Hi guys.
I need help with the Arcam A70.
The amp runs very very hot. Bias is 250-300mV for both channels and does not tolerate adjustment.
There is no service manual for this model, but there is one for the A80, which according to some users is the same as the A70, but with a larger transformer.
What do you think the problem might be? I read that it could be the emitter resistors that are built into the SAP15N/P transistors.
scheme for A80:
I need help with the Arcam A70.
The amp runs very very hot. Bias is 250-300mV for both channels and does not tolerate adjustment.
There is no service manual for this model, but there is one for the A80, which according to some users is the same as the A70, but with a larger transformer.
What do you think the problem might be? I read that it could be the emitter resistors that are built into the SAP15N/P transistors.
scheme for A80:
Those output transistors do have a bit of a reputation... but both channels with the same fault. That seems a bit odd but is possible I suppose.
I wouldn't look to deeply without trying a new pair of whatever the latest fitment is. The only other things to check are the supply voltages (always check those as the first thing in any fault-finding) and you could also measure the voltage across C425 and see how low it can pulled with the bias preset. That's just for interest really.
I wouldn't look to deeply without trying a new pair of whatever the latest fitment is. The only other things to check are the supply voltages (always check those as the first thing in any fault-finding) and you could also measure the voltage across C425 and see how low it can pulled with the bias preset. That's just for interest really.
Have the output transistors been changed? If the old ones failed, it is likely heavy current flowed through D401 causing it to short circuit.
NEVER fit fleabay replacements as you will never know if they are as described, unless you have dealt with the supplier before and are willing to chance it.
SAP15/16 pairs have been out of production for many years!
NEVER fit fleabay replacements as you will never know if they are as described, unless you have dealt with the supplier before and are willing to chance it.
SAP15/16 pairs have been out of production for many years!
I'm sorry, but IMHO, the amp schematic and/or design is suspect.
I don't dispute suspicion about D401 current flow being being related to excess current. But output DC offset is typically governed by global feedback biasing. This design purports to have DC servo that should take output DC to near 0V. I'd take a quick look at servo amp IC400B, pin 7. I expect it's stuck on a rail.
Worse, feedback topology at IC400A seems like nonsense. If this schematic were an English composition, it would be full of sentences that don't parse.
Sorry for pessimistic assessment.
I don't dispute suspicion about D401 current flow being being related to excess current. But output DC offset is typically governed by global feedback biasing. This design purports to have DC servo that should take output DC to near 0V. I'd take a quick look at servo amp IC400B, pin 7. I expect it's stuck on a rail.
Worse, feedback topology at IC400A seems like nonsense. If this schematic were an English composition, it would be full of sentences that don't parse.
Sorry for pessimistic assessment.
I took post #1 to mean the bias current was excessive with it running hot and was assuming DC offset was not an issue. Perhaps the op can clarify,
Hi Mooly,
I wasn't pointing to a IC400B as a possible failure; rather, I doubt it can servo as intended because of other circuit problems. But an easy test of the health of that opamp: make sure power supply voltages to the opamp are appropriate, and confirm pin 5 is about 0V. Then use meter probe to short pins 7 and 6 to each (i.e. shorting across C422). The opamp output should return to ground (i.e. pin 5 voltage), then ramp to a correction voltage when the short to pin 6 is released.
I wasn't pointing to a IC400B as a possible failure; rather, I doubt it can servo as intended because of other circuit problems. But an easy test of the health of that opamp: make sure power supply voltages to the opamp are appropriate, and confirm pin 5 is about 0V. Then use meter probe to short pins 7 and 6 to each (i.e. shorting across C422). The opamp output should return to ground (i.e. pin 5 voltage), then ramp to a correction voltage when the short to pin 6 is released.
It would need a lot of untangling as to how it all comes together around that front end.
What we need is the op to return and confirm just where the issue/s are with regard to offset and bias current.
What we need is the op to return and confirm just where the issue/s are with regard to offset and bias current.
I'm following the thread, but I don't have a chance to work on the amp right now.
I'll do it as soon as I get a chance.
Greetings to all!
I'll do it as soon as I get a chance.
Greetings to all!
Hi lim0nade,
I recommend you continue your search for good documentation on this amp. Perhaps other members can contribute. If nothing else is available, you could try to confirm the hardware is well represented by the schematic.
A disturbing detail is that the schematic appears to be a work in progress. Perhaps the most egregious example is the R447 through R450 network of series/parallel resistors. Typically, this would be a gain setting resistor that would be negative feedback to an input differential transistor pair. But this schematic shows the feedback tied directly to the output of IC400A opamp. This make no sense at all and doesn't bode well for design quality in general. Hence the wish for a schematic that looks more tenable.
Good luck!
I recommend you continue your search for good documentation on this amp. Perhaps other members can contribute. If nothing else is available, you could try to confirm the hardware is well represented by the schematic.
A disturbing detail is that the schematic appears to be a work in progress. Perhaps the most egregious example is the R447 through R450 network of series/parallel resistors. Typically, this would be a gain setting resistor that would be negative feedback to an input differential transistor pair. But this schematic shows the feedback tied directly to the output of IC400A opamp. This make no sense at all and doesn't bode well for design quality in general. Hence the wish for a schematic that looks more tenable.
Good luck!
I sort of got the feeling it isn't drawn correctly as well and I can't really see how it would/could work as it stands. The op does say this diagram isn't even for the model in question.
Hi again guys.
After opening the thread and before getting your advice's I randomly checked a few capacitors and found some with high ESR.
I then removed the board for a more thorough measurement. I found about 20 pcs. with bad ESR 7-8 up to 40 Ohm.
I replaced these capacitors today and the BIAS adjustment is back to normal. The offset is 2/2.5 mV.
The amp does not overheat and sounds great.
Thanks again everyone!
Cheers!
After opening the thread and before getting your advice's I randomly checked a few capacitors and found some with high ESR.
I then removed the board for a more thorough measurement. I found about 20 pcs. with bad ESR 7-8 up to 40 Ohm.
I replaced these capacitors today and the BIAS adjustment is back to normal. The offset is 2/2.5 mV.
The amp does not overheat and sounds great.
Thanks again everyone!
Cheers!
Yes, well done. I wonder if it was unstable with high esr caps and that caused the high current.
Excellent 👍
Excellent 👍
It seems that the trouble has been resolved.
If the ESR of the capacitor was really the cause, it may have been oscillating, judging from the symptoms of abnormal heat generation, increased offset, and inability to adjust the bias.
Hello BSST
You seem to have doubts about the schematic, but I don't think the schematic is wrong.
The feedback resistance of R447 to R450 from the output is connected to the output of TL072 by the local current feedback to the output stage (SEPP) of TL072.
The non-inverting input from the output of TL072 is also fed back by R445 and R446, but this is also local feedback from the viewpoint of the entire circuit.
The 22R looks very heavy to the TL072, but most of the current comes from the output via R447-R450, so the current output by the TL072 is negligible.
After all, the gain of this circuit operates as (820 + 22 + 22) / (22 + 22).
Another TL072 unit also functions properly as a DC servo.
If the output offset becomes the positive side → IC400B ⑦ shifts to the negative side → Increases the negative power supply current of the IC400 ④ → Increases the current of TR404, 416, 417, 411 → Pulls the output offset back to the negative side
best regard
If the ESR of the capacitor was really the cause, it may have been oscillating, judging from the symptoms of abnormal heat generation, increased offset, and inability to adjust the bias.
Hello BSST
You seem to have doubts about the schematic, but I don't think the schematic is wrong.
The feedback resistance of R447 to R450 from the output is connected to the output of TL072 by the local current feedback to the output stage (SEPP) of TL072.
The non-inverting input from the output of TL072 is also fed back by R445 and R446, but this is also local feedback from the viewpoint of the entire circuit.
The 22R looks very heavy to the TL072, but most of the current comes from the output via R447-R450, so the current output by the TL072 is negligible.
After all, the gain of this circuit operates as (820 + 22 + 22) / (22 + 22).
Another TL072 unit also functions properly as a DC servo.
If the output offset becomes the positive side → IC400B ⑦ shifts to the negative side → Increases the negative power supply current of the IC400 ④ → Increases the current of TR404, 416, 417, 411 → Pulls the output offset back to the negative side
best regard
You could be right...
The one thing that looked wrong to me was the impedance at TR404 and TR408 base, specifically the effective clamping of their bases and whether it would prevent the rail current to the opamp varying in response to the audio... but it won't.
The front end is very reminiscent of an ancient Elektor design using a 741 in a not so dissimilar configuration.
Yes, TR404 and TR408 work as a cascode (base common circuit) to transmit only the power supply current of OP-AMP to the current mirror beyond it without exceeding the withstand voltage of OP-AMP.
The prototype of this method seems to have been announced using 741. It seems that it was not an audio amplifier but a motor driver.
The Alexander circuit is famous as an audio amplifier.
https://www.analog.com/media/en/tec...tes/58052492001115525484056221917334an211.pdf
