Yes the R129 voltage drop points to a problem. I'd pull that transistor Q119 and try to power it up again. Without a signal Q117 can handle operating the amp by itself without Q119 in parallel.
I pulled Q119 and found a little spec of solder in the small area between the silpad and the theaded hole in the heatsink. I think that was shorting the back metal tap of the transistor to chassis.
With it removed, the amp powers up as normal. Keep in mind the right channel is still disconnected. I tested the removed Q119 and it appears nominal. The rail is now at 78.5v.
Then I powered down, reinstalled Q119, measured OL between the transistor tab and the heat sink, and connected it back up and powered it on. The DBT remains slightly slight, and I measured the rails and they're about +/-45v. It makes me wonder if the silpads are just squashed down and the back tab is still making a tenuous connection on Q119. Or perhaps Q119 is slightly blown? Thoughts?
With it removed, the amp powers up as normal. Keep in mind the right channel is still disconnected. I tested the removed Q119 and it appears nominal. The rail is now at 78.5v.
Then I powered down, reinstalled Q119, measured OL between the transistor tab and the heat sink, and connected it back up and powered it on. The DBT remains slightly slight, and I measured the rails and they're about +/-45v. It makes me wonder if the silpads are just squashed down and the back tab is still making a tenuous connection on Q119. Or perhaps Q119 is slightly blown? Thoughts?
I wouldn't expect any added current draw with Q119 installed or not, so I think if the bulb is glowing slightly with it installed and not glowing at all with it out of the circuit then Q119 is probably bad.
I'd remove Q119 and try to begin biasing up the left channel to see if its going to come up properly. Leaving the DBT in circuit you'll not be able to fully bias it up, but you can at least test it before taking the DBT out of the circuit. Make sure Q126 is in-circuit because the bias test point relies on that transistor being installed.
Put you meter on TP11 and begin turning the bias potentiometer up slowly. Once you start to see a rise in voltage at TP11 slowly bring the bias up to say 5 - 10mV. The DBT will begin glowing. Now check your DC offset at TP12, is it still good like < 50mV? If you check the DC rail now its going to be low so don't worry about that. I'd let the amp sit there for a good half an hour at 10mV still on the DBT and let it idle for a while. Just to see if anything goes poof while under a load.
If after a half hour of idling you still have reasonably low DC offset then you can take the DBT out of circuit and bring the left channel bias up to its full operating level of 40mV.
Then its time to turn your attention back to the R ch...
If your lucky then it was just not coming up due to low DC rail voltages and the higher voltages will allow it to come up. You'll just have to reconnect it and check again. You may need to turn the L ch bias back down while you're testing the R ch on the DBT.
I'd remove Q119 and try to begin biasing up the left channel to see if its going to come up properly. Leaving the DBT in circuit you'll not be able to fully bias it up, but you can at least test it before taking the DBT out of the circuit. Make sure Q126 is in-circuit because the bias test point relies on that transistor being installed.
Put you meter on TP11 and begin turning the bias potentiometer up slowly. Once you start to see a rise in voltage at TP11 slowly bring the bias up to say 5 - 10mV. The DBT will begin glowing. Now check your DC offset at TP12, is it still good like < 50mV? If you check the DC rail now its going to be low so don't worry about that. I'd let the amp sit there for a good half an hour at 10mV still on the DBT and let it idle for a while. Just to see if anything goes poof while under a load.
If after a half hour of idling you still have reasonably low DC offset then you can take the DBT out of circuit and bring the left channel bias up to its full operating level of 40mV.
Then its time to turn your attention back to the R ch...
If your lucky then it was just not coming up due to low DC rail voltages and the higher voltages will allow it to come up. You'll just have to reconnect it and check again. You may need to turn the L ch bias back down while you're testing the R ch on the DBT.
With Q119 removed, the DBT dims completely. Measuring TP11, it stays at 2mV and doesn't budge as I adjust the pot. TP12 measures 0.8v. Q139 is in-circuit, I haven't touched that other than removing it from the heat sink when removing the PCB. Think Q139 could be blown?
Check the voltage at the bases of Q116 & 119 and see if its going up with the potentiometer adjustment. It needs to get up well above +/- 1V in order to start bringing the bias current up in the output transistors.
If Q139 were blown or shorted from E to C it would present a short across polarity of the driver transistor Q116/119 bases so the voltage difference would be 0 between them. Its always possible somethings going on there, but go ahead and check the Q116/119 base voltage adjustment range with the bias pot.
If Q139 were blown or shorted from E to C it would present a short across polarity of the driver transistor Q116/119 bases so the voltage difference would be 0 between them. Its always possible somethings going on there, but go ahead and check the Q116/119 base voltage adjustment range with the bias pot.
To measure the base of Q119, I'm measuring on the transistor side of R163. I see it vary a bit as I adjust the pot, from about -1.10v to -1.13v.
To measure the base of Q116, I'm measuring on the transistor side of R122. I see it vary a bit as I adjust the pot, from about 1.13 to 1.16v.
With the amp powered off, I measured bit bias pot and it seems to stuck at about 0.5 ohms no matter how many times I turn it. This is one of those multi turn Bourns pots, and I turned it a bunch and it's not budging more than a 0.2 ohms or so. I'm positive I've measured it in-circuit before and tried to settle it at about 500 ohms when I first powered on. So I find that very strange. I have the old pot I could throw back in I guess. Could something else in parallel be shorting it out?
To measure the base of Q116, I'm measuring on the transistor side of R122. I see it vary a bit as I adjust the pot, from about 1.13 to 1.16v.
With the amp powered off, I measured bit bias pot and it seems to stuck at about 0.5 ohms no matter how many times I turn it. This is one of those multi turn Bourns pots, and I turned it a bunch and it's not budging more than a 0.2 ohms or so. I'm positive I've measured it in-circuit before and tried to settle it at about 500 ohms when I first powered on. So I find that very strange. I have the old pot I could throw back in I guess. Could something else in parallel be shorting it out?
Make sure that the rails are fully discharged when you're measuring the resistance of the bias adjustment pot. If they're fully discharged then it sounds like that pot is probably bad.
Yes throw the old pot back in if you still have it.
Those base voltages are just under what you need to start biasing the output stage. You'll need about two diode drops worth of voltage there, so about + & - 1.2V give or take a few hundred mV.
Yes throw the old pot back in if you still have it.
Those base voltages are just under what you need to start biasing the output stage. You'll need about two diode drops worth of voltage there, so about + & - 1.2V give or take a few hundred mV.
BTW I forgot to tell you before, nice work finding that solder blob shorting out your Q119.
Those kind of problems can be a bear to find sometimes. I was scratching my head wondering how you could have close to 0V at that point in the circuit and .35V on the other side of that 1 ohm resistor but a short from the collector to ground (the heatsink) will sure do it.
Keep plugging away at it, you're almost there.
Those kind of problems can be a bear to find sometimes. I was scratching my head wondering how you could have close to 0V at that point in the circuit and .35V on the other side of that 1 ohm resistor but a short from the collector to ground (the heatsink) will sure do it.
Keep plugging away at it, you're almost there.
Alright, I swapped out the pot to the old one and measured the resistance and it seems reasonable. I also found something interesting: one of the wires going to the little circuit board holding Q139 had broken, so Q139 was probably not in-circuit in my last measurements. I went ahead and fixed that. I suppose all of the wear and tear of installing and removing the PCB is taking its toll.
With only the left channel installed and Q119 still removed, the DBT dims completely. TP11 still doesn't really adjust. It powers up at around 2mV and stays there. I measured TP12, and when I adjust the pot I see it vary from 0.28v to 0.46v.
Measuring the base voltages again while adjusting the pot, I get:
Q119 base: -0.777v to -0.592v
Q116 base: 0.612v to 0.806v
With only the left channel installed and Q119 still removed, the DBT dims completely. TP11 still doesn't really adjust. It powers up at around 2mV and stays there. I measured TP12, and when I adjust the pot I see it vary from 0.28v to 0.46v.
Measuring the base voltages again while adjusting the pot, I get:
Q119 base: -0.777v to -0.592v
Q116 base: 0.612v to 0.806v
You're probably correct on the bias transistor probably being out of the circuit before. Especially now that your bias voltage readings have dropped significantly.
I'm not sure why the bias voltages aren't responding. The DBT dropping the rail voltage may be messing with it, but I would expect it should still be able to bias up. What is the rail voltage right now?
If the biasing transistor was out of circuit and you were only getting +/- 1.1V before reconnecting it then that was your max bias voltage. It will not get better than that.
If the dc rails are still significantly low then you may want to turn the bias setting to its minimum level and take it off of the DBT and bring it up connected directly to the mains to see how the bias voltage responds with the higher DC rail volts. If the DC rail voltage is only a little bit low right now when on the DBT, like say 5 - 10V low, then this will not correct it either. We'll have to look elsewhere.
I'm not sure why the bias voltages aren't responding. The DBT dropping the rail voltage may be messing with it, but I would expect it should still be able to bias up. What is the rail voltage right now?
If the biasing transistor was out of circuit and you were only getting +/- 1.1V before reconnecting it then that was your max bias voltage. It will not get better than that.
If the dc rails are still significantly low then you may want to turn the bias setting to its minimum level and take it off of the DBT and bring it up connected directly to the mains to see how the bias voltage responds with the higher DC rail volts. If the DC rail voltage is only a little bit low right now when on the DBT, like say 5 - 10V low, then this will not correct it either. We'll have to look elsewhere.
After thinking about it for a bit I have another thing for you to check. The fact that TP12 is moving when you're adjusting the bias pot is not a good thing. It makes me wonder if something is wrong with another one of the Q116 - 119 drivers.
Pull one end of R126, R127 & R128. I know Q119 still removed so you don't need to pull its resistor.
Now use your diode check on your DMM to test the Base to Emitter voltage drop on Q116, 117 & 118. Check it both ways. If they all test normal then that's fine. But I expect the B to E drop should be in the 550 - 650mV range for all of them.
Pull one end of R126, R127 & R128. I know Q119 still removed so you don't need to pull its resistor.
Now use your diode check on your DMM to test the Base to Emitter voltage drop on Q116, 117 & 118. Check it both ways. If they all test normal then that's fine. But I expect the B to E drop should be in the 550 - 650mV range for all of them.
The DC rails are 77.7v and -77.7. So, higher than what the schematic says by 3 volts or so, so I'm assuming the DBT is not drooping them.
I wound up having to completely remove R126, R127, and R128. I figured I would measure them while they were out, and they all appear intact at about 1.4 ohms.
Q116 B to E: 0.492v
Q116 E to B: OL
Q117 B to E: OL
Q117 E to B: 0.519v
Q118 B to E: 0.492v
Q118 E to B: OL
So, a little lower than what you were thinking they should be. What do you think?
I wound up having to completely remove R126, R127, and R128. I figured I would measure them while they were out, and they all appear intact at about 1.4 ohms.
Q116 B to E: 0.492v
Q116 E to B: OL
Q117 B to E: OL
Q117 E to B: 0.519v
Q118 B to E: 0.492v
Q118 E to B: OL
So, a little lower than what you were thinking they should be. What do you think?
Because they're all showing low Vbe then they're probably fine. Didn't you say earlier that you thought you still had a couple of new output transistors? If so you should check them to see what the new one measures for the same test.
The 1.4 ohms of those resistors is taking into account your test lead resistance so that too is normal.
When you put the 1 ohm resistors back into the circuit leave out R128 which is the resistor for Q118 which is opposite from Q119 which is removed. This will leave that transistor out of circuit. So you'll only be using Q116/117 for the front end drivers for now. I'm having you do this so as to avoid any imbalance that might result from having two driver transistors in the circuit on one polarity and one on the opposite polarity. It shouldn't matter just for preliminary testing, but just to be safe leave that resistor out for now.
Do you have a 1k ohm resistor? If so install it into the R163 location which is currently a 750ohm resistor in the bias network. This increased resistance should allow for a somewhat higher bias voltage adjustment range. Turn the bias pot to its minimum position and retry powering the amp up with the DBT in circuit.
The 1.4 ohms of those resistors is taking into account your test lead resistance so that too is normal.
When you put the 1 ohm resistors back into the circuit leave out R128 which is the resistor for Q118 which is opposite from Q119 which is removed. This will leave that transistor out of circuit. So you'll only be using Q116/117 for the front end drivers for now. I'm having you do this so as to avoid any imbalance that might result from having two driver transistors in the circuit on one polarity and one on the opposite polarity. It shouldn't matter just for preliminary testing, but just to be safe leave that resistor out for now.
Do you have a 1k ohm resistor? If so install it into the R163 location which is currently a 750ohm resistor in the bias network. This increased resistance should allow for a somewhat higher bias voltage adjustment range. Turn the bias pot to its minimum position and retry powering the amp up with the DBT in circuit.
Ok, I swapped out R163 for a 1.1k resistor (I'm out of 1ks). Then I reinstalled R126 and R127.
The bias is still stuck at 1.9mV. TP12 varies from 0.275v to 0.394v.
Q119 base: -0.740v to -0.592v
Q116 base: 0.618v to 0.775v
My lone new A1492 measures:
BJT-PNP
hFE=112
Ie=6.3mA
Vbe=568mV
The Q119 I pulled out measures:
BJT-PNP
hFE=106
Ie=6.3mA
Vbe=557mV
The bias is still stuck at 1.9mV. TP12 varies from 0.275v to 0.394v.
Q119 base: -0.740v to -0.592v
Q116 base: 0.618v to 0.775v
My lone new A1492 measures:
BJT-PNP
hFE=112
Ie=6.3mA
Vbe=568mV
The Q119 I pulled out measures:
BJT-PNP
hFE=106
Ie=6.3mA
Vbe=557mV
Sorry for the delayed response. I've been off of work this week and catching up on outside projects. Replacing rotting trim pieces and wood on the house today. Why builders ever decided to use masonite as the exterior trim on homes is beyond me. I guess profit is the utmost importance not a long life. I'm guessing that you do not have that problem in Santa Fe. 🙂
It looks like I had you go the wrong way with R163, so if you have something a bit lower resistance than 750 ohms you could try that.
But first,
Check and verify that R124/125 are still measuring 3 ohms. After that see what your voltage drop is across R124/125 with the bias pot at its max position.
Just looking at the manual and Nak is showing +/- 0.8V for the base readings of Q116/117 so you should be there of very close. Back when you had > +/- 1V when the bias transistor was disconnected you should've been dropping some voltage across R124/125. Its possible one or both of those were damaged. They are fusible, so if they get over driven then they will open up.
It looks like I had you go the wrong way with R163, so if you have something a bit lower resistance than 750 ohms you could try that.
But first,
Check and verify that R124/125 are still measuring 3 ohms. After that see what your voltage drop is across R124/125 with the bias pot at its max position.
Just looking at the manual and Nak is showing +/- 0.8V for the base readings of Q116/117 so you should be there of very close. Back when you had > +/- 1V when the bias transistor was disconnected you should've been dropping some voltage across R124/125. Its possible one or both of those were damaged. They are fusible, so if they get over driven then they will open up.
Seriously no worries about getting some stuff done around the house. It's funny, out place has harieplank siding, which is similar to masonite but is hopefully much more durable due to its structure. We'll see. I spent the day (my day off) getting a bunch of landscaping gravel and sand. Very not fun. Ah well.
Both R124 and R125 are 3.3ohms. Do you think it's possible the bias transistor was damaged when it was partially disconnected?
I swapped out R163 for a 560ohm resistor.
The bias is still stuck at 2mV. TP12 varies from 0.303v to 0.695v.
Q119 base: -1.02v to -0.646v
Q116 base: 0.672v to 1.058v
Both R124 and R125 are 3.3ohms. Do you think it's possible the bias transistor was damaged when it was partially disconnected?
I swapped out R163 for a 560ohm resistor.
The bias is still stuck at 2mV. TP12 varies from 0.303v to 0.695v.
Q119 base: -1.02v to -0.646v
Q116 base: 0.672v to 1.058v
Funny our builder used hardiplank on certain areas of the house and then mixed in masonite in others. Needless to say the hardiplank is holding up fine.
What's your DC voltage drop directly across R124/125 with bias at min/max?
I think the bias transistor is okay. It's job is just to essentially turn on more and lower resistance between the pos & neg drive as the temp of the heatsink drives its hFE higher. This will lower the DC bias voltage as heat increases which keeps temp runaway from occurring. The positive temp coeff thermistor they added into that circuit assists the bias transistor. Many amps do not have the thermistor.
What's your DC voltage drop directly across R124/125 with bias at min/max?
I think the bias transistor is okay. It's job is just to essentially turn on more and lower resistance between the pos & neg drive as the temp of the heatsink drives its hFE higher. This will lower the DC bias voltage as heat increases which keeps temp runaway from occurring. The positive temp coeff thermistor they added into that circuit assists the bias transistor. Many amps do not have the thermistor.
bias pot fully right:
R124 rail side: 79.0v
R124 transistor side: 79.0v
R125 rail side: -79.1v
R125 transistor side: -79.1v, but takes a long time to charge up to that voltage, it starts around -50v and slowly ramps up to the rail voltage.
bias pot full left:
R124 rail side: 78.7v
R124 transistor side: 78.7v
R125 rail side: -78.7v
R125 transistor side: -78.5v, no delay in charging up to this voltage
R124 rail side: 79.0v
R124 transistor side: 79.0v
R125 rail side: -79.1v
R125 transistor side: -79.1v, but takes a long time to charge up to that voltage, it starts around -50v and slowly ramps up to the rail voltage.
bias pot full left:
R124 rail side: 78.7v
R124 transistor side: 78.7v
R125 rail side: -78.7v
R125 transistor side: -78.5v, no delay in charging up to this voltage
Thats very odd. So the rail side comes up to -79.1 right away? However the transistor side of the 3 ohm resistor takes time to come up?
So it sounds like Q117 is conducting for a short period of time when you power the amp up. Or the base of one of the output neg polarity transistors is momentarily conducting or breaking down.