Those readings sound good. That at least shows the output stage is biased correctly.
As you have a scope we can do more detailed tests. No speakers attached initially.
You should apply the test signal (1kHz sine for example) to one of the inputs, and then look at the output voltage on R101 and R102. Adjust the volume to give around 4 volts peak level corresponding to 1 watt into 8 ohms.
Do this for both channels and confirm the signal is distorted or not.
Also you can check that both channels are receiving the same input signal and that this signal is pure and undistorted. Measure on R71 and R72 for this.
If the output is good for channels then you now need to test with a load attached, preferably an 8 ohm resistor and see if the distortion only appears under load.
Before you do any of this, have you proved the speaker relay is OK ??? you can easily short it out when the amp is playing and see if the distortion stops.
As you have a scope we can do more detailed tests. No speakers attached initially.
You should apply the test signal (1kHz sine for example) to one of the inputs, and then look at the output voltage on R101 and R102. Adjust the volume to give around 4 volts peak level corresponding to 1 watt into 8 ohms.
Do this for both channels and confirm the signal is distorted or not.
Also you can check that both channels are receiving the same input signal and that this signal is pure and undistorted. Measure on R71 and R72 for this.
If the output is good for channels then you now need to test with a load attached, preferably an 8 ohm resistor and see if the distortion only appears under load.
Before you do any of this, have you proved the speaker relay is OK ??? you can easily short it out when the amp is playing and see if the distortion stops.
Okay yesterday I realized that mechanical tapping on the amp had an effect on the crackling, sometimes there was crackling on the left channel and sometimes there wasn't. I tried to find the source of this but it seemed to happen allmost regardless of where I tapped. So I resoldered a lot of the connections and cleaned all of the pots.
After putting the amp back together there were changes in the behaviour:
- I measured the voltage between TP7 and TP8 again and this was now 0.000 VDC so I tried to readjust it with trimp pot VR1 but this didn't have any effect. I desoldered the VR1 and it's working as it should.
- When I have an input signal on the amp and play sound either over terminal A or B and turn up the volume I hear the relay click and the sound is gone for a few seconds, then the relay clicks and the sound is back and this behaviour repeats. It happens only when theres a speaker connected to the left channel. I measured the supply voltage of the relay and at low volume it's 24VDC. When I turn the volume up above about 50% the voltage drops to zero leading the relay to switch back. Now I'm not quite sure what to look for? Is this some over current protection (maybe related to IC13 (µPC1237HA) or something different?
By the way after I put the amp back together I couldn't see any effect of tapping anymore.
After putting the amp back together there were changes in the behaviour:
- I measured the voltage between TP7 and TP8 again and this was now 0.000 VDC so I tried to readjust it with trimp pot VR1 but this didn't have any effect. I desoldered the VR1 and it's working as it should.
- When I have an input signal on the amp and play sound either over terminal A or B and turn up the volume I hear the relay click and the sound is gone for a few seconds, then the relay clicks and the sound is back and this behaviour repeats. It happens only when theres a speaker connected to the left channel. I measured the supply voltage of the relay and at low volume it's 24VDC. When I turn the volume up above about 50% the voltage drops to zero leading the relay to switch back. Now I'm not quite sure what to look for? Is this some over current protection (maybe related to IC13 (µPC1237HA) or something different?
By the way after I put the amp back together I couldn't see any effect of tapping anymore.
It sounds like you have a few things going on with this amp. Losing the bias voltage between the test points suggests that something has happened during your resoldering.
IC13 will disconnect the speakers if either a DC offset appears or an overcurrent situation (or both).
You need to check all the work you have done and make sure no solder blobs or anything like that have occurred.
IC13 will disconnect the speakers if either a DC offset appears or an overcurrent situation (or both).
You need to check all the work you have done and make sure no solder blobs or anything like that have occurred.
it doesn't hurt to look at all solder joints under a magnifying glass either (i check my own work this way) things like cracked solder joints will cause all sorts of problems and don't always correspond to physical stimulus and from the behavior your describing, the cycling of the output relay with increased volume means something is going open under current like filter cap ground's or somesuch.
You've replaced UPC1298V ? Those aren't manufactured for a long time. Are You sure they are genuine parts ? One day I had a Sony Amp for repairing and couldn't find those ICs, well I could find, but they they had an obnoxious price. So I replaced the board with two LM3886 boards and the costumer is happy.
http://www.tecnotre.it/audio/tecnica/upc1237.pdf
The cap on pin 4 is small physical diameter cap and prone to drying out and this will render the overvoltage protection overly sensitive.
I have seen this fault numerous times over 35 years.
Not to say this is your fault without reading the rest of the thread.
Dan.
Okay, I'm back again 
I soldered the power transistors out (Q21..26) and checked them with the multimeter --> several didn't test correct relating to the routine I found online to check transitors. I think I might have destroyed them myself by accidently creating a short circuit while measuring. So I ordered all 6. Yesterday I soldered them in again and it seems like it fixed the whole problem. No distorted sound anymore, actually it's sounding very nice!
There's one thing that's still bothering me: I tried to set the bias voltage again and I'm still measuring 0.00 V independent of the trim pot setting (at both channels). But since the amp sounds good to me and it's not even buzzing at standby I guess I will just stick to it.
Thanks for this answer, too. I will keep this mind. Are you talking about C51/C52 or about C53/C54?
I soldered the power transistors out (Q21..26) and checked them with the multimeter --> several didn't test correct relating to the routine I found online to check transitors. I think I might have destroyed them myself by accidently creating a short circuit while measuring. So I ordered all 6. Yesterday I soldered them in again and it seems like it fixed the whole problem. No distorted sound anymore, actually it's sounding very nice!
There's one thing that's still bothering me: I tried to set the bias voltage again and I'm still measuring 0.00 V independent of the trim pot setting (at both channels). But since the amp sounds good to me and it's not even buzzing at standby I guess I will just stick to it.
Thanks for this answer, too. I will keep this mind. Are you talking about C51/C52 or about C53/C54?
That's great news
(It is up to you if you want to investigate the lack of bias adjustment, just say if you do
Okay, after using the amp now for some time I realized that there's a little distortion on higher frequencies on the right channel. While listening to music it's hard to hear but when I play Zelda on the Switch it gets quite disturbing in the ambience sounds. Also the amp has a little hum, nothing major but I can hear it when the amp is idling at a higher volume setting.
So I would indeed like to investigate the problem.
What would you suggest to start with?
So no bias adjustment.
Lets reconfirm what is going on first.
1/ With no speakers attached and no signal present, is the voltage between TP7 and TP8 always zero, even when the pot is turned ?
2/ What is the voltage across C83 with the preset at maximum and at minimum setting ? (so two readings).
3/ What is the voltage between C and E of Q25 with the preset at maximum and at minimum setting ? (so again, two readings).
Lets reconfirm what is going on first.
1/ With no speakers attached and no signal present, is the voltage between TP7 and TP8 always zero, even when the pot is turned ?
2/ What is the voltage across C83 with the preset at maximum and at minimum setting ? (so two readings).
3/ What is the voltage between C and E of Q25 with the preset at maximum and at minimum setting ? (so again, two readings).
Okay, I just took the measurements
1. Voltage between TP7 and TP8 is 0V all the time, same for TP9 and 10. Neither the volume setting nor the trimmer pot seem to make any differece.
2. I guess by "preset" you mean the settig of the trim pot / idle adjustment?
Voltrage accross C83 (left channel)
0% --> 0.31V
100% --> 0.41V
Voltage across C84 (right channel)
0% --> 0.000 V
100% --> 0.005 V
3.Q25 and Q26 are 2SC4137 in my case (remember I have the non-SE version)
Voltage between C and E for Q25 (left channel)
0% --> 1.78V
100% --> 2.13V
Voltage between C and E for Q26 (right channel)
0% --> 0.000 V
100% --> 0.000 V
What do you think?
1. Voltage between TP7 and TP8 is 0V all the time, same for TP9 and 10. Neither the volume setting nor the trimmer pot seem to make any differece.
2. I guess by "preset" you mean the settig of the trim pot / idle adjustment?
Voltrage accross C83 (left channel)
0% --> 0.31V
100% --> 0.41V
Voltage across C84 (right channel)
0% --> 0.000 V
100% --> 0.005 V
3.Q25 and Q26 are 2SC4137 in my case (remember I have the non-SE version)
Voltage between C and E for Q25 (left channel)
0% --> 1.78V
100% --> 2.13V
Voltage between C and E for Q26 (right channel)
0% --> 0.000 V
100% --> 0.000 V
What do you think?
Last edited:
The measurements confirm the problem but not the cause. (and yes, trim pot for bias )
The standout reading is the voltage across C and E of Q26 as this is the bias generator. It would be worth switching the amp off and doing a resistance check between pins 7 and 8 of the chip (which is the same as across C and E).
Does it read short circuit ?
It is very unusual to have a problem around that stage in any amplifier and I also see you mentioned earlier that the bias was initially OK but that there was a physical issue where you could tap the amp and get it to crackle. It has to be worth having a look again at the soldering and fitment around the bias transistor.
)The standout reading is the voltage across C and E of Q26 as this is the bias generator. It would be worth switching the amp off and doing a resistance check between pins 7 and 8 of the chip (which is the same as across C and E).
Does it read short circuit ?
It is very unusual to have a problem around that stage in any amplifier and I also see you mentioned earlier that the bias was initially OK but that there was a physical issue where you could tap the amp and get it to crackle. It has to be worth having a look again at the soldering and fitment around the bias transistor.
Oh I confused pins 7 and 8 and test point 7 and 8. I was pretty tired yesterday...
I just checked again. There really was a short circuit on the right channel (Q26), C and E had a tiny solder connection. I fixed that and now I can set the bias for this channel again. Thanks for that!
On the left channel the problem is still there. I checked the resistance between C and E of Q25 and its 6.8kOhm (as is now at Q26). The trim pot itself seems to be okay, no short circuit. I also resoldered Q25 and also swapped IC1 (µPC1298V) with the one that was in the right channel at the beginning. All of this had no effect.
I just checked again. There really was a short circuit on the right channel (Q26), C and E had a tiny solder connection. I fixed that and now I can set the bias for this channel again.
Thanks for that!On the left channel the problem is still there. I checked the resistance between C and E of Q25 and its 6.8kOhm (as is now at Q26). The trim pot itself seems to be okay, no short circuit. I also resoldered Q25 and also swapped IC1 (µPC1298V) with the one that was in the right channel at the beginning. All of this had no effect.
Excellent 
Just remind me I thought the left channel sounded OK earlier (post #30) and that is was just the right channel that had the distortion... which would be due to the low bias fault.
Are we still looking for a no/low bias condition on the right channel based on you not being able to get any measurable bias voltage across the test points for the right channel ?
Just remind me
I thought the left channel sounded OK earlier (post #30) and that is was just the right channel that had the distortion... which would be due to the low bias fault.Are we still looking for a no/low bias condition on the right channel based on you not being able to get any measurable bias voltage across the test points for the right channel ?
Exactly! I just hooked the amp up again and the right channel sounds good now. The left channel already sounded good, it's just that I can't set any bias (I measure 0.001V). I can also feel that the µPC1298V of the left side is a lot cooler than on the right which was the reason I swapped the right one months ago (probably unnecessarily).
I just tried the tapping again but couldn't hear any effect, I think this stopped when I had changed Q21 to 26.
So I guess if Q25, IC1, VR1 and C83 is okay then it's probably Q21 or Q23?
It would be cool if I could find the fault but it doesn't really seem to be necessary at this point. It's nice that I could fix the other channel, though
I just tried the tapping again but couldn't hear any effect, I think this stopped when I had changed Q21 to 26.
So I guess if Q25, IC1, VR1 and C83 is okay then it's probably Q21 or Q23?
It would be cool if I could find the fault but it doesn't really seem to be necessary at this point. It's nice that I could fix the other channel, though
So... more questions
Have the output transistors been replaced on the left channel ?
Do the type numbers fitted for the left channel outputs match the right channel ?
The readings in post #32 show that the left channel was basically functioning, so how does the voltage across C83 now compare to C84. You need around 1.2 volts across the cap in order to begin to get the outputs to conduct current.
Also compare the voltage you now get across C and E of Q26 in the right channel to that of Q25 in the left. Is Q26 in the good channel higher ?
Have the output transistors been replaced on the left channel ?
Do the type numbers fitted for the left channel outputs match the right channel ?
The readings in post #32 show that the left channel was basically functioning, so how does the voltage across C83 now compare to C84. You need around 1.2 volts across the cap in order to begin to get the outputs to conduct current.
Also compare the voltage you now get across C and E of Q26 in the right channel to that of Q25 in the left. Is Q26 in the good channel higher ?
A1693 C4466 Power Transistor 80V 6A 60W (pair) | eBay
I bought two pairs of these as a replacement for both channels and I think if I accidently swapped two it shouldn't work at all since one is NPN and the other PNP.
For Q25 and Q26 I bought two new 2SC4137. I will check the numbers again when I get home and will take the measurements that you proposed.
The lack of bias adjustment already started before I swapped the six transistors. I don't think it was the resoldering though because this was basically on the input/equalizer pcb since I thought the mechanical tapping problem was somewhere in that area. On the main PCB I basicly just resoldered the µPC1298V, the six other transistors, the trim pots and maybe a few other things.
I bought two pairs of these as a replacement for both channels and I think if I accidently swapped two it shouldn't work at all since one is NPN and the other PNP.
For Q25 and Q26 I bought two new 2SC4137. I will check the numbers again when I get home and will take the measurements that you proposed.
The lack of bias adjustment already started before I swapped the six transistors. I don't think it was the resoldering though because this was basically on the input/equalizer pcb since I thought the mechanical tapping problem was somewhere in that area. On the main PCB I basicly just resoldered the µPC1298V, the six other transistors, the trim pots and maybe a few other things.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.