I just installed Q815, Q816 and Q821. I did that not long after the surge occured. The powwer supply caps and diodes are all new. D803 is non exsistent. It has been omitted and has never been present. Only getting 12vdc across ZD817. 21vdc across ZD816 and 23.5 across ZD815. The paet number is XZ225. I cant find much on it although I read a blog where it was thought that it was a 22 volt zener but after further analysis it was determined to be a 25 volt zener. I am a little hesitant to order anything at this point
The Zener is critical because it is the reference voltage used to set the final voltage. We know the negative side seems OK with -55v on the emitter.
The Zener in the positive reg needs to be the same voltage as the neg one. 21 volts across the Zener sounds about right.
The maths works as well as proof. This is the negative reg. We ignore the polarity for this.
1. Voltage across Zener = 21 volts. Voltage across R824 (22k) is 21 + 0.6 volts (B/E voltage of Q820).
2/ Current in R824 is 21.6/22000 which is 0.99 milliamps.
3/ Same current flows in R822 (33k) developing a voltage of 0.99ma * 33000 which is 32.7
4/ Add the two which is 21.6 + 32.7 and we get the output of the reg which is 54.3 volts.
So the figures agree closely with what you measure.
Doing the same for your measurement of 23.5v across the Zener gives a rail voltage of over 60 volts
If you have fitted a 25 volt Zener you can increase the value of the 22k or reduce the value of the 33k. You can try a 27k for R821 to reduce the voltage a little or alternatively fit a 27k for R823. One or the other, not both together.
----------------------
The 12 volt rail needs a determination on current draw.
Is R803 OK?
Is R803 hot?
What voltage do you measure across R803?
What voltage do you see on the collector of Q821?
And if you isolate connection '22' does the 12 volts come up on the transistor? with around 25 volts on the collector?
The Zener in the positive reg needs to be the same voltage as the neg one. 21 volts across the Zener sounds about right.
The maths works as well as proof. This is the negative reg. We ignore the polarity for this.
1. Voltage across Zener = 21 volts. Voltage across R824 (22k) is 21 + 0.6 volts (B/E voltage of Q820).
2/ Current in R824 is 21.6/22000 which is 0.99 milliamps.
3/ Same current flows in R822 (33k) developing a voltage of 0.99ma * 33000 which is 32.7
4/ Add the two which is 21.6 + 32.7 and we get the output of the reg which is 54.3 volts.
So the figures agree closely with what you measure.
Doing the same for your measurement of 23.5v across the Zener gives a rail voltage of over 60 volts
If you have fitted a 25 volt Zener you can increase the value of the 22k or reduce the value of the 33k. You can try a 27k for R821 to reduce the voltage a little or alternatively fit a 27k for R823. One or the other, not both together.
----------------------
The 12 volt rail needs a determination on current draw.
Is R803 OK?
Is R803 hot?
What voltage do you measure across R803?
What voltage do you see on the collector of Q821?
And if you isolate connection '22' does the 12 volts come up on the transistor? with around 25 volts on the collector?
I mixed up R821 and R823 but it doesn't matter because swapping either for 27k is what we do for a 25 volt Zener. Going cross eyed looking at the circuit and typing at the same time 🙂
Ill just go with the 22 volt Zener because your proves that they are 22 volt. I will check the resistor R803.
All the power supply parts have been changed. That's the first thing I did when the surge occurred. The only thing I didn't change was the bridge rectifiers but I do have a supply of those if you think they might need to be swapped out
All the power supply parts have been changed. That's the first thing I did when the surge occurred. The only thing I didn't change was the bridge rectifiers but I do have a supply of those if you think they might need to be swapped out
R804 has 425mv across it. 31vdc at either side of it
28vdc on the collector
11.6 on emitter
12.2 on the base
These measurments are still with the forced zero bias
28vdc on the collector
11.6 on emitter
12.2 on the base
These measurments are still with the forced zero bias
R803 🙂 That all sounds good at this point. Sounds like the low readings before were due to the raw supply being low because of the bias current pulling things down.
So all good so far as long as you can get the 56 volt reg up to speed. Once all the supplies are OK we can then see where the rest of it is at.
So all good so far as long as you can get the 56 volt reg up to speed. Once all the supplies are OK we can then see where the rest of it is at.
Replaced ZD815 and ZD816 as well as R821 and R823 because I had them in my parts bin. Removed the jumpers to get zero bias. Made no difference. Getting 44 vdc both pos and neg. Will keep testing and replacing parts as needed.
I just remembered some time ago I had removed and replaced the differential unit on the left channel. I replaced all 4 transistors with KSC1845 and KSA992 transistors. I am going to pull that board and double check that I put it back together correctly. Could that pull that voltage down??
Be logical 🙂 Wasn't one of the two OK a few days ago?
Have you enough voltage going into the regs? What is the collector voltage of Q815 and 816?
What voltage do you actually measure across the Zener's?
Plugged directly into a power source and not the DBT.
21.34 vdc across both Zeners.
Q815 78.2 vdc at the collector
Q816 79.8 vdc at the collector
And what of the emitters as those are the stabilised outputs? The manual shows -/+56 volts as the target value (in the block diagrams). 54 to say 58 volts would be acceptable. Are they in the right ball park is one of them off by miles like it was before.
I had to go back to DBT as the outputs were getting very hot and it blew a fuse. Ill never do that again!!! Damnit
Again as I am learning, why would the collector voltage sky rocket like that from 44vdc to darn near 80vdc by going into the direct voltage source?
I feel dumb asking these questions but I understand the DBT is sucking up current flow that the regulator is drawing but what in the regulator is drawing that much current or perhaps its not the regulator doing that and something on one of the main boards is a miss?
I feel dumb asking these questions but I understand the DBT is sucking up current flow that the regulator is drawing but what in the regulator is drawing that much current or perhaps its not the regulator doing that and something on one of the main boards is a miss?
Read this topic diagonally swiftly, but no info about U701, which is in the amplification loop and also (obvious) connected to the supply rails.
Will read less tilted tomorrow.
Will read less tilted tomorrow.
The collectors of the regulators are fed of 'raw' but rectified AC from the transformer. The collector voltage needs to be quite a bit higher than the desired output voltage from the regulator. Here we have a nominal 80 volts 'raw' input voltage and a stable target value for the regulator output of 56 volts.
The raw voltage will vary a lot with a DBT. If the bulb drops say 30 volts across it because of current draw anywhere in the amp and glows dimly then that means the AC voltage applied to the transformer is reduced by that amount. So on 120 volt mains the transformer is seeing only 90 volts. This means all the secondary voltages are reduced by this percentage fall of 25% and so the rectified DC is also reduced.
In practice the regulator output should be stable at 56 volts from a collector voltage of say 65 volts and higher.
What we are looking for is the correct output voltage from the regulator.
If you had 80 volts on the collector then all that matters at this point is checking quickly the output was correct at 56 volts on each transistor Q815 and 816.
If you had forced zero bias then this should not occur unless either there is a problem with the outputs themselves or you had a load attached and there was a significant DC offset.
Always test with no load attached.
If you have it configured as in post #15 the outputs should stay cold under any test condition and no matter what DC fault may or may not be present.
I'm blind.
R712 (Direct short, +/-20mV) and R703 (Direct open, +/-20mV) are both involved in the offset adjustment.
If 1.62 is the result, the current from the differentials (U701, ahum) from pins 7 and 2 are different (should be equalled by R712), or the input offset from R703 cannot set on zero volt.
Did the 134's survive? You're sure?
The offset adustments seems to work (lowest you can get), the 134's can leak some current causing the output to rise. And can heat up too. This current from the 134's has to go somewhere and the (closed loop) amplifier tries to correct this by shifting the gate voltages.
Absolutely!
No current could flow through the mosfet's in this condition. So why are the mosfet's heating up? One, both, all?
R804 is a bleeder to discharge the neg main supply rail for the right channel, should be 58V across.
Nothing save bad cap's in the regulators.
The main supplies (to the mosfet's) are separated, the higher supplies rails for the main boards are shared.
Can you disconnect the main supplies to the faulty left channel (in accord with #15!) ?
Same thoughts as #35 here.
Thanks for the help guys. Unfortunately I had to take a break from the hobby. My father in law was in a bad accident and is now paralyzed from the neck down. Prognosis is not so good. My wife had to leave town to go help dad and home alone with my boys. That takes precedence and I had a long awaited project that I had to put a cap on. 2004 Jeep Grand Cherokee is finally back on the road after being parked for 2 years.
I will get back at it in the next 2 weeks after the situation with dad is more settled
Phil
I will get back at it in the next 2 weeks after the situation with dad is more settled
Phil