Hi everyone,
I have a Quad 303 (serial 45343 - schematic here) that underwent an upgrade after burning up its outputs. Here's the subs I used, and what was originally in there:
TR1 MJ15003 2N3055
TR2 MJ15003 RCA 38494
TR100 BC560 BC214C
TR101 BC549B BC184
TR102 BC546B C1007
TR103 BC546B C1007
TR104 BC556B E5458
TR105 BC461 38496
TR106 BC141 38495
TR107 BC549B BC184
TR3 MJ15003 2N3773
TR200 BC556B C1593
TR201 BC141 40409
I fired it up on a 40W light bulb limiter and everything looks good initially. I adjust the power supply to 67V no problem, I adjust the half-rail to 33.5V no problem, but when I begin to increase the bias, it quickly jumps from around 5mV up to over 200mV and the limiter bulb starts pulsating brightly, at which point I back off the bias to zero again. I've replaced the diodes and the bias trimmers so the wiper isn't going open-circuit and sometimes I can get both L & R channels biased to around 7mV but just touching my meter probes to pins 4 & 6 for measuring sometimes sets it off and the bias runs away again. I've double-checked the orientation of all the transistors and everything's good there, replaced the small electrolytic caps to no effect (the large caps had previously been replaced and test fine). I'm all out of ideas on this one so I thought I'd seek help here! Hope someone can lend some advice...please!
Thanks.
I have a Quad 303 (serial 45343 - schematic here) that underwent an upgrade after burning up its outputs. Here's the subs I used, and what was originally in there:
TR1 MJ15003 2N3055
TR2 MJ15003 RCA 38494
TR100 BC560 BC214C
TR101 BC549B BC184
TR102 BC546B C1007
TR103 BC546B C1007
TR104 BC556B E5458
TR105 BC461 38496
TR106 BC141 38495
TR107 BC549B BC184
TR3 MJ15003 2N3773
TR200 BC556B C1593
TR201 BC141 40409
I fired it up on a 40W light bulb limiter and everything looks good initially. I adjust the power supply to 67V no problem, I adjust the half-rail to 33.5V no problem, but when I begin to increase the bias, it quickly jumps from around 5mV up to over 200mV and the limiter bulb starts pulsating brightly, at which point I back off the bias to zero again. I've replaced the diodes and the bias trimmers so the wiper isn't going open-circuit and sometimes I can get both L & R channels biased to around 7mV but just touching my meter probes to pins 4 & 6 for measuring sometimes sets it off and the bias runs away again. I've double-checked the orientation of all the transistors and everything's good there, replaced the small electrolytic caps to no effect (the large caps had previously been replaced and test fine). I'm all out of ideas on this one so I thought I'd seek help here! Hope someone can lend some advice...please!
Thanks.
Don't use the Mains Bulb Tester while trying to increase the bias current.
But that may have nothing to do with the sudden increase in power being consumed.
But that may have nothing to do with the sudden increase in power being consumed.
Pins 4 and 6 are the collector and emitter of the two output transistors I believe. If simply touching this point with a meter causes a sudden change in bias then everything points to high frequency instability (oscillation).
Output triples such as these are notoriously difficult to stabilise. I wonder if the ancient (ancient as in original production and low Ft) 38494 was an integral part of the design. Both outputs should have been the same so the 2N3055 suggests previous repair work.
Is the zobel network intact ?
Output triples such as these are notoriously difficult to stabilise. I wonder if the ancient (ancient as in original production and low Ft) 38494 was an integral part of the design. Both outputs should have been the same so the 2N3055 suggests previous repair work.
Is the zobel network intact ?
The 303 output triples absolutely depend on SLOW output devices for stability. The 15003 is way too fast. The original circuit used a slow version of the 2N3055
- TR1: 38494 - 2N3055 MJ15003
- TR2: 38494 - 2N3055 MJ15003 (or Push-pull alternative* MJ15004)
- TR3: 40411 - MJ802
- TR100: BC154 - BC560 BC214C
- TR101/TR107: BC109 - BC109 BC237C BC184K
- TR102/TR103: U17219 - BC546B ZTX304
- TR104: U17229 - BC556B ZTX504
- TR105: 38496 - BC461 2N5322
- TR106: 38495 - BC441 2N5320
- TR200: U17229 - BC556
- TR210: 38495 - BC441
- MR 100/101/105/106 can be replaced by 1N4148
- Replacing semiconductors is a foolish practice sometimes.
I agree with the comment about instability. Check the Zobel netwrok.
NB I have been fitting MJ15003s to Quad 303s for over ten years. They are not 'too fast' for the 303. The original version used 38494s which were 2N3055 selected for larger SOA, not lower speed.
NB I have been fitting MJ15003s to Quad 303s for over ten years. They are not 'too fast' for the 303. The original version used 38494s which were 2N3055 selected for larger SOA, not lower speed.
- Status
- Not open for further replies.