QSilver
Super Member
From looking at the schematic, I think that's where Pins 5&9 and 6&10 are reading from. each is coming from the 0.5Ohm emitter resistors. That's where I was measuring from. Unless I was measuring it in a wrong way?
QM-800 Help
- Thread starter QSilver
- Start date
From looking at the schematic, I think that's where Pins 5&9 and 6&10 are reading from. each is coming from the 0.5Ohm emitter resistors. That's where I was measuring from. Unless I was measuring it in a wrong way?
QSilver
Super Member
I managed to connect a speaker up to each amp one by one to see how it sounded. The left channels amp board sounds okay and reproduced a 1Khz sine wave without any noticable distortion.
The right channels amp board has audible distortion in both channels. I hooked up a scope but I couldn't see any distortion or issues but it sounds a bit like clipping. I'm going to take a look at the amp boards and check for dry solder joints and at least change the old 4.7k trimmers for new multiturn 5k trimmers to try and get some more stability for now. I checked the power supply caps for ripple too but that seemed to be okay...
Does anyone have any suggestions on how best to get access to the amp board? There are wires on either side - one side is the power to the outputs and the other side is everything else. I think that I'll have to take off the output B+ and B- wires so I can turn the board upside down to work on it or is that a bad idea?
Starting to take it the amp board off its mountings it looks like someone's been in here before with a soldering iron.... melted wires - two of which are the B+ and B- feeds to the front of the board on pins 11 & 14. I'm glad they hadn't melted through!
The right channels amp board has audible distortion in both channels. I hooked up a scope but I couldn't see any distortion or issues but it sounds a bit like clipping. I'm going to take a look at the amp boards and check for dry solder joints and at least change the old 4.7k trimmers for new multiturn 5k trimmers to try and get some more stability for now. I checked the power supply caps for ripple too but that seemed to be okay...
Does anyone have any suggestions on how best to get access to the amp board? There are wires on either side - one side is the power to the outputs and the other side is everything else. I think that I'll have to take off the output B+ and B- wires so I can turn the board upside down to work on it or is that a bad idea?
Starting to take it the amp board off its mountings it looks like someone's been in here before with a soldering iron.... melted wires - two of which are the B+ and B- feeds to the front of the board on pins 11 & 14. I'm glad they hadn't melted through!
Attachments
Last edited:
QSilver
Super Member
While I'm waiting on advice on how to proceed, I took some more measurements around the boards for comparisons. My B+ is 31V and B- is -31V. The only thing I've found so far with a difference to the schematics given value is the emitters on Q1 through to Q4 which read out at 700mV and 730mV on the bad amp board. Is this a clear sign of needing those KSA992's?
QSilver
Super Member
I hope so... When I looked at Noodles old QM-800 thread I saw MTF mention setting the idle current to 100mA but then I think noodle told me that his readings on his at the moment are 20mV so maybe there is something wrong with the documented procedure. If not, then I can't get the unit to 100mV on any channel I don't think, and it does get hot when you start to get nearer to 100mV...
I'm starting to think about whether it might be a good idea to go through at least the amp boards and repalce the transistors - as a longevity measure. If I need to start ordering transistors it might be cost effective to get the whole lot but I'd have to look into it first...
I've found some spare KSA992's in my left over parts so I can put in a new set on the bad amp and see if that effects the emitter voltage...
I'm starting to think about whether it might be a good idea to go through at least the amp boards and repalce the transistors - as a longevity measure. If I need to start ordering transistors it might be cost effective to get the whole lot but I'd have to look into it first...
I've found some spare KSA992's in my left over parts so I can put in a new set on the bad amp and see if that effects the emitter voltage...
Last edited:
The amps are small to medium power. I think the formula for the idles, although for a different amp, is close to the real specs. A 30mA draw is nothing.
What readings do you get on pins 10 - 8, 8 - 6, 9 - 7, 7 - 5?
The 992's can't hurt. They should be HFE matched for best results.
What readings do you get on pins 10 - 8, 8 - 6, 9 - 7, 7 - 5?
The 992's can't hurt. They should be HFE matched for best results.
QSilver
Super Member
On the good amp I get:
10 - 8: 10mV
8 - 6: -12mV
9 - 7: 11mV
7 - 5: -13.5mV
On the bad amp I get:
10 - 8: 17mV
8 - 6: 19mV
9 - 7: 15mV
7 - 5: 17mV
They're quite steady once the amp has warmed up for a minute or so - while I was taking the meaurements the protection clicked in and out a couple times.
I'll try to gain match them - I have an HFE setting on my DMM but not sure how good it is...
I'm still trying to think of a good way to get access to the board... I don't want to cut the wire wraps to the B+ and B- supplies...
10 - 8: 10mV
8 - 6: -12mV
9 - 7: 11mV
7 - 5: -13.5mV
On the bad amp I get:
10 - 8: 17mV
8 - 6: 19mV
9 - 7: 15mV
7 - 5: 17mV
They're quite steady once the amp has warmed up for a minute or so - while I was taking the meaurements the protection clicked in and out a couple times.
I'll try to gain match them - I have an HFE setting on my DMM but not sure how good it is...
I'm still trying to think of a good way to get access to the board... I don't want to cut the wire wraps to the B+ and B- supplies...
Can you unsolder the pins and pull the hole pin with the wire-wrap from the board?
I use a Hakko FR300 to suck most of the solder from the trace connection. Then rest the tip of the Hakko on the top of the pin (Trace side) heat the pin, loosening the remaining solder and push it out (With the Hakko tip) at least to the board trace. Suck the rest of the solder and use pliers to pull the pin from the board. No twisting!
Looking at problematic transistor's:
The 2SC945's in the protection can be replaced with the KSC1845's.
The 2SC983s in the protection can be replaced with the KSC2310's.
Other's?
Amplifier Board:
2SA497's replacement KSC2383 or KSC2690
2SC627's replacement KSC1845
Output's 2SC793's replacement's; Mouser 3N3442's
Turn the idles down to match the good board.
And keep looking
I use a Hakko FR300 to suck most of the solder from the trace connection. Then rest the tip of the Hakko on the top of the pin (Trace side) heat the pin, loosening the remaining solder and push it out (With the Hakko tip) at least to the board trace. Suck the rest of the solder and use pliers to pull the pin from the board. No twisting!
Looking at problematic transistor's:
The 2SC945's in the protection can be replaced with the KSC1845's.
The 2SC983s in the protection can be replaced with the KSC2310's.
Other's?
Amplifier Board:
2SA497's replacement KSC2383 or KSC2690
2SC627's replacement KSC1845
Output's 2SC793's replacement's; Mouser 3N3442's
Turn the idles down to match the good board.
And keep looking
QSilver
Super Member
Did you mean these transistors? - 863-2N3442G
I saw on Noodles thread that MTF recommended 863-MJ21194G
Is one better than the other?
I'm not sure about the posts. I think they're sort of stapled into the board... Looks to be a cut-out square in the pin near the base that the board goes through....?
I am looking at that thread.
The pins look like the "h" style that engage the board with two parts. I'm inclined to belay the pin removal unless really hard pressed.
If MTF recommends the MJ21194's, that's the replacement. The On-Semi can handle alot of power. They are the same price as well.
The pins look like the "h" style that engage the board with two parts. I'm inclined to belay the pin removal unless really hard pressed.
If MTF recommends the MJ21194's, that's the replacement. The On-Semi can handle alot of power. They are the same price as well.
Last edited:
QSilver
Super Member
After removing some solder around pin 15, I can confirm they are as you said, "h" style. Looks like it curls around through the board. I don't think I'll try to move those....
Ok, if I need to, I'll order the MJs then - I've just changed the trimmers on the bad amp board - will try putting in the new KSA992's. I'll try the hfe on my DMM and see if I can get some close matches...
Ok, if I need to, I'll order the MJs then - I've just changed the trimmers on the bad amp board - will try putting in the new KSA992's. I'll try the hfe on my DMM and see if I can get some close matches...
QSilver
Super Member
I've fitted the new trimmers and the matched pairs of KSA992's and powered up - set the DC offset and idle current (20mV) which was much easier now...
Connected up a speaker and I've still got the distortion on both channels. The rear channel (towards the back of the chassis) has no disotrtion at very low signal levels but crackles and sometimes the signal disapears. If you up the signal it distorts and continues to crackle.
The front amp (towards front of chasis) distorts at all levels and breaks up occasionally. I increased the idle current to 30mV to see if that would help but the distortion didn't change....
I checked the voltages present on the new Q1 to Q4 and the emitter voltage is down to 620mV on both.
I then compared readings on their legs with the good amp board and the only real difference I've found is that the Base voltages on the good amp board are almost double what they are on the bad board.
Good amp Q1 B: 60mV
Good amp Q3 B: 61mV
Bad amp Q1 B: 30mV
Bad amp Q3 B: 25mV
Connected up a speaker and I've still got the distortion on both channels. The rear channel (towards the back of the chassis) has no disotrtion at very low signal levels but crackles and sometimes the signal disapears. If you up the signal it distorts and continues to crackle.
The front amp (towards front of chasis) distorts at all levels and breaks up occasionally. I increased the idle current to 30mV to see if that would help but the distortion didn't change....
I checked the voltages present on the new Q1 to Q4 and the emitter voltage is down to 620mV on both.
I then compared readings on their legs with the good amp board and the only real difference I've found is that the Base voltages on the good amp board are almost double what they are on the bad board.
Good amp Q1 B: 60mV
Good amp Q3 B: 61mV
Bad amp Q1 B: 30mV
Bad amp Q3 B: 25mV
Last edited:
QSilver
Super Member
Between pin 7 and ground I have 6mV that slowly drops as the amp warms up. It settles around 0.6mV.
While I had the board out for the trimmers I changed all but the coupling cap. I found that most of the capacitors were out of spec - double their specified capacitance. At least their done now and it did stop the amp from crackling intermitently... still having the same distortion though...
I just thought I'd check the rear amp and at first I had the distortion and then after a couple of crackles it disapeared and the signal sounded clean until I incrased the volume to 0 on the VU on the 20dB scale - I started to hear it coming back... I think that if I had let it play long enough, the distortion would have crackled back in...
While I had the board out for the trimmers I changed all but the coupling cap. I found that most of the capacitors were out of spec - double their specified capacitance. At least their done now and it did stop the amp from crackling intermitently... still having the same distortion though...
I just thought I'd check the rear amp and at first I had the distortion and then after a couple of crackles it disapeared and the signal sounded clean until I incrased the volume to 0 on the VU on the 20dB scale - I started to hear it coming back... I think that if I had let it play long enough, the distortion would have crackled back in...
Last edited:
QSilver
Super Member
Just made a discovery. I was testing voltage over the boards and comparing and as I passed a test tone through the bad amp, I tapped the heatsink and the distortion changed! I kept tapping and eventually gave it a sharp tap in the centre and the distortion vanished. I tapped again and it came back. So I'm guessing I have a dry solder joint there.... going to go reflow the output transistors connections and see if that cures it...
QSilver
Super Member
I've narrowed it down I think.... When I get the distortion, I took a wooden dowel and tapped around the amp. It seems that if I tap D4 (STV-3H) or R40 the distortion dissapears and then creeps back at random. but a tap in either of those palces fixes it. I think if I tap R40 it's more reliable... any suggestions? I'll try reflowing the emitter resistor connections - I already have reflowed both STV-3H on the board.
Swap the parts with ones from a good channel. If the problem moves, problem solved. Start with the resistor. Treat the STV part with extra care. You can't get them anymore. The leads are brittle and break off at the part body. Try not to bend or stress them at all.
Enjoy,
Rich P
QSilver
Super Member
I wasn't sure if it was the emitter resistor, D4 or R39. I repalced R39 and R38 with new and tried agian with no change. I very carefully removed STV-3H as I thought that maybe the legs had cracked or deteriorated but they hadn't and were in good condition. Both tested good and I VERY carefully reinserted them into the circuit. I have heard MTF mention that if the STV's go, the amp can go into meltdown and so I did think that these should be ok... correct me if I'm wrong.
Next, I finally got out my scope and started to probe around. The distortion seemed to appear around the driver/output stage - I attached the probe to the collector of Q10 and I could see one half of the sine wave then the flat off as it dropped to idle and then back up. When I had distortion, I found that the waveform would have a slight kink just over the top of the half of the sine wave. If the distortion got any worse and the sound almost cut out, I could see that the wave form stopped coming out of the transistor. I proped around the resistors and components connecting to the transistor on the Emitter, Base and Collector and the input wave forms looked clean at all times. The only place that the waveform distorted was on the collector of Q10. If I tap Q10 directly the distortion does alter or disapear... I noticed that the distortion was easier to hear if I dropped the audio generator to around 320Hz instead of 1kHz...
I also probed the collector of Q14 and I could see a completed waveform the whole time which I didn't fully understand.... but I am guessing that if the input to Q10 looks clean and it does in the loew stages of the amplifier, that the driver stages might need replacing? I should be able to get the replacement sets tomorrow. Does that sound like a good next move?
Again, if you suspect a component, borrow one from a good channel. See if you fixed it. If you did not, you can put it back the way it was and carry on. Transistors can be vibration sensitive when going bad.
Rich P
Rich P