what would cause the centre to wander.Ive been through everything and cant realy find much that differs from the good channel
fluctuates anything from -150mv ,back to 0v and up to +50mv
i have fitted a new 20k trimmer as the other was a it loose.
fluctuates anything from -150mv ,back to 0v and up to +50mv
i have fitted a new 20k trimmer as the other was a it loose.
It could be many things. Have any other parts around the front end been changed?
Here is how I would approach it (assuming that the other channel really is OK and steady).
1/ Scope. Prove by measurement that the output is only fluctuating as a slow changing 'DC variable' and not because of HF oscillation.
2/ Measure over time (with a meter) the voltage across both diodes (D601 and D603) to make sure they are steady.
It really could be 'anything'. Q601, Q603 or any of the small caps including C511 could do this. Compressed disc ceramics are always under suspicion for going intermittently leaky. If the midpoint voltage is jumping around then its more likely a component fault, if it is a slow drift from -150 to +50 mv and then it stays put then it may just be on the edge of normal. That would point to a semiconductor rather than a cap or resistor.
Freezer spray dripped onto any suspect part might give a clue.
Here is how I would approach it (assuming that the other channel really is OK and steady).
1/ Scope. Prove by measurement that the output is only fluctuating as a slow changing 'DC variable' and not because of HF oscillation.
2/ Measure over time (with a meter) the voltage across both diodes (D601 and D603) to make sure they are steady.
It really could be 'anything'. Q601, Q603 or any of the small caps including C511 could do this. Compressed disc ceramics are always under suspicion for going intermittently leaky. If the midpoint voltage is jumping around then its more likely a component fault, if it is a slow drift from -150 to +50 mv and then it stays put then it may just be on the edge of normal. That would point to a semiconductor rather than a cap or resistor.
Freezer spray dripped onto any suspect part might give a clue.
D602 (0v to ground)and D604 both stable(its the other channel)
all of the transistors check out ok, i did replace one of the output transitors originaly as it was blown and also R643 open circuit(these always are(blackened) on these so i change them from 1/4 to 1/2 watt
nothing else changed as the voltages around seem ok
so this could be pre amp then?
This is the big question though. Is it slow drift or is it jumping around?
Nothing in the output stage should do this. Preamp can be eliminated by removing C601.
If it is drift and not jumping around then I bet its normal. You could compare voltages across R627 and R628 and check that this is constant. Any drift in the current here will cause the midpoint to drift. If it drifts may be because of matching between Q602 and Q604. Even putting a finger on one of those transistors may heat it enough to shift the midpoint.
There is no easy way to pin it down I'm afraid because everything interacts and the amp being single ended is not designed for tight DC precision. It relies on drift in one semiconductor (normal drift by temperature) being cancelled by similar drift elsewhere that can cancel each other out... but it's not very precise.
Nothing in the output stage should do this. Preamp can be eliminated by removing C601.
If it is drift and not jumping around then I bet its normal. You could compare voltages across R627 and R628 and check that this is constant. Any drift in the current here will cause the midpoint to drift. If it drifts may be because of matching between Q602 and Q604. Even putting a finger on one of those transistors may heat it enough to shift the midpoint.
There is no easy way to pin it down I'm afraid because everything interacts and the amp being single ended is not designed for tight DC precision. It relies on drift in one semiconductor (normal drift by temperature) being cancelled by similar drift elsewhere that can cancel each other out... but it's not very precise.
so its not drifting slowly it is in big jumps, 0-50mv then 50-150mv them 150-20mv etc etc
R627 and 8 only come into play if you use the LAB input.i have tried this and it is no different.
The simpler way to eliminate the pre-amp is just to pull the pre-out / norm-in jumpers.
In my experience with the 3020, it's not normal to have anything like that amount of drift in DC offset - a few mV here or there maybe, but not a 200mV range in (I assume) a few minutes, or "big jumps".
I don't know which schematic is appropriate here (?), but as far as I can see none of my notes from various 3020's refer to R643 being burnt / blackened. If it's the channel where you changed an output, I'd be more suspicious that further components were taken out when the output went.
In my experience with the 3020, it's not normal to have anything like that amount of drift in DC offset - a few mV here or there maybe, but not a 200mV range in (I assume) a few minutes, or "big jumps".
I don't know which schematic is appropriate here (?), but as far as I can see none of my notes from various 3020's refer to R643 being burnt / blackened. If it's the channel where you changed an output, I'd be more suspicious that further components were taken out when the output went.
yep i agree, i have had quite a few with this resistor gone, might just have been the same fault.i have checked all the components directly around and can find nothing realy wrong yet
Jumping around suggests a fault of some kind.
R627 and 628 on the 3020 diagrams I'm looking at are the current setting resistor for the constant current source transistor. The voltage across these should be steady.
so this is the one im working to, its closest to the amp im working on, which is why our R627 and 8 are different
what im trying to work out now is there has to be a break or poor connection somewhere as it jumps around even more if you twist the chassis a bit in either direction, but im blowed if i can find it just yet
what im trying to work out now is there has to be a break or poor connection somewhere as it jumps around even more if you twist the chassis a bit in either direction, but im blowed if i can find it just yet
Differences 🙂
So R630 and R629 on that one.
Faults like this are where you earn your crust now 😉 and where it gets interesting. Very careful measurement is where I would begin. You have to try and prove whether the network around the preset is altering the setting or whether something else is at play.
My choice at this point would be freezer spray dripped on those two transistors in turn and also on the non electrolytic caps around that stage.
If that proves nothing I would probably swap the two transistors out as a test, its quicker than wondering about them 😉
If still getting nowhere I would force a fixed voltage onto the base of Q602 from a low impedance source. Easiest way looks to be to rig up a diode and resistor across the rail, select a resistor to give say 5 milliamp through the diode. That would give a 600mv reference similar to D602.
If the voltage still jumped around after doing that then it would look as though the issue was elsewhere, again measurement comes into play and you have to try and find where its falling apart.
At this point freezer is the tool of choice.
So R630 and R629 on that one.
Faults like this are where you earn your crust now 😉 and where it gets interesting. Very careful measurement is where I would begin. You have to try and prove whether the network around the preset is altering the setting or whether something else is at play.
My choice at this point would be freezer spray dripped on those two transistors in turn and also on the non electrolytic caps around that stage.
If that proves nothing I would probably swap the two transistors out as a test, its quicker than wondering about them 😉
If still getting nowhere I would force a fixed voltage onto the base of Q602 from a low impedance source. Easiest way looks to be to rig up a diode and resistor across the rail, select a resistor to give say 5 milliamp through the diode. That would give a 600mv reference similar to D602.
If the voltage still jumped around after doing that then it would look as though the issue was elsewhere, again measurement comes into play and you have to try and find where its falling apart.
At this point freezer is the tool of choice.
so i thought i was getting somewhere nailing it down to Q 601 as all the voltages up to that point were ok and then went wild
i didnt have one of these transistors 2sc1400 so i thought i would swap it over with the good channel temporarily, and it was all good, i was able to adjust the centre, well i thought i could, then on came the bulb light(very bright) so i dont know what happened there and i now have even more problems, so.....thats all for today folks cos ive had enough lol
i didnt have one of these transistors 2sc1400 so i thought i would swap it over with the good channel temporarily, and it was all good, i was able to adjust the centre, well i thought i could, then on came the bulb light(very bright) so i dont know what happened there and i now have even more problems, so.....thats all for today folks cos ive had enough lol
This is the fun of fault-finding, honing your skills.
So it looks as though the 2SC1400 might/maybe/possibly be faulty if it seemed OK when swapped out... or... the heat from the iron temporarily 'fixed' something like a flaky C614 or C618. It happens. This is where freezer is so useful in faults like this.
It might be worth setting the bias current low while working on it.
So it looks as though the 2SC1400 might/maybe/possibly be faulty if it seemed OK when swapped out... or... the heat from the iron temporarily 'fixed' something like a flaky C614 or C618. It happens. This is where freezer is so useful in faults like this.
It might be worth setting the bias current low while working on it.
i did actualy use the freezer spray as you suggested and the voltages went down, thats why i took this out.
TBH i wouldnt normaly even consider swaping parts, i learned my lesson years ago doing that with mechanical things
i just didnt want to turn it back on again without that one back in the other channel as i didnt know what damage it might have done without anything fitted.
still whats done is done, ill carry on tomorrow, but now i have even more to do ,but i wont give up👍
Voltages changing as you cool or heat semiconductors is normal, if they suddenly go crazy though and jump up and down then that is not normal. With freezer try and practice pressing the nozzle so little that you get single drops of the stuff dangling of the straw. A few of those drops on the transistor or cap is more than enough.
If you spray a part it can cause issues with condensation (which is conductive) and that can give weird effects.
Swapping parts is usually the last resort but sometimes it can be the quickest option to prove or disprove something.
Faults like this are usually one offs and sometimes you have to devise ways to try and prove things.
If you spray a part it can cause issues with condensation (which is conductive) and that can give weird effects.
Swapping parts is usually the last resort but sometimes it can be the quickest option to prove or disprove something.
Faults like this are usually one offs and sometimes you have to devise ways to try and prove things.
am i right in that if i remove fuses 3 and 4 (either causes the lamp to come on) that i can still check acuratly the lower voltage side
now i have checked all the transistors on both sides this morning and all are ok with basic checks.
i replaced the ones i swapped last night temporarily with some BD139's that are close enough for now, but the originals also check out ok basicaly.
i wanted to assertain that the low voltage side of things is ok first or does removing those fuses (3 and 4) prove it is only the high rails that are at fault?
i dont realy want to test with the lamp on ,as bright as it is, or is this ok?
im convinced it is something simple and i am determined not to end up going round in circles, which is what tends to happen with me eventualy, especialy after working on it all day.
now i have checked all the transistors on both sides this morning and all are ok with basic checks.
i replaced the ones i swapped last night temporarily with some BD139's that are close enough for now, but the originals also check out ok basicaly.
i wanted to assertain that the low voltage side of things is ok first or does removing those fuses (3 and 4) prove it is only the high rails that are at fault?
i dont realy want to test with the lamp on ,as bright as it is, or is this ok?
im convinced it is something simple and i am determined not to end up going round in circles, which is what tends to happen with me eventualy, especialy after working on it all day.
is there a way to isolate one channel from the other power wise? other than cutting track? if i can at least determine which side is the issue its a start
so update-one thing i didnt check this morning was the output transistors and the right pair had blown,how i dont know, must have been that swap last night
anyway replaced both of those and now i have no lamp on but -25v at the rear, so at least i now have something to go on
the right channel has -1.5v, but at least thats a stable voltage, so again ill find out whats causing that and let you know later 👍
anyway replaced both of those and now i have no lamp on but -25v at the rear, so at least i now have something to go on
the right channel has -1.5v, but at least thats a stable voltage, so again ill find out whats causing that and let you know later 👍
If the low voltage rails come up OK then yes you can check those. They should do and the regulated supplies should be correct.
If the current causing the lamp to light is because the output transistors are conducting then they should be fine, they can handle that level of power as long as they don't physically get to hot. If the current lighting the bulb is flowing elsewhere (not in the outputs) then it may or may not still damage something else... it all depends what is pulling the current. Most times the outputs are conducting. Easy to check, just measure the current as you would for a bias check. If its high then the outputs are conducting. If its low then its something else.
Probably not easily but if you know the outputs and the drivers are OK then we can say with certainty that if you over ride the vbe multiplier then the outputs and drivers can not conduct current from rail to rail.