Thanks for the suggestion. Ill try to resolder all of the transformer leads to the main board first as I had difficulty with limited space - there may be a poor connection there. If that doesn't make a difference, ill have to do what you suggested. Otherwise I'm fresh out of ideas what could be wrong if that doesn't fix it
I just redid all 13 power terminals on the board and unfortunately it did not resolve the issue.
I also measured the DC offset from the speaker terminals (with no load) and they read around 50mV on the left channel and just above 70mV on the right channel. Not sure if this helps but those values seem high.
Looks like I'm going to have to look at the input board next and see what's going on there as per your suggestion.
I also measured the DC offset from the speaker terminals (with no load) and they read around 50mV on the left channel and just above 70mV on the right channel. Not sure if this helps but those values seem high.
Looks like I'm going to have to look at the input board next and see what's going on there as per your suggestion.
Check the input board for dry joints etc. One way you could see if it's amp or input board is to bypass the input board and pass the signal into the preamp directly.
That this point we're almost out of easy (non-scope) suggestions.
I'm just brainstorming guesses:
High DC offset would point to a dead output component unbalancing the two sides of the output devices. A bad amp design will result in DC offset if the bias circuit is out of whack and can't balance itself (ie if there's a servo circuit and it's dead).
The A220 is a simple beast so there's no servo. If I use the brymen I can see there's little DC (1.6mV) with some AC on there too. However if I use the crest function (simply watches for max with fast measurement) it shows 6mVdc on but if I switch off that the amp puts out 20mVdc max.
So that does seem high..
That this point we're almost out of easy (non-scope) suggestions.
I'm just brainstorming guesses:
High DC offset would point to a dead output component unbalancing the two sides of the output devices. A bad amp design will result in DC offset if the bias circuit is out of whack and can't balance itself (ie if there's a servo circuit and it's dead).
The A220 is a simple beast so there's no servo. If I use the brymen I can see there's little DC (1.6mV) with some AC on there too. However if I use the crest function (simply watches for max with fast measurement) it shows 6mVdc on but if I switch off that the amp puts out 20mVdc max.
So that does seem high..
update: I bypassed the input board and fed the audio signal directly to the main board. Same result. ****.
I measured the offset again - 72mV on right channel, and 56mV on left. I'm beginning to think there is a component that was damaged somewhere... the issue is that buzzing is on BOTH channels suggesting a general power/ground input issue rather than a single component failing.
Another note, when I disconnected the left and right signal leads from the input board and left them, I noticed there was a strong EMI when my hand came close to each of the cables and touched them. As I moved away, it disappeared. I'm not sure if this is to be expected, but it was noticeable.
Further note, I did test to see if it was my preamp or streamer/DAC by swapping electronics along with swapping cables including power cables - no change. It is the power amp.
Lastly, I did notice that when my streamer/dac was completely disconnected from the system but powered it on, it did in fact create very slight interference being played through my speakers via the power amp (it is located two shelves above the amp, but connected to the same wall socket) For example, I could wiggle the wi-fi dongle and there would be an ever slight corresponding noise being played through the speakers indicating EMI/ground issue. Again, output from the streamer/DAC is not physically connected.
I really am out of ideas at this point, so any further is suggestions are all welcomed. The only thing I can think of that is probably irrelevant is that I used Gelid GC-Extreme thermal CPU past on the bridge rectifiers that are bolted to the chassis for heat exchange. The paste is not electrically conductive so I doubt this is an issue....
I measured the offset again - 72mV on right channel, and 56mV on left. I'm beginning to think there is a component that was damaged somewhere... the issue is that buzzing is on BOTH channels suggesting a general power/ground input issue rather than a single component failing.
Another note, when I disconnected the left and right signal leads from the input board and left them, I noticed there was a strong EMI when my hand came close to each of the cables and touched them. As I moved away, it disappeared. I'm not sure if this is to be expected, but it was noticeable.
Further note, I did test to see if it was my preamp or streamer/DAC by swapping electronics along with swapping cables including power cables - no change. It is the power amp.
Lastly, I did notice that when my streamer/dac was completely disconnected from the system but powered it on, it did in fact create very slight interference being played through my speakers via the power amp (it is located two shelves above the amp, but connected to the same wall socket) For example, I could wiggle the wi-fi dongle and there would be an ever slight corresponding noise being played through the speakers indicating EMI/ground issue. Again, output from the streamer/DAC is not physically connected.
I really am out of ideas at this point, so any further is suggestions are all welcomed. The only thing I can think of that is probably irrelevant is that I used Gelid GC-Extreme thermal CPU past on the bridge rectifiers that are bolted to the chassis for heat exchange. The paste is not electrically conductive so I doubt this is an issue....
Those bridge rectifiers are typically electrically insulated between the back and the chassis they're bolted to, unless the bolt through them is touching something.
One option here - that the output devices are grounding to the heatsink. I can't find any datasheet mention but the SAP15 seems to have a metal back (a good indication that it's electrically connected too).
I had this with my hybrid tube amp - from the metal back to the grounded heatsink blew both BJT.
To find out if one is dead or shorted you should be able to test in a number of ways:
a) test chassis continuity (if one of the pins is connected to the back) - power off for this one (and let it bleed down).
b) test the resistor for the Vcc or the load (if there's one) and see how much current is going through it (measure the voltage across the resistor as V=IR in ohms law, then use V/R to find the current flowing through). Look for any one that is zero, low or high than the others.
If you have a diode setting you can test the transistors around the board (power off and let it bleed down).
What are the power rail voltages?
One option here - that the output devices are grounding to the heatsink. I can't find any datasheet mention but the SAP15 seems to have a metal back (a good indication that it's electrically connected too).
I had this with my hybrid tube amp - from the metal back to the grounded heatsink blew both BJT.
To find out if one is dead or shorted you should be able to test in a number of ways:
a) test chassis continuity (if one of the pins is connected to the back) - power off for this one (and let it bleed down).
b) test the resistor for the Vcc or the load (if there's one) and see how much current is going through it (measure the voltage across the resistor as V=IR in ohms law, then use V/R to find the current flowing through). Look for any one that is zero, low or high than the others.
If you have a diode setting you can test the transistors around the board (power off and let it bleed down).
What are the power rail voltages?
I will check everything you've mentioned but while you were posting that, I went off on a tangent following something I thought was unusual. I checked the AC section, specifically the power distribution board and measured for continuity. My multimeter went off showing continuity between live and neutral.
I checked to see if all three toroidal wires were shorting somehow and it turns out it wasn't the wires, it is the toroidal transformers that show continuity between the neutral and 120v wire. The two large transformers are showing around 1ohm reading, which I assume is fairly normal. The third smaller transformer is showing 36 ohms. This is with disconnecting all the additional wires going to the main board. I'm going to assume that isn't normal and could very well be the problem of my issues. Luckily it's the much smaller toroid of the two, but now I have to worry about sourcing a proper replacement. What do you guys think?
I checked to see if all three toroidal wires were shorting somehow and it turns out it wasn't the wires, it is the toroidal transformers that show continuity between the neutral and 120v wire. The two large transformers are showing around 1ohm reading, which I assume is fairly normal. The third smaller transformer is showing 36 ohms. This is with disconnecting all the additional wires going to the main board. I'm going to assume that isn't normal and could very well be the problem of my issues. Luckily it's the much smaller toroid of the two, but now I have to worry about sourcing a proper replacement. What do you guys think?
Last edited:
Hi, the small toroidal supplies AC to the Low Voltage Power Supply Unit (LV PSU) half wave rectifiers D16 and D17 and also the 'AC Sense' mute circuit rectifier D15 (you will have to confirm the component numbering, I'm going off the photo's).
I expect it's ok, just check DC output of the (7812 +12V?) regulator.
I expect it's ok, just check DC output of the (7812 +12V?) regulator.
NickKUK, yes I figured that. I just wanted to show the board and wiring where I showed measurements of continuity between live and neutral wires. I did not realize it was normal for Transformers to show this reading. I'm assuming the 36ohm reading on the smaller transformer also checks out? If so, I'll reassemble everything back and continue to diagnose with yours and Protegimus' suggestions.
Quick update: Just tested for continuity on the SAP15. The SAP15 PY models (6 total in a set of 3 on either side) have continuity to ground on pin 3 while the SAP15 NY models do not.
I'll check for resistor values shortly and report back. I will also remove the aluminum heatsinks on the transistors as you wondered about earlier o remove all variables.
NPN and PNP will have a different reading.
The diode setting on the DMM will return a reading - there's plenty of YT examples of working out if a transistor is dead. A darlington driver is two transistors chained, so it should still work.
No - keep the aluminium HS attached. One pin usually connects to the pad at the back. What you want to check is if any leg and the heatsink is shorted. Typically if it was you would have blown some resistors. So for now - leave it as it is. Work out the diode setting to test the transistors.
The diode setting on the DMM will return a reading - there's plenty of YT examples of working out if a transistor is dead. A darlington driver is two transistors chained, so it should still work.
No - keep the aluminium HS attached. One pin usually connects to the pad at the back. What you want to check is if any leg and the heatsink is shorted. Typically if it was you would have blown some resistors. So for now - leave it as it is. Work out the diode setting to test the transistors.
Tested the JE350 and JE340 transistors:
TR6 (JE350) - Positive terminal on c: b = 0.633 and e = 1.6
Negative Terminal on c: b = 0L and e = 0L
TR7 (JE340) - Positive Terminal on c: b = 0L and e = 0l
Negative Terminal on c: b = 0.625 and e = 0L
TR107 (JE 340) - Positive Terminal on c: b = 0 and e = 0L
Negative Terminal on c: b= 0.623 and e = 0L
TR106 (JE350) - Positive Terminal on c: b = 0.632 and e = 1.61
Negative Terminal on c: b = 0 and e = 0
Fairly consistent readings and I have checked for any shorts on the heatinks to the legs with no issue. Any further suggestions on what should be checked? I'm officially dumbfounded aside from a very last idea that the chassis was not properly grounded for whatever reason at the three screws near the bottom of the board specifically for grounding. It may have been slightly off and/or the screw ring terminals were slightly worn/covered with chassis paint that rubbed off. I've lightly sanded down the chassis where the board rests on top of these screw-ring terminals and very lightly sanded the contacts on the board and wiped down with isopropyl. If that is the issue all along....
Anyway, let me know if there's anything else I should check before putting everything back together with a prayer
TR6 (JE350) - Positive terminal on c: b = 0.633 and e = 1.6
Negative Terminal on c: b = 0L and e = 0L
TR7 (JE340) - Positive Terminal on c: b = 0L and e = 0l
Negative Terminal on c: b = 0.625 and e = 0L
TR107 (JE 340) - Positive Terminal on c: b = 0 and e = 0L
Negative Terminal on c: b= 0.623 and e = 0L
TR106 (JE350) - Positive Terminal on c: b = 0.632 and e = 1.61
Negative Terminal on c: b = 0 and e = 0
Fairly consistent readings and I have checked for any shorts on the heatinks to the legs with no issue. Any further suggestions on what should be checked? I'm officially dumbfounded aside from a very last idea that the chassis was not properly grounded for whatever reason at the three screws near the bottom of the board specifically for grounding. It may have been slightly off and/or the screw ring terminals were slightly worn/covered with chassis paint that rubbed off. I've lightly sanded down the chassis where the board rests on top of these screw-ring terminals and very lightly sanded the contacts on the board and wiped down with isopropyl. If that is the issue all along....
Anyway, let me know if there's anything else I should check before putting everything back together with a prayer
Last edited:
They were Nick and I've already mentioned that, however thewiredninja did bypass the board completely with an input and said he still experience the same symptoms.
I've asked him to take a look at C27 and C32, C32 was
replaced with
metal film. C32 is the REG1 +12v regulator reservoir capacitor, metal film are not usually recommended for this purpose as their very low ESR can cause instability. 12v rail does measure 11.9v, don't know if AC has been checked through a 0.1uF. +12v regulator supplies the input PCB, not sure what else.
I raised this as from the photo's that's the only thing I've spotted so far that looks incorrect.
There is the buzz, but also the weird distortion that varies with music that's playing. Buzz could be recorded on a PC and the resulting audio waveform frequency (120Hz?) confirmed in a basic sound editor.
C27 appears to be part of the 'AC sense' mute circuit (RC component on the base of transistor TR?).
As thimios mentioned the DC offset doesn't appear anything to be overly concerned about, it is very similar on my A308 INT and when you adjust output stage bias it doesn't affect DC offset at all.
Question @thewiredninja : what types / values are capacitors C5, C8, C9 and C105, C108 & C109 replaced with?
Image is with replaced component numbers overlayed:
I've asked him to take a look at C27 and C32, C32 was
| Jamicon TK | 10 | uF | 50 | v |
replaced with
|
metal film. C32 is the REG1 +12v regulator reservoir capacitor, metal film are not usually recommended for this purpose as their very low ESR can cause instability. 12v rail does measure 11.9v, don't know if AC has been checked through a 0.1uF. +12v regulator supplies the input PCB, not sure what else.
I raised this as from the photo's that's the only thing I've spotted so far that looks incorrect.
There is the buzz, but also the weird distortion that varies with music that's playing. Buzz could be recorded on a PC and the resulting audio waveform frequency (120Hz?) confirmed in a basic sound editor.
C27 appears to be part of the 'AC sense' mute circuit (RC component on the base of transistor TR?).
As thimios mentioned the DC offset doesn't appear anything to be overly concerned about, it is very similar on my A308 INT and when you adjust output stage bias it doesn't affect DC offset at all.
Question @thewiredninja : what types / values are capacitors C5, C8, C9 and C105, C108 & C109 replaced with?
Image is with replaced component numbers overlayed:
C5 and c105 are 16v 100uf bipolar
C8 and c108 are 16v 220uf
C9 and c109 are 160v 1uf
I also managed to put the board back together after making sure ground conditions were good and same result ended up happening.
Next in line - I'm thinking a resistor or something was damaged when twisting out the PSU caps occurred. Only reason for all of this it seems. No cracks, stress marks or damage can be observed anywhere on the board however but Ill report back soon and any suggestions in the meantime are always welcomed.
C8 and c108 are 16v 220uf
C9 and c109 are 160v 1uf
I also managed to put the board back together after making sure ground conditions were good and same result ended up happening.
Next in line - I'm thinking a resistor or something was damaged when twisting out the PSU caps occurred. Only reason for all of this it seems. No cracks, stress marks or damage can be observed anywhere on the board however but Ill report back soon and any suggestions in the meantime are always welcomed.
Last edited: