Hi, I have an Eagle A6400 amplifier (circa 1975) that has had quite a major blow, I need help please with finding suitable part substitutes and identifying resistors that have burnt. output transistors are all bad (TO3 type) and there is little hope of a schematic
Audio Karma Slideshow by grabread | Photobucket
Pictures of amplifier board here
Any help gratefully received, Thank You
Audio Karma Slideshow by grabread | Photobucket
Pictures of amplifier board here
Any help gratefully received, Thank You
See the old amp repair manual thread vintage amplifier repair/upgrade manual - diyAudio
Don't forget the light bulb box in series with the AC line testing after "repairs". My light bulb is in a grounded metal box with a circuit breaker. You check the speaker output is not whanging to one rail or the other before getting off the light bulb. Wear safety glasses, used transistors tend to explode. Ones where Vcb <.5v on the DVM are damaged IMHO.
Rail voltages will help you pick output transistors. In the US I use On semi ones. MJ802/MJ4502 would usually be good enough for anything that old. If there are no pnp output transistors, might be all 2n3055/mj15015.
Use new mica washers & grease, I like the ones I'm getting from newark the multicomp kits with plastic ferrules included.
After working replace all electrolytic caps. 40 years is too long on rubber parts. Clean the heat sink and if there is a fan, replace it.
Usually there is one emitter resistor not burnt. Measure it I tend to favor .39 or .47 ohm over anything lower for better transistor Vceon matching, although lower resistors can allow more power for a little while. Calibrate your meter with a known value resistor, they are not usually accurate below 10 ohm. Your picture shows one .27 ohm resistor, a bit unstable choice IMHO.
Look for obvious things like diode open or 0 ohms, capacitors 0 ohms, resistors way off value.
I replaced 124 parts in a PV-1.3k. Most were salvage out of old ATX power supplies, not the output and driver transistors of course. It is a fun hobby but not productive with class D amps going at $50 for 100 watts stereo.
Mine was blown all the way back into the input op amp and DDT current amp. Root cause of blow ups on one channel was in fact a bad solder joint on the input op amp from the factory in 1996. If your amp has been reliable since 1975 you won't have that problem. Just the joints you make yourself (I have trouble too).
I don't need a scope or a transistor matcher, but I do trace AC music in some cases with a VOM with a 2 VAc and 20 VAC scales. Oscillations will show up as AC voltage with no music going in and input shorted. I match transistor with a 12 v power supply, a 10 ohm resistor to collector, and a 100 ohm resistor to base. Match Vce as much as possible, and buy a few spares in your OT collection to waste the outliers. I had two .11 volt transistors in 24, the rest were .13 or .14 volts Vce. I needed 20 output transitors, wasted 2.
Buy transistor from an authorized distributor like newark(farnell), digikey, mouser, alliedelec, avnet, RS in UK,
don't waste your time on e-bay trash tested with a silk screen and a spray painter. No used e-caps either. surplus resistors can be okey, especially the 8 ohm logs (200 watt resistors) you need to test the final wattage performance before putting it on speakers. Nearing the end I energize with a transistor radio on a rock station, for the beats in the AC pointer prove it is music, not oscillation. I initially use speakers protected by 4700 uf back to back capacitors series in case a DC event occurs unexpectedly. (I use salvage e caps for that).
Don't forget the light bulb box in series with the AC line testing after "repairs". My light bulb is in a grounded metal box with a circuit breaker. You check the speaker output is not whanging to one rail or the other before getting off the light bulb. Wear safety glasses, used transistors tend to explode. Ones where Vcb <.5v on the DVM are damaged IMHO.
Rail voltages will help you pick output transistors. In the US I use On semi ones. MJ802/MJ4502 would usually be good enough for anything that old. If there are no pnp output transistors, might be all 2n3055/mj15015.
Use new mica washers & grease, I like the ones I'm getting from newark the multicomp kits with plastic ferrules included.
After working replace all electrolytic caps. 40 years is too long on rubber parts. Clean the heat sink and if there is a fan, replace it.
Usually there is one emitter resistor not burnt. Measure it I tend to favor .39 or .47 ohm over anything lower for better transistor Vceon matching, although lower resistors can allow more power for a little while. Calibrate your meter with a known value resistor, they are not usually accurate below 10 ohm. Your picture shows one .27 ohm resistor, a bit unstable choice IMHO.
Look for obvious things like diode open or 0 ohms, capacitors 0 ohms, resistors way off value.
I replaced 124 parts in a PV-1.3k. Most were salvage out of old ATX power supplies, not the output and driver transistors of course. It is a fun hobby but not productive with class D amps going at $50 for 100 watts stereo.
Mine was blown all the way back into the input op amp and DDT current amp. Root cause of blow ups on one channel was in fact a bad solder joint on the input op amp from the factory in 1996. If your amp has been reliable since 1975 you won't have that problem. Just the joints you make yourself (I have trouble too).
I don't need a scope or a transistor matcher, but I do trace AC music in some cases with a VOM with a 2 VAc and 20 VAC scales. Oscillations will show up as AC voltage with no music going in and input shorted. I match transistor with a 12 v power supply, a 10 ohm resistor to collector, and a 100 ohm resistor to base. Match Vce as much as possible, and buy a few spares in your OT collection to waste the outliers. I had two .11 volt transistors in 24, the rest were .13 or .14 volts Vce. I needed 20 output transitors, wasted 2.
Buy transistor from an authorized distributor like newark(farnell), digikey, mouser, alliedelec, avnet, RS in UK,
don't waste your time on e-bay trash tested with a silk screen and a spray painter. No used e-caps either. surplus resistors can be okey, especially the 8 ohm logs (200 watt resistors) you need to test the final wattage performance before putting it on speakers. Nearing the end I energize with a transistor radio on a rock station, for the beats in the AC pointer prove it is music, not oscillation. I initially use speakers protected by 4700 uf back to back capacitors series in case a DC event occurs unexpectedly. (I use salvage e caps for that).
Last edited:
Thanks for the quick reply, output transistors are d371 and b531 and the obviously blown transistor is a 9013g
Those are private label numbers and don't mean anything.
I pick transistors by package type, source voltage (Vceo has to be higher) wattage.
MPSA06/56, BD139/140, TIP41c/42c, MJE15032/33, MJ21193/94 usually cover everything but jfets and mosfets.
I pick transistors by package type, source voltage (Vceo has to be higher) wattage.
MPSA06/56, BD139/140, TIP41c/42c, MJE15032/33, MJ21193/94 usually cover everything but jfets and mosfets.
Last edited:
The first two would seem to be 2SD/2SB types though. The datasheets should give you an idea of what to replace them with (or look what other people have used as a substitute). And it looks like the 9013G is a real thing, too. (Is that the bias spreader transistor in the middle between the TO-3s? That one isn't actually terribly critical just as long as thermal tracking is good and you adjust the bias. People tend to use TO-126s for that these days.)
It looks like tracing out the board should be quite doable. It is an almost mirrored layout with partnumber(LHS) = partnumber(RHS) + 1 (or - 1 in some cases, apparently?), so you should also be able to identify almost all the resistor values.
Just by layout and heatsinking alone, I'd be guessing that Q301/303 are the input LTP (usually pnp), Q305 is the VAS transistor, and Q313 and its complement over on the other side are drivers. Pretty standard "Lin topology" amp of its time there (actually quite modern, as a lot of manufacturers were still using diode string biasing rather than a Vbe multiplier). Q309/311 may just be overcurrent protection.
It looks like tracing out the board should be quite doable. It is an almost mirrored layout with partnumber(LHS) = partnumber(RHS) + 1 (or - 1 in some cases, apparently?), so you should also be able to identify almost all the resistor values.
Just by layout and heatsinking alone, I'd be guessing that Q301/303 are the input LTP (usually pnp), Q305 is the VAS transistor, and Q313 and its complement over on the other side are drivers. Pretty standard "Lin topology" amp of its time there (actually quite modern, as a lot of manufacturers were still using diode string biasing rather than a Vbe multiplier). Q309/311 may just be overcurrent protection.
Thanks sgrossklass, D307 is also blown can you suggest a suitable replacement? You can probably tell my repair skills are still in their infancy.
Tiny diodes rectifiers 1n4148 (out of dead ATX supply)
Fatter lead rectifiers any similar sized lead rectifier from same source or 1n4007.
Real fat lead rectifiers mr805 or similar sized lead rectifiers standing up in ATX supply.
Zener diodes have to be replaced same voltage and bigger wattage.
Have to draw out the circuit and see what it is doing. Hint - zener diodes go to the base of a transistor, also through a resistor to ground or a supply rail. Zeners mostly have a wiggly line on the symbol on the board.
Fatter lead rectifiers any similar sized lead rectifier from same source or 1n4007.
Real fat lead rectifiers mr805 or similar sized lead rectifiers standing up in ATX supply.
Zener diodes have to be replaced same voltage and bigger wattage.
Have to draw out the circuit and see what it is doing. Hint - zener diodes go to the base of a transistor, also through a resistor to ground or a supply rail. Zeners mostly have a wiggly line on the symbol on the board.
Thanks Indianajo, looks like those atx supplies are very useful for parts also lots of bits of wire to use. Would you have any idea what value the resistor is in my photo's
many thanks
many thanks
Yeah, new ATX supplies are $20 here plus freight. Almost worth it if you didn't blow up a PC supply already & save it. The heat sink circuit breaker & fan are worth $12.
Good diodes are ~.62 v forwards, 1999 backwards.
For the resistor you'll have to draw out the position in the circuit. 12???? If we knew the position we could guess.
For a sample circuit look at the build document for the honeybadger up top sticky thread. It has the names of all the transistors on the schematic.
Tiny TI is a free circuit drawing utility for windows. Linux op system has its own, part of the ubuntu suite.
Good diodes are ~.62 v forwards, 1999 backwards.
For the resistor you'll have to draw out the position in the circuit. 12???? If we knew the position we could guess.
For a sample circuit look at the build document for the honeybadger up top sticky thread. It has the names of all the transistors on the schematic.
Tiny TI is a free circuit drawing utility for windows. Linux op system has its own, part of the ubuntu suite.
I'm getting a better idea of the parts I need now, for the 2SD371/ 2SB531 MJ802/MJ5402 and for the 9013g MPSA06 , the small diode 1N4148.
The 260 ohm resistor seems to be rare I have 250 or 270 ohm in stock so up or down there?
I'm still struggling with that burnt resistor though any ideas? 1.2, 12, 120 , or 120k ? start with the highest first and work down or vice versa
any suggestions welcome please
The 260 ohm resistor seems to be rare I have 250 or 270 ohm in stock so up or down there?
I'm still struggling with that burnt resistor though any ideas? 1.2, 12, 120 , or 120k ? start with the highest first and work down or vice versa
any suggestions welcome please
- Status
- Not open for further replies.