-13.7VDC output from Pioneer VSX-409RDS caused by multiple grounds
I bought a broken amplifier again, but this time it behaves really strange. Now I understand the basic problems of amplifiers and this one doesn't seem to comply to those. It is a pretty cheap 5-channel amplifier from Pioneer (1999) using a 2 channel IA and a 3-channel IA.
The first problem I found was a blown fuse in the standby 12V+ part. I replaced it and it booted up, but I measured -13.7V at the Front Left and Right outputs. I guessed that this problem was most likely caused by a broken Integrated Amplifier and I replaced the PAC010A (2-channel IA). This, however didn't fix the problem, so me and my father searched for the problem, but couldn't find it. I did find a strange interference of two voltages (the IAs use a 55V+, 55V-, 35V+ and 35V- input). I read on the 35V- pin 45V-. When I removed the 35V- and 35V+, I still read 45V- on the 35V- pin. I found 7V+ on the 35V+ pin too. This only completely went away after removing the 55V+ and 55V- pins. When I removed the PAC011A (3-channel IA), I read 35V- on the 35V- pin and when I removed the connector that transferred the audio-input to the IAs, it was 38V-, so I just left it there. With that kind of solved, the main problem still wasn't solved. We found this amplifier had 2 separate grounds with a difference of 13.7V, when using the speaker ground as the main ground it was -13.7VDC. When again removing the connector that transfers the audio-input, it outputs -50VDC on the speaker outputs, so, again, I just left it there. What we can't understand is: how can those two separate grounds cause an output of -13.7VDC on the speaker output.
The Service Manual is here: PIONEER VSX-409RDS SM Service Manual download, schematics, eeprom, repair info for electronics experts
I bought a broken amplifier again, but this time it behaves really strange. Now I understand the basic problems of amplifiers and this one doesn't seem to comply to those. It is a pretty cheap 5-channel amplifier from Pioneer (1999) using a 2 channel IA and a 3-channel IA.
The first problem I found was a blown fuse in the standby 12V+ part. I replaced it and it booted up, but I measured -13.7V at the Front Left and Right outputs. I guessed that this problem was most likely caused by a broken Integrated Amplifier and I replaced the PAC010A (2-channel IA). This, however didn't fix the problem, so me and my father searched for the problem, but couldn't find it. I did find a strange interference of two voltages (the IAs use a 55V+, 55V-, 35V+ and 35V- input). I read on the 35V- pin 45V-. When I removed the 35V- and 35V+, I still read 45V- on the 35V- pin. I found 7V+ on the 35V+ pin too. This only completely went away after removing the 55V+ and 55V- pins. When I removed the PAC011A (3-channel IA), I read 35V- on the 35V- pin and when I removed the connector that transferred the audio-input to the IAs, it was 38V-, so I just left it there. With that kind of solved, the main problem still wasn't solved. We found this amplifier had 2 separate grounds with a difference of 13.7V, when using the speaker ground as the main ground it was -13.7VDC. When again removing the connector that transfers the audio-input, it outputs -50VDC on the speaker outputs, so, again, I just left it there. What we can't understand is: how can those two separate grounds cause an output of -13.7VDC on the speaker output.
The Service Manual is here: PIONEER VSX-409RDS SM Service Manual download, schematics, eeprom, repair info for electronics experts
Last edited:
That's a particularly hard-to-read schematic (yea! Pioneer!), so I'm too lazy to trace down the lines. For what it's worth, though, don't assume that the speaker black or [-] terminals is the same as the amplifier's input ground, there are many amplifiers in which that is not so. So when measuring amplifier speaker output voltages, measure between only each speaker's two terminals, the voltage from either to the chassis or input connector shields isn't necessarily 0.
Yes, I did indeed learn that those were different and I don't automatically assume they are the same. I did measure between the two speaker terminals everytime when mentioning output DC voltage. Everytime I mentioned a power supply voltage, I measured between the power supply ground and the respective +. When measuring the difference between grounds, I put the power supply ground, which is connected to the speaker ground, on the - of the meter. What bothers me is that the voltage on the speaker terminals seems to equal the difference between the grounds, although no physical connection can be found.
New measurements
We did some new measurements just now and did find something new. We removed the other IA; now there are no IAs connected anymore.
We found that the difference between the grounds was still -13.4VDC, when putting the negative voltmeter end on the speaker/power supply ground. However when removing the amplifier input connector (CN601) it was around -54VDC (It might be that this was this way all along and I simply forgot to measure this way). This seems to prove our hypothesis that the speaker output outputs the difference between the grounds (when removing the amplifier input connector, the output voltage is around -50VDC; I first thought the IA had some weird design flaw, which caused it to need the input or something).
There is a simple circuit which seems to equal the grounds and none of its components are broken. The other ground (U-ground) is supplied by the amplifier input connector. The -54VDC looks a lot like the -55VDC from the power supply and I know this doesn't make sense, but It looks like there is a connection.
The weird -38V/-45V from the -35VDC supply has disappeared after removing all the IAs.
We also tried to test the input capacitors from the PAC010A which all seem fine.
My father told me that he found a broken Zener diode (D605) and he replaced it (I forgot to mention this in my initial post and it might be important, because both PAC010As have been exposed to this broken Zener).
Now we have two hypotheses: The IAs are both broken (one of them a second time) and the difference between the grounds is supposed to be -13.4VDC (or it doesn't matter) or there is something faulty somewhere in this complete circuit which causes the difference between the grounds to be so big that the equalizing circuit can't handle it.
We did some new measurements just now and did find something new. We removed the other IA; now there are no IAs connected anymore.
We found that the difference between the grounds was still -13.4VDC, when putting the negative voltmeter end on the speaker/power supply ground. However when removing the amplifier input connector (CN601) it was around -54VDC (It might be that this was this way all along and I simply forgot to measure this way). This seems to prove our hypothesis that the speaker output outputs the difference between the grounds (when removing the amplifier input connector, the output voltage is around -50VDC; I first thought the IA had some weird design flaw, which caused it to need the input or something).
There is a simple circuit which seems to equal the grounds and none of its components are broken. The other ground (U-ground) is supplied by the amplifier input connector. The -54VDC looks a lot like the -55VDC from the power supply and I know this doesn't make sense, but It looks like there is a connection.
The weird -38V/-45V from the -35VDC supply has disappeared after removing all the IAs.
We also tried to test the input capacitors from the PAC010A which all seem fine.
My father told me that he found a broken Zener diode (D605) and he replaced it (I forgot to mention this in my initial post and it might be important, because both PAC010As have been exposed to this broken Zener).
Now we have two hypotheses: The IAs are both broken (one of them a second time) and the difference between the grounds is supposed to be -13.4VDC (or it doesn't matter) or there is something faulty somewhere in this complete circuit which causes the difference between the grounds to be so big that the equalizing circuit can't handle it.
Difference between grounds
Does anybody know if this circuit is really supposed to equalize the grounds?
The -VF line goes to the chip PDG246A. We think that -VF draws a certain current until the voltage between D and UG is equal to 35.5V; evening the grounds in the process. Could it be that the circuitry inside of the chip meant to draw a certain current is defect? The rest of the chip seems to be working fine.
My personal hypothesis is that some voltage spike in the net destroyed a few components, because randomly there seem to be broken parts: a fuse, a zener and maybe this chip.
Does anybody know if this circuit is really supposed to equalize the grounds?
The -VF line goes to the chip PDG246A. We think that -VF draws a certain current until the voltage between D and UG is equal to 35.5V; evening the grounds in the process. Could it be that the circuitry inside of the chip meant to draw a certain current is defect? The rest of the chip seems to be working fine.
My personal hypothesis is that some voltage spike in the net destroyed a few components, because randomly there seem to be broken parts: a fuse, a zener and maybe this chip.
You must find the cause for the differing ground potential first, then go about repairing any remaining faults. Loss of continuity in the ground connection can cause the most random and bizarre problems.
From what you write, it sounds like power amp input and feedback ground differs from speaker terminal / output ground. They are supposed to be tied together somewhere (sometimes via a low-value resistor, e.g. 10 ohms). You'll have to follow these connections around and determine where the break is - start with any two points that should be grounded, determine whether there is continuity, maybe mark connection on printout of schematic. Move one end to another point and repeat until you get to a spot that is clearly not connected, then attempt to narrow things down.
Typical causes for breaks would include: Bad solder joints, bad connections at flex cables, cracked PCB, wire jumpers eaten by killer glue.
From what you write, it sounds like power amp input and feedback ground differs from speaker terminal / output ground. They are supposed to be tied together somewhere (sometimes via a low-value resistor, e.g. 10 ohms). You'll have to follow these connections around and determine where the break is - start with any two points that should be grounded, determine whether there is continuity, maybe mark connection on printout of schematic. Move one end to another point and repeat until you get to a spot that is clearly not connected, then attempt to narrow things down.
Typical causes for breaks would include: Bad solder joints, bad connections at flex cables, cracked PCB, wire jumpers eaten by killer glue.
I have already thoroughly inspected the board and it doesn't look broken anywhere; the soldering seems fine everywhere or I have already resoldered it myself.
I don't think they are supposed to be tied together through a low value resistor, because we found the circuit as shown in my previous thread. This has the ability to create a separate ground which is determined relative to the speaker ground. We don't fully understand why they would go to such lengths, but this seems to be the case. This circuit seems to be fine, so the -VF might be the culprit. What do you think?
The Zeners you have circled are nothing to do with the power amps - they are for the DC bias on the AC feed (FLAC1 and FLAC2) to the vacuum flourescent display. This is a pretty typical arrangement for VFDs
From the schematic it looks as if all speaker -VE's should be grounded at 0V, and this is taken to the midpoint of the power supply capacitors (which it should be). I would check that the "star ground" point is actually connected solidly to the centre tap of the transformers. It might also be worth looking for any damaged track or cracked solder joints in that area
From the schematic it looks as if all speaker -VE's should be grounded at 0V, and this is taken to the midpoint of the power supply capacitors (which it should be). I would check that the "star ground" point is actually connected solidly to the centre tap of the transformers. It might also be worth looking for any damaged track or cracked solder joints in that area
My bets would still be on one of the multiple cable connections in this unit requiring cleaning and reseating.
From what I can tell, ground (GND) is connected to chassis at the FL/FR/C speaker terminals. From there it makes its way to power supply star ground and power amplifier output ground on AWX7388, where the main power transformer secondary connects via TRANS3 (AWX7392).
Power amplifier input ground (AG) comes in via CN291 / CN290, and is then heavily used on the main board (AWX7381). It makes its way to the regulator board (AWX7389) from there via CN101 / CN803, where is is being merged with DG (digital ground), ADG (analog power ground?) and UG (misc. stuff).
The circuit that the OP was so mystified by does not "equalize the grounds" at all - it supplies voltages to the VFD and gets its ground (UG) from the standby power supply area on AWX7388, which in turn connects back to the regulator board AWX7389 via CN53 / CN801.
Regulator board ground also comes in via CN801 / CN53 (REG G), running on AWX7388 to 701 / 851 on TRANS2 (AWX7391).
All the input jacks appear to be isolated with RF bypass caps to chassis like in the olden days, which is somewhat baffling on a device that necessarily has quite a few of those.
The only remaining question is where the connection between output and regulator ground is. It has got to be somewhere. You can't just leave two different grounds flappin' in the breeze like that.
One possible candidate might be the pin 2 connection on the power amp hybrids, which travels via a chunky trace and IC603 (looks like a 1A picofuse - possible replacement) to mysterious terminal KN601. If that connections to chassis / output ground, our mystery would be solved.
So perhaps check IC603 for continuity, too.
From what I can tell, ground (GND) is connected to chassis at the FL/FR/C speaker terminals. From there it makes its way to power supply star ground and power amplifier output ground on AWX7388, where the main power transformer secondary connects via TRANS3 (AWX7392).
Power amplifier input ground (AG) comes in via CN291 / CN290, and is then heavily used on the main board (AWX7381). It makes its way to the regulator board (AWX7389) from there via CN101 / CN803, where is is being merged with DG (digital ground), ADG (analog power ground?) and UG (misc. stuff).
The circuit that the OP was so mystified by does not "equalize the grounds" at all - it supplies voltages to the VFD and gets its ground (UG) from the standby power supply area on AWX7388, which in turn connects back to the regulator board AWX7389 via CN53 / CN801.
Regulator board ground also comes in via CN801 / CN53 (REG G), running on AWX7388 to 701 / 851 on TRANS2 (AWX7391).
All the input jacks appear to be isolated with RF bypass caps to chassis like in the olden days, which is somewhat baffling on a device that necessarily has quite a few of those.
The only remaining question is where the connection between output and regulator ground is. It has got to be somewhere. You can't just leave two different grounds flappin' in the breeze like that.
One possible candidate might be the pin 2 connection on the power amp hybrids, which travels via a chunky trace and IC603 (looks like a 1A picofuse - possible replacement) to mysterious terminal KN601. If that connections to chassis / output ground, our mystery would be solved.
So perhaps check IC603 for continuity, too.
Hello, I have the same problem, the 5 outputs have those negative 13V, the equipment in question was powered with 220Vac, the original power being 120Vac. A fuse blew and blew the filter on the stand-by source. After leaving the power supply operating, this problem appeared, there is a lot of ripple in the negative low voltage voltage of 33v and the FM reception does not work very well.
Any suggestions ?. Thanks
Any suggestions ?. Thanks