ok, so here are the measurements. some are close, others are pretty far off. so those that dont have schematic handy, ill put in parenthesis what it SHOULD be.
R108: 6.15 (4-5)
R106: 5.17 (4-5)
Q105: 6.00 (4-5)
Z103: 8.64 (9)
R120: 1.27 (.5)
R121: 1.55 (.5)
R122: 1.39 (.5)
R123: 1.4 (.5)
so there you have it. the power resistors are really far off. they should be a lot lower. any ideas???
R108: 6.15 (4-5)
R106: 5.17 (4-5)
Q105: 6.00 (4-5)
Z103: 8.64 (9)
R120: 1.27 (.5)
R121: 1.55 (.5)
R122: 1.39 (.5)
R123: 1.4 (.5)
so there you have it. the power resistors are really far off. they should be a lot lower. any ideas???
Looks like the input constant current source is working but the zener voltage is a bit on the low side. I suspect its some wiring issue or PCB problem, and I do notice that your input ground does not make it back to the star ground as per your photo unless ther is some other wire hidden, but it should really ground at the star point. Does it connect there? If not use a clip lead to connect it and see what happens.
Wiring mistakes are VERY easy to overlook and when I brought up my first Aleph 2 on the variac I had the rails reversed....
. Oops! It blow one output device and R-15 but that was it. Replaced the parts corrected the mistake and they've been running great for almost two years now.
Although the voltages are close on the input stage it is possible that you may killed an input device..... that can happen. Triple check your wiring as compared to Algar_emi's drawing and if thats all ok then check the traces on the pcb, if good then I'd replace the input devices and see what happens. If you don't get anywhere I'm just a short senic drive to Salt Lake City!
Mark
Wiring mistakes are VERY easy to overlook and when I brought up my first Aleph 2 on the variac I had the rails reversed....
. Oops! It blow one output device and R-15 but that was it. Replaced the parts corrected the mistake and they've been running great for almost two years now. Although the voltages are close on the input stage it is possible that you may killed an input device..... that can happen. Triple check your wiring as compared to Algar_emi's drawing and if thats all ok then check the traces on the pcb, if good then I'd replace the input devices and see what happens. If you don't get anywhere I'm just a short senic drive to Salt Lake City!
Mark
The way you have the input ground connected sounds fine. It also sounds like the input pair is working of they are heating up evenly. The voltages are not all that critical for the circuit to work and its open for a wide variation and alot of experimentation as per Nelson. Getting that zener voltage up to 9 volts would be a good idea and that might bring the input stage readings closer to what they should be. I've had bad zeners before....one when I built me X's did the same thing.... a tad on the low side.
Does the output stage get hot after a short time? With voltage readins that high on the resistirs it almost sounds like a biasing problem, check the parts around and including Q104 and Q105. After this stuff I'm at a loss without it in front of me on the bench to see whats going on.
Mark
P.S. Another unlikely thing that just dawned on me is if your input jack might be shorted?
Does the output stage get hot after a short time? With voltage readins that high on the resistirs it almost sounds like a biasing problem, check the parts around and including Q104 and Q105. After this stuff I'm at a loss without it in front of me on the bench to see whats going on.
Mark
P.S. Another unlikely thing that just dawned on me is if your input jack might be shorted?
ill check that. they do heat up mighty quick, and im my opinion, get far too hot. i dont remember it getting THIS hot before. i could work around it when it was running, but now you will burn yourself on the heatsinks...
i believe the input jack isnt shorted, it a NICE cardas and pretty hard to short.
ill check Q104 and 105. I am not 100% sure HOW ill check them, but i will check them
i know it has a low gain, but i cant even get any readings out of it really... i have a little "thing" that i made up that will measure output power. its basicall a dummy load with meter on it for measuring power output. it doesnt even register when its given a full signal. before the ground issue, it was measuring right at 30 watts.
im almost sure the resistor values are correct. when i had my first problem with the amp, i re-tested all the resistors. they were fine. they havent changed. i bet its a zener or a MPSA18.
is it safe to rule out the FETS and the differential pair for now? but it could still be Q103 though?
i guess im asking this:
since i know the resistor values are ok, and it seems to be wired fine, and its generating an immense amount of heat, we can only conclude its over-biased (maybe?) and the problem is Q103, Q104, or Q105, or a zener?
i believe the input jack isnt shorted, it a NICE cardas and pretty hard to short.
ill check Q104 and 105. I am not 100% sure HOW ill check them, but i will check them
i know it has a low gain, but i cant even get any readings out of it really... i have a little "thing" that i made up that will measure output power. its basicall a dummy load with meter on it for measuring power output. it doesnt even register when its given a full signal. before the ground issue, it was measuring right at 30 watts.
im almost sure the resistor values are correct. when i had my first problem with the amp, i re-tested all the resistors. they were fine. they havent changed. i bet its a zener or a MPSA18.
is it safe to rule out the FETS and the differential pair for now? but it could still be Q103 though?
i guess im asking this:
since i know the resistor values are ok, and it seems to be wired fine, and its generating an immense amount of heat, we can only conclude its over-biased (maybe?) and the problem is Q103, Q104, or Q105, or a zener?
I forgot it was working fine (hum aside) before messing about. Your bias is also way to high. (3X) What's the dc offset look like. That would explain burned speaker. I doubt there is a problem with mpsa. The diff pair is static sensitive and easy to blow. The outputs too, but are all in the same range, which makes me believe they are OK. Recheck what you rewired in the ground rework
the dc offset is fine. i THINK i checked it after all this happened, and i remember it at 7mv, same as before.
the wiring seems like it could be an issue, but i only redid 2 wires really, and ive checked them both at least 10 times (its not a hard thing to check )
its going to be something really stupid, i know. could the diff pair get blown by touching them while inside the circuit? i have heatsinks on them, and im sure i might have touched one of the heatsinks when it was soldered in there... if that could have blown it, its a possibility.
the wiring seems like it could be an issue, but i only redid 2 wires really, and ive checked them both at least 10 times (its not a hard thing to check
)its going to be something really stupid, i know. could the diff pair get blown by touching them while inside the circuit? i have heatsinks on them, and im sure i might have touched one of the heatsinks when it was soldered in there... if that could have blown it, its a possibility.
check q104 and q105 out of circuit. if your multimeter has a diode function set it there..it should have 0.4 to 0.6 V across B-E. with resistance function it should be low ohm one way and high on reverse. check r120 thru r123 too they might be toasty and have drifted in R. with 1.55V each is passing about 3A at 5W. check q106 to q109 the internal diode across D-S should also be 0.6V using diode function.
another update:
i got some time to work in it today, but i dont think im much closer to figuring it out.
this is what i did. i had a second board for the other monoblock, and since the original one is pretty messed up (broken traces, etc from so much unsoldering), i just repopulated a new board with ALL new parts. i tested each part that went in and had an anti-static band on the whole time.
it went together pretty quick. i also did a bit of re-wiring and ground-wire work. its a big neater now. i kicked it on, and it actually works. i hooked up a meter and saw that the DC offset was only 2mv (unloaded). i figured i could hook up a speaker. I did, but it was a bit too soon. i got sound, and it was actually amplifying it (not like before where it was really quiet), but i was getting crazy DC through the outputs. at a low volume, my meter was jumping around 200mv or so. any higher in volume and it would go out of range. this made my poor test speakers smoke again.
the things that havent changed are: the wiring (its just been updated, but nothing is different really), the fets (they are off on the heatsinks, and they are the same), the power supply (which is supplying correct voltage).
so, i assume its the fets. i did some testing using the diode function on my meter that blues suggested. i got consistantly about 0.48v out of them. i had a new batch sitting around that tested about 0.53v. on both sets, all four test very close. i could try swapping out all the fets, but i would like to know for sure first, because its a lot of rewiring to switch them out (i dont want to spend 4 hours working on that to find out its something else).
i havent done any formal measurements yet, becuase i know there will be many things off, and i prefer not to run the amp, or run the risk of shorting something before i get a bit more educated as to what is going on. i did notice when it was on that the power resistors (R120 - R127) are getting VERY hot as blues guessed.
if no one has any ideas, i guess ill go to work swapping out the fets. but since the assumed bad set and the assumed good set measure so close, im skeptical that its the problem. however, maybe .1v is a big deal in this case.
i got some time to work in it today, but i dont think im much closer to figuring it out.
this is what i did. i had a second board for the other monoblock, and since the original one is pretty messed up (broken traces, etc from so much unsoldering), i just repopulated a new board with ALL new parts. i tested each part that went in and had an anti-static band on the whole time.
it went together pretty quick. i also did a bit of re-wiring and ground-wire work. its a big neater now. i kicked it on, and it actually works. i hooked up a meter and saw that the DC offset was only 2mv (unloaded). i figured i could hook up a speaker. I did, but it was a bit too soon. i got sound, and it was actually amplifying it (not like before where it was really quiet), but i was getting crazy DC through the outputs. at a low volume, my meter was jumping around 200mv or so. any higher in volume and it would go out of range. this made my poor test speakers smoke again.
the things that havent changed are: the wiring (its just been updated, but nothing is different really), the fets (they are off on the heatsinks, and they are the same), the power supply (which is supplying correct voltage).
so, i assume its the fets. i did some testing using the diode function on my meter that blues suggested. i got consistantly about 0.48v out of them. i had a new batch sitting around that tested about 0.53v. on both sets, all four test very close. i could try swapping out all the fets, but i would like to know for sure first, because its a lot of rewiring to switch them out (i dont want to spend 4 hours working on that to find out its something else).
i havent done any formal measurements yet, becuase i know there will be many things off, and i prefer not to run the amp, or run the risk of shorting something before i get a bit more educated as to what is going on. i did notice when it was on that the power resistors (R120 - R127) are getting VERY hot as blues guessed.
if no one has any ideas, i guess ill go to work swapping out the fets. but since the assumed bad set and the assumed good set measure so close, im skeptical that its the problem. however, maybe .1v is a big deal in this case.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.