47uF cap
Checking the board I checked the 47uF caps by putting my VOM across the leads with the meter set on the Kohms setting. I see that 3 of these caps charge and go off of the scale in about 1 second. But one of them shows a very slow charge rate. Even after 3-4 second the meter reading has not gone off the scale. I pulled the cap and did the same test out of circuit and it does the same thing. However, I checked it on a tester and it reads an OK value 43uF. What do you think?
Checking the board I checked the 47uF caps by putting my VOM across the leads with the meter set on the Kohms setting. I see that 3 of these caps charge and go off of the scale in about 1 second. But one of them shows a very slow charge rate. Even after 3-4 second the meter reading has not gone off the scale. I pulled the cap and did the same test out of circuit and it does the same thing. However, I checked it on a tester and it reads an OK value 43uF. What do you think?
Which cap?
Well which ever - the cap may be leaky, and really not work in circuit espcially if it has more voltage across it (than with your VOM).
When things have not just blown up always suspect electolytics - one of the least reliable components. They espcially do not like heat it seems.
Good luck sleuthing
Bill
Well which ever - the cap may be leaky, and really not work in circuit espcially if it has more voltage across it (than with your VOM).
When things have not just blown up always suspect electolytics - one of the least reliable components. They espcially do not like heat it seems.
Good luck sleuthing
Bill
Hi mikebarg,
That cap is a bit leaky it seems. If it bypasses the signal to ground from the feedback network the offset will increase and the gain may drop. May as well change all of those since you have the amp apart. Just make sure you get new caps, magic parts are not needed.
All four parts may not short at the same time. It only takes one protection transistor to short to kill the bias. Current from a failed output out the base is what normally takes these out.
-Chris
That cap is a bit leaky it seems. If it bypasses the signal to ground from the feedback network the offset will increase and the gain may drop. May as well change all of those since you have the amp apart. Just make sure you get new caps, magic parts are not needed.
All four parts may not short at the same time. It only takes one protection transistor to short to kill the bias. Current from a failed output out the base is what normally takes these out.
-Chris
What about Q4 neighborhood?
How did you do with inspection of Q4 surroundings?
In case you have not found a short, here are more things to look for.
Measure the voltage across Q14 collector and emitter in both amplifiers. Normal operating voltage whould be almost 0V.
Measure voltage across the 47uF capacitor connected to the diode and 18K resistor at the emitters of Q1-Q2. Compare for both amplifiers, it should be almost the positive rail voltage.
Rodolfo
How did you do with inspection of Q4 surroundings?
In case you have not found a short, here are more things to look for.
Measure the voltage across Q14 collector and emitter in both amplifiers. Normal operating voltage whould be almost 0V.
Measure voltage across the 47uF capacitor connected to the diode and 18K resistor at the emitters of Q1-Q2. Compare for both amplifiers, it should be almost the positive rail voltage.
Rodolfo
found nothing so far
Comparing reading between the two boards I have not yet found anything. I must be missing something. I am going to go back and be very sytematic in my checks. I am going to take the schematics and number the diode, resistors and caps also, so I can post it and reference which parts I have checked by number.
I will do some voltage checks next.
Comparing reading between the two boards I have not yet found anything. I must be missing something. I am going to go back and be very sytematic in my checks. I am going to take the schematics and number the diode, resistors and caps also, so I can post it and reference which parts I have checked by number.
I will do some voltage checks next.
It is still possible that the Q4 is shorted under working voltage, which may not show up with the meter.
You may want to replace it for a quick check, even with one that is not exactly the same, to see if that is the case. But if you do, make sure to set the bias pot initially to lowest bias as described above.
You could even replace it temporarely with two diodes in series.
A-------C-A------C
Coll---------------Emitter
If that gives you some 1.2V bias, that shows that the circuit is OK and it very probably is the Q4.
Jan Didden
You may want to replace it for a quick check, even with one that is not exactly the same, to see if that is the case. But if you do, make sure to set the bias pot initially to lowest bias as described above.
You could even replace it temporarely with two diodes in series.
A-------C-A------C
Coll---------------Emitter
If that gives you some 1.2V bias, that shows that the circuit is OK and it very probably is the Q4.
Jan Didden
Re: replacing Q4
If I remember well it is npn too, only silghtly better in voltage/power, so it will go OK.
Did you measure and compare resistance readings as I suggested in a previous post?
Did the failing amplifier work OK before this situation?
Rodolfo
Edit: I mean this post
mikebarg said:
If I remember well it is npn too, only silghtly better in voltage/power, so it will go OK.
Did you measure and compare resistance readings as I suggested in a previous post?
Did the failing amplifier work OK before this situation?
Rodolfo
Edit: I mean this post
resistance tests
I compared resistance reading from each board across Q4,Q5,Q6 as well as across the diodes. I did not find any differences in values. I will check the voltage across the 47uF cap you suggested. I currently have Q4 removed. I will have to reinstall it first(or could this board be safely power up with it removed?
I don't know the previous condition of the amp. I bought this from a guy as a non working amp as a project to help develop my trouble shooting skills. I actually thought it was totally not working when I bought it(which would most likely be easier to diagnose than this) but I found one channel was work normally and the other was distorted. Someone had opened the amp and discontect power from the bad channel. He was most likely using it on one channel only. It certainly has turned out to be a bigger challenge than I expected. But, the point is for me to learn so I really appreciate everyone's assistance and patience.
I compared resistance reading from each board across Q4,Q5,Q6 as well as across the diodes. I did not find any differences in values. I will check the voltage across the 47uF cap you suggested. I currently have Q4 removed. I will have to reinstall it first(or could this board be safely power up with it removed?
I don't know the previous condition of the amp. I bought this from a guy as a non working amp as a project to help develop my trouble shooting skills. I actually thought it was totally not working when I bought it(which would most likely be easier to diagnose than this) but I found one channel was work normally and the other was distorted. Someone had opened the amp and discontect power from the bad channel. He was most likely using it on one channel only. It certainly has turned out to be a bigger challenge than I expected. But, the point is for me to learn so I really appreciate everyone's assistance and patience.
Re: resistance tests
***DON'T POWER UP WITHOUT Q4***
You may blow the output or driver transistors.
I suggest you either replace Q4 with a known good (ej TIP 31) or 2 diodes as suggested by Jan as a first step, and check voltage across what were Q4 collector and emitter pads. If nonzero, verify audio changes.
Don't be afraid to ask. We enjoy helping (particularly across a whole continent!!)
Rodolfo
mikebarg said:
***DON'T POWER UP WITHOUT Q4***
You may blow the output or driver transistors.
I suggest you either replace Q4 with a known good (ej TIP 31) or 2 diodes as suggested by Jan as a first step, and check voltage across what were Q4 collector and emitter pads. If nonzero, verify audio changes.
Don't be afraid to ask. We enjoy helping (particularly across a whole continent!!)
Rodolfo
replaced Q4 - futher tests
I replaced Q4 with a tip31. The result was no change in voltage readings. I have begun to look closer at other transistors and diodes and have found something. I removed Q7 and have found a problem. Please verfy that my test would be acceptable. I am testing across the leads with my VOM on the diode test setting.
E-C - looks OK - does not conduct in either direction, B-C looks OK , conducts in one direction. E-B does not conduct in either direction. Looks like it's open, right?
I replaced Q4 with a tip31. The result was no change in voltage readings. I have begun to look closer at other transistors and diodes and have found something. I removed Q7 and have found a problem. Please verfy that my test would be acceptable. I am testing across the leads with my VOM on the diode test setting.
E-C - looks OK - does not conduct in either direction, B-C looks OK , conducts in one direction. E-B does not conduct in either direction. Looks like it's open, right?
need to order
I don't have a 2n3440. I will have to order one. The only place locally that will sell me one over the counter has the same hours as my work. Getting small quantity parts locally sure is difficult. Even in a high tech town like Austin Texas. I'll see what else I can put on the order. I hate to pay shipping and handling just for one transistor.
I don't have a 2n3440. I will have to order one. The only place locally that will sell me one over the counter has the same hours as my work. Getting small quantity parts locally sure is difficult. Even in a high tech town like Austin Texas. I'll see what else I can put on the order. I hate to pay shipping and handling just for one transistor.
Replaced Q7 - found Q10 problem
I got a 2n3440 and installed it. The tech at the place I bought it suggested looking for a bad resistor also. But, in the mean time I discovered that the leg on collector of the socket for Q10 is loose. You can actually move it up and down in the hole on the board. I don't know if this is not conducting and causing part of the problem. I've discovered, one issue that comes up more often on amps used by musicians to gig is they get moved around a lot and can have a problem like this. This board has really tryed my patience, but I'm not ready to give up on it yet.
I got a 2n3440 and installed it. The tech at the place I bought it suggested looking for a bad resistor also. But, in the mean time I discovered that the leg on collector of the socket for Q10 is loose. You can actually move it up and down in the hole on the board. I don't know if this is not conducting and causing part of the problem. I've discovered, one issue that comes up more often on amps used by musicians to gig is they get moved around a lot and can have a problem like this. This board has really tryed my patience, but I'm not ready to give up on it yet.
sounding much better
I re-soldered the socket for Q10 and another resistor that I found one end loose( one I had de-soldered earlier to test). Things are now sounding a lot better. I am only running on very low power right now because I don't have all of the heat sinking reattached and I am concerned about proper biasing. The SWR bias adjust procedure is to adjust while observing the crossover notch on a scope. Well I don't have a scope. Are there accurate alternitive methods? According to the SWR documention over biasing will cause heating up of the amp at ideal. At least I can watch for that and make sure I'm not currently over biased. What will be the result if the amp is under biased?
I re-soldered the socket for Q10 and another resistor that I found one end loose( one I had de-soldered earlier to test). Things are now sounding a lot better. I am only running on very low power right now because I don't have all of the heat sinking reattached and I am concerned about proper biasing. The SWR bias adjust procedure is to adjust while observing the crossover notch on a scope. Well I don't have a scope. Are there accurate alternitive methods? According to the SWR documention over biasing will cause heating up of the amp at ideal. At least I can watch for that and make sure I'm not currently over biased. What will be the result if the amp is under biased?
Final Post
I guess this will rap it up on this one. I have the amp up and playing on both channels. It sounds good and output levels on both channels sound equal. I am running a CD player into the effects return direct which by-passes the preamp. I have run it at moderate volume levels(my speakers can't handle the amps power) and noticed no heating on either amp channel. Thanks to everyone for your help. I hope I am able to contribute in the future.
I guess this will rap it up on this one. I have the amp up and playing on both channels. It sounds good and output levels on both channels sound equal. I am running a CD player into the effects return direct which by-passes the preamp. I have run it at moderate volume levels(my speakers can't handle the amps power) and noticed no heating on either amp channel. Thanks to everyone for your help. I hope I am able to contribute in the future.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.