So the SAP15's transplant was a success... albeit with a few hiccups in matching them, but I've settled on what I could get. Luckily I bought several (used) sets in advance as backup. Here are the values as of now at idle with no load:
NPN:
TR27 - 13.6mV
TR21: 9.0mV
TR15: 8.7mV
PNP:
TR16: 10.4mV
TR22: 8.6mV
TR28: 12.4mV
- DC offset is steady with exactly 1.0mV. AC voltage across terminals averages 535mV. All values set with variac for steady 120v.
Originally when I hooked it up I had TR27 reading at over 32mV when the rest of the NPN's are at values stated above. So I swapped that out with another and an improvement was seen. Similar case with TR22 originally reading 4.4mV and now it's reading better than the other PNP's at roughly 8.6mV. I now see why Musical Fidelity had a special deal (or so I've read) with the manufacturer to send them matched pairs. Needless to say, I'll be looking forward to seeing the values after testing the left channel with its original SAP15's.
Interesting to note how I originally started with a DC offset of around 60mV and it's gone down 1.0mV!! That's quite a difference! Next up, I plan on hooking up a pair of cheap speakers to see if the amp plays music cleanly with zero hum. I then plan on stress testing it with an 8ohm 200watt resistor and measuring the bias again as per your suggestions. I'm beyond happy that I got it to this point versus giving up and sending it to the 'professionals' for months with zero results!
I know I may be celebrating a bit early, but thanks to you guys for everything! I've learned a hell of a lot.
NPN:
TR27 - 13.6mV
TR21: 9.0mV
TR15: 8.7mV
PNP:
TR16: 10.4mV
TR22: 8.6mV
TR28: 12.4mV
- DC offset is steady with exactly 1.0mV. AC voltage across terminals averages 535mV. All values set with variac for steady 120v.
Originally when I hooked it up I had TR27 reading at over 32mV when the rest of the NPN's are at values stated above. So I swapped that out with another and an improvement was seen. Similar case with TR22 originally reading 4.4mV and now it's reading better than the other PNP's at roughly 8.6mV. I now see why Musical Fidelity had a special deal (or so I've read) with the manufacturer to send them matched pairs. Needless to say, I'll be looking forward to seeing the values after testing the left channel with its original SAP15's.
Interesting to note how I originally started with a DC offset of around 60mV and it's gone down 1.0mV!! That's quite a difference! Next up, I plan on hooking up a pair of cheap speakers to see if the amp plays music cleanly with zero hum. I then plan on stress testing it with an 8ohm 200watt resistor and measuring the bias again as per your suggestions. I'm beyond happy that I got it to this point versus giving up and sending it to the 'professionals' for months with zero results!
I know I may be celebrating a bit early, but thanks to you guys for everything! I've learned a hell of a lot.
Last edited:
Awesome - I think you learnt more about electronics, the amp and its design that you were originally expecting with the recap!
Good to see you've got there in the end
The show's not over folks!
Put it all back together and no sound on right channel. Well, at least I know that I'll never need to buy a lottery ticket because that is proof I have terrible luck!
The good is news is there is zero hum from both channels and the left channel works perfectly. I did swap the RCA cables (and then the speaker cables) to confirm that there is no sound being played on right channel. I measured the output transistors voltages and the left working channel shows the voltages 'dancing' as it should as music is being played. The right channel is steady however, and I suppose that indicates the signal isn't getting through somewhere.
At first I thought it was the relay, but I have confirmed they both 'click' on and off. Which leads me to again ask for your guys help. Could the very low DC bias be a clue to the problem as the left channel still shows around 50mv while the right is at 0.5mV - 1.0mV... My only other hunch is that it may be TR8 (ZTX605) which I did not replace as I could not source one - what do you guys think? Otherwise I'll just have go on and literally replace every other transistor that I did not get to.
Time for a beer.
Put it all back together and no sound on right channel. Well, at least I know that I'll never need to buy a lottery ticket because that is proof I have terrible luck!
The good is news is there is zero hum from both channels and the left channel works perfectly. I did swap the RCA cables (and then the speaker cables) to confirm that there is no sound being played on right channel. I measured the output transistors voltages and the left working channel shows the voltages 'dancing' as it should as music is being played. The right channel is steady however, and I suppose that indicates the signal isn't getting through somewhere.
At first I thought it was the relay, but I have confirmed they both 'click' on and off. Which leads me to again ask for your guys help. Could the very low DC bias be a clue to the problem as the left channel still shows around 50mv while the right is at 0.5mV - 1.0mV... My only other hunch is that it may be TR8 (ZTX605) which I did not replace as I could not source one - what do you guys think? Otherwise I'll just have go on and literally replace every other transistor that I did not get to.
Time for a beer.
Yes, Musical Fidelity even has an outline of the transistors shape printed on the boards to help with direction. I decided to spend several hundred dollars and get an OWON HDS2102S 3-in-1 Oscilloscope, signal generator, and multimeter. I've never used these before, so here's another skill I'll have try and learn from very little knowledge. It's my understanding that I'll be able to send a signal to the input and trace that signal throughout the circuit to see if the sound wave changes/stops to pinpoint the problem. How exactly I will do this is still something I'll have to learn - but I've gotten this far, so I may as well dig in deeper!
BTW - if anyone can tell me or point me in the direction of what sort of amplitude (voltages) and Hertz, sine wave is best for testing, please let me know.
Nice addition, the scope will likely prove invaluable.
0.894V peak-to-peak (-10dBV) sine wave signal at the RCA input of your pre-ampp @1kHz will be a safe reference for tracing the signal.
PM'd you re the ZTX605.
Well... here's another 'interesting' update to say the least. This has me completely baffled but bear with me to explain exactly what I did, and no, it still isn't working.
I used the signal generator output connected with alligators clips to stripped RCA cable which is plugged into the right channel of my preamp. I connected the amplifier to the preamp as usual with another set of left and right RCA cables and hooked up the 200watt 8 ohm dummy load to the right channel only. Everything looked good and using the probe (negative connected to chassis and probe connected to positive at speaker terminal) I even got a signal at the speaker terminals! I can adjust amplitude of the sine wave with my preamp volume and I bought it up right to maximum without showing any clipping. Here it is with a picture to confirm (yes, that is the broken right channel and you'll notice that this is 'zoomed' out as the voltage position is at 20v AC):
I confirmed everything with my left working channel by taking the signal generator from right channel to the left and I got the same results. Again, adjusting the volume changes the amplitude of the wave (Note: picture was taken when left channel was at nominal volume on preamp and not maximum, but you get the point that it is the same):
I thought "GREAT, let's hook up the speakers". I did just that and the right channel is still out while the left plays music. Here's where it gets 'interesting'. I measured at the back of the amplifier terminals with everything still hooked up and the signal is still showing strong. To be clear, I am not using the signal generator now and I'm playing music watching the sine wave move around to the music being played. Left channel is playing music and I can view the music through the oscilloscope. The right speaker will show the EXACT SAME THING but will not move the speaker and is silent, despite seeing movement on the oscilloscope and also adjusting the amplitude with volume. Why and how isn't this playing music?
Thinking it was the cables, I swapped for another pair to be used on my speaker. Nothing. Swapped out my speaker from the left working channel to the right, nothing. So it's not the speakers and not the cables (and yes, I even tried another really crappy speaker to confirm this).
I'm not sure what sort of curse is upon me but this is truly baffling why the amp is receiving and outputting a signal but doesn't seem to drive the one speaker... does this make any sense to you guys?
I used the signal generator output connected with alligators clips to stripped RCA cable which is plugged into the right channel of my preamp. I connected the amplifier to the preamp as usual with another set of left and right RCA cables and hooked up the 200watt 8 ohm dummy load to the right channel only. Everything looked good and using the probe (negative connected to chassis and probe connected to positive at speaker terminal) I even got a signal at the speaker terminals! I can adjust amplitude of the sine wave with my preamp volume and I bought it up right to maximum without showing any clipping. Here it is with a picture to confirm (yes, that is the broken right channel and you'll notice that this is 'zoomed' out as the voltage position is at 20v AC):
I confirmed everything with my left working channel by taking the signal generator from right channel to the left and I got the same results. Again, adjusting the volume changes the amplitude of the wave (Note: picture was taken when left channel was at nominal volume on preamp and not maximum, but you get the point that it is the same):
I thought "GREAT, let's hook up the speakers". I did just that and the right channel is still out while the left plays music. Here's where it gets 'interesting'. I measured at the back of the amplifier terminals with everything still hooked up and the signal is still showing strong. To be clear, I am not using the signal generator now and I'm playing music watching the sine wave move around to the music being played. Left channel is playing music and I can view the music through the oscilloscope. The right speaker will show the EXACT SAME THING but will not move the speaker and is silent, despite seeing movement on the oscilloscope and also adjusting the amplitude with volume. Why and how isn't this playing music?
Thinking it was the cables, I swapped for another pair to be used on my speaker. Nothing. Swapped out my speaker from the left working channel to the right, nothing. So it's not the speakers and not the cables (and yes, I even tried another really crappy speaker to confirm this).
I'm not sure what sort of curse is upon me but this is truly baffling why the amp is receiving and outputting a signal but doesn't seem to drive the one speaker... does this make any sense to you guys?
EDIT: Just realized the whole time that the dummy load resistor was cool the whole time. (and yes, it is hooked up properly with one end on positive and the other negative on the terminals) I reconfirmed this a second time just now, and it's from my understanding is that these things should get scalding hot. The amplifier is indeed amplifying the input signal but it seems it cannot draw a current / drive the speaker. Let me know if you can figure this one out!
Huggygood, yes everything was triple checked on the dead channel to make sure I didn't miss anything.
I will tap the right channel with a DC battery and see if anything happens.
Huggygood, yes everything was triple checked on the dead channel to make sure I didn't miss anything.
I will tap the right channel with a DC battery and see if anything happens.
I did the battery trick and it did produce static sound when I tapped the terminals. Signal and connections are all there. This is so bizarre... at first I thought it was the relay is stuck on mute or something of that nature, but the scope shows amplitude change when I turn the volume up or down - it also shows the wave dancing to music when played. If I quickly swap the speaker to the left channel, music is heard. Again, they both show the same response on the scope to volume, music and amplitude change but the right channel is mute.
Other notes: There is little to no DC offset on the right channel even the left has 60mV, maybe this is an indicator of something? Could the relay still be at fault despite everything? Can a faulty transistor produce this strange situation? I've got my eye on that ZTX605 and just changing out every other transistor that I haven't already as a last resort.
Last edited:
Be careful, this is a fairly common mistake, you risk getting lost by doing this.
leave the amp and yourself for a few days then go back to the work methodically, starting from what you are sure of and what works.
Thank you for the concern and I agree I'll need opinions on how to approach this next. It's a very strange situation. It seems the amp is not outputting any current on that channel despite everything. I'm not sure what would be responsible for this and I'm running out of options.
The only thing I can think of is fake darlington SAP15's that I purchased and installed - but that seems like an easy thing to blame. My only other option is to replace the other transistors as I don't see any other approach. I'm ofcourse open to any and all suggestions.
The only thing I can think of is fake darlington SAP15's that I purchased and installed - but that seems like an easy thing to blame. My only other option is to replace the other transistors as I don't see any other approach. I'm ofcourse open to any and all suggestions.
When you quickly tap the 9V battery on the terminals it should make a small spike so the speakers react. It should sound like a loud clear pop. There shouldn't be any sizzling or static other than the pop or any intermittent connection noise for the battery terminal.
If the speaker is attached and no sound comes out and the speaker seems to have a problem getting to maximum voltage swing it could be the speaker is drawing too much current due to a short in a voice coil. However as you've switched both speakers and the same behaviour occurs that sounds more like the amp.
As huggygood has said, give it a couple of days and then come back to it. Impatience is not a good thing with amp repair as it takes a logical step by step analysis of each stage.
Last edited:
My description of "static" was most likely incorrect as I have a really, really cheap speaker attached so it may have sounded like static when it would otherwise have been popping. The speaker is good enough to determine if the amp plays music or not and is used in conjunction with the 200w resistor. The battery application did affect the speaker and it was distinct.
Back to the amp, the fact that when the resistor was hooked up and was cool to the touch, and despite the sine wave dancing when music is being played.... and the fact that the amp seems change the amplitude of the signal, leads me to believe it has to do with a current problem. This is where I'll need advice from professionals (you lot!) as I'm not even sure where to start. I am determined to fix this amplifier but at this point, it's far beyond my expertise to come up with a solution aside from my last resort of replacing all transistors - and even then I'm not sure that would fix what seems to be an ampere issue.
Back to the amp, the fact that when the resistor was hooked up and was cool to the touch, and despite the sine wave dancing when music is being played.... and the fact that the amp seems change the amplitude of the signal, leads me to believe it has to do with a current problem. This is where I'll need advice from professionals (you lot!) as I'm not even sure where to start. I am determined to fix this amplifier but at this point, it's far beyond my expertise to come up with a solution aside from my last resort of replacing all transistors - and even then I'm not sure that would fix what seems to be an ampere issue.
The dummy load should get hot. So i agree there’s a problem.
The voltage swing is needed to cause the current flow. V=I*R even if they’re BJT there should be some predictable resistance so measuring voltage to probe the current flowing rather than inserting the meter into the circuit.
Power (watts)= I^2 * R so if there’s no current then there’s no power dissipated as heat.
So i think you’re on the right track.
The voltage swing is needed to cause the current flow. V=I*R even if they’re BJT there should be some predictable resistance so measuring voltage to probe the current flowing rather than inserting the meter into the circuit.
Power (watts)= I^2 * R so if there’s no current then there’s no power dissipated as heat.
So i think you’re on the right track.
With the amp running, dummy loads attached and a 1Khz signal input, start at the input signal. Then work through each device to the output terminals.
BJTs are current devices so you will need to measure the voltages and possibly use the resistances to work out the actual current flowing.
When you get to the transistor - check all the voltages to understand how the current is flowing. From that you should find if there’s a weaker than expected current flow or a unexpected difference (this could he the wrong beta for example, wrong bias, wrong collector/emitter voltages).
Just work through the amp, stage by stage. Make a note on a schematic as you work through.
Sorry for the second post, but the 30 minute edit window expired.
BJTs are current devices so you will need to measure the voltages and possibly use the resistances to work out the actual current flowing.
When you get to the transistor - check all the voltages to understand how the current is flowing. From that you should find if there’s a weaker than expected current flow or a unexpected difference (this could he the wrong beta for example, wrong bias, wrong collector/emitter voltages).
Just work through the amp, stage by stage. Make a note on a schematic as you work through.
Sorry for the second post, but the 30 minute edit window expired.