I shall check the caps... As you rightly say they are underrated for the voltage being input.
turk, the resistors fry as soon as a load is connected to +/- 18v rails [if you see my post above, a 10k resistor across one of the rails caused the same thing to happen.] If just the transformer is connected they are fine.
turk, the resistors fry as soon as a load is connected to +/- 18v rails [if you see my post above, a 10k resistor across one of the rails caused the same thing to happen.] If just the transformer is connected they are fine.
well you could save yourself subsequent resistor replacements with a lamp limiter as suggested.
your problem seems to point to the darlington pass transistors.
your problem seems to point to the darlington pass transistors.
I'll have to put one together, I used to have one but lost it in a house move 🙄
I've just been measuring voltages across the caps & every single one is a good 2v higher than the caps are specified for... I'll order up some replacements with higher voltage ratings. Also just pulled the darlingtons & measured them... They both test OK with my dmm but should I order some new one's "just in case" ?
I've just been measuring voltages across the caps & every single one is a good 2v higher than the caps are specified for... I'll order up some replacements with higher voltage ratings. Also just pulled the darlingtons & measured them... They both test OK with my dmm but should I order some new one's "just in case" ?
you need to explain to me how you checked a darlington with a multimeter?
i did think that was possible.
i did think that was possible.
Are we talking about the same thing here ? The bipolar transistors used in the 18v circuit [Q853/Q854]
Exactly what I did: DMM in "diode" mode. Connected the red meter lead to base, Connected the black meter lead to emitter. 2SD1417 [NPN] read a junction drop of .788. Moved the black lead to the collector. The reading was the same as the previous test. Reversed the meter leads and repeated with the meter showing OL both times... Then did it all again but the opposite polarity for the 2SB1022 [PNP] and got a junction drop of .922
Did I do it right or have I lost the plot ? Thanks for all the continued help guys, I owe you all a beer
Exactly what I did: DMM in "diode" mode. Connected the red meter lead to base, Connected the black meter lead to emitter. 2SD1417 [NPN] read a junction drop of .788. Moved the black lead to the collector. The reading was the same as the previous test. Reversed the meter leads and repeated with the meter showing OL both times... Then did it all again but the opposite polarity for the 2SB1022 [PNP] and got a junction drop of .922
Did I do it right or have I lost the plot ? Thanks for all the continued help guys, I owe you all a beer
Ah, my mistake ! A quick look up on Wiki has set me straight.
I wonder why Marantz chose this arrangement over basic 7818/7918 regs used in cheaper models [that don't go wrong!]
I wonder why Marantz chose this arrangement over basic 7818/7918 regs used in cheaper models [that don't go wrong!]
cuz 7800 series regulators can't pass the current that the darlington's can.
if your forward bias readings are correct i'd put my money on the 2sd1417 as the culprit.i'd also change the cap's fore and aft of the pass element.
if your forward bias readings are correct i'd put my money on the 2sd1417 as the culprit.i'd also change the cap's fore and aft of the pass element.
Are you sure they were 10k and not something else ? 10k at 18 volts draws under 2 milliamps. Something really basic and fundamental is going wrong with all this.
I wondered that too...
It could be coincidence to fail when resistor is inserted. Voltage first drop to 9V, therefore something suddenly shorted, at the same time resistors burns. On lower voltage short circuit opens and capacitors discharge.
If 10k resistor can cause damage it is due to oscillations. In that case I would suspect capacitors too.
I can assure you that nothings connected apart from the transformer & yes the resistor was 10k 1/2w. Here is a picture of the board in question... There are two AC connection points, one for the 24v/18v regs [J852] & one for the power amplifier front end [J851] only J852 has been connected during testing.
I've just changed the cap values that I have to hand [100uf, 10uf & 1uf] but needed to order the larger values so they'll be here soon hopefully.
An externally hosted image should be here but it was not working when we last tested it.
I've just changed the cap values that I have to hand [100uf, 10uf & 1uf] but needed to order the larger values so they'll be here soon hopefully.
There must be a fundamental problem somewhere. Nothing apart from a shorted rectifier or cap across a rail could take out those 1 ohms. And again if you look at the circuit, R861 and 862 can not be influenced by the 18 volt circuitry.
when it comes to capacitor faults the behavior that bewilders me most is when; under voltage they exhibit momentary breakdown and then snap back to normal making the culprits hard to trace because standard ohm meter tests(resistance/continuity) aren't a reliable indicator of of how it behaves with voltage applied (i feel the same applies to semiconductors)
That's true and its an effect most often seen with self healing film caps which are designed to "self repair" by the film vaporising. Electroylitics don't normally behave like that, more often they can be leaky particularly as temperature rises.
There are very few parts on that PCB that can take out the 1 ohms. Caps across the rails and rectifiers. Everything else is at too high an impedance.
now i remember why the bjt test can fool you into thinking a darlington is ok!
if the second transistor becomes a short but the first transistor survives you'd simply be proving one element survived!
would that not account for the fusing resistors opening?(as in the bad darlington=more current)
if the second transistor becomes a short but the first transistor survives you'd simply be proving one element survived!
would that not account for the fusing resistors opening?(as in the bad darlington=more current)
Last edited:
mooly
i know electrolytic's don't normally do that but we are talking about end of life cycle parts.
i guess i've spend to many years and i guess braincells on intermittant cap troubles!
i know electrolytic's don't normally do that but we are talking about end of life cycle parts.
i guess i've spend to many years and i guess braincells on intermittant cap troubles!
🙂 I know.
I'm an ex service tech, and electroylitics account for many faults but rarely (in fact I can't just recall any) failing intermittently short or low impedance.
I'm an ex service tech, and electroylitics account for many faults but rarely (in fact I can't just recall any) failing intermittently short or low impedance.
Surely if the darlington transistors themselves were at fault the +/-20v wouldn't be there on the outputs ? Or am I wrong ?
This amp is around 20 years old, all original caps are 85c rated & have been run over their rated voltage for a long time so I'm hoping the "fundamental problem" is capacitor related. I'll pull what's left of the original caps this evening & stick the meter across them [I've just bought a new battery for my LCR meter...]
Do we feel it worth splashing out an extra £4 on a new darlington pair ? Maybe I should take it one step at a time 🙂
This amp is around 20 years old, all original caps are 85c rated & have been run over their rated voltage for a long time so I'm hoping the "fundamental problem" is capacitor related. I'll pull what's left of the original caps this evening & stick the meter across them [I've just bought a new battery for my LCR meter...]
Do we feel it worth splashing out an extra £4 on a new darlington pair ? Maybe I should take it one step at a time 🙂
Last edited:
One step at a time. Remember that the darlingtons in themselves can't blow the 1 ohms unless they failed C-B short in which case they would take the zener out too.
Also remember in the first post that you said R861 was open circuit. That resistor and circuitry it feeds is unrelated to the 18 volt rails.
You need to work on this methodically. Start again and take it one step at a time. With the board connected only to the transformer and with 10k resistors loading the -/+18 volts rails it should all sit there happily. Remember that the unregulated 60 volt rails will stay charged up.
Also remember in the first post that you said R861 was open circuit. That resistor and circuitry it feeds is unrelated to the 18 volt rails.
You need to work on this methodically. Start again and take it one step at a time. With the board connected only to the transformer and with 10k resistors loading the -/+18 volts rails it should all sit there happily. Remember that the unregulated 60 volt rails will stay charged up.
You are right, 37V on collector, 20V emitter is ok with no load, zener voltage.