheating - it turned out to be transistors Q25,Q25 and Q2.. i think i left out the insulating washer on Q16 so I replaced Q16 and took Q27 out and measured it but it seems to measure up fine. Those transistors still get very hot immediately after turn on and there is an offset of about -10v on the output

how can I test to find out the problem?

also, if there are any web resources for fixing these amplifiers please let me know - i am fairly experienced with electronics but not these kinds of amplifiers, but with a little reading I might be able to get up to speed
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.