Ampeg B-100 problems

[Jazz P Bass]

At what point in replacing all of those parts did Q4 start overheating?

Or was this an original issue when the amp was first worked on?

[g1]

60mA should not be enough to make that transistor overheat. So if it's not defective, or incorrect orientation, there must be another current path other than R24.

[loudthud]

I calculate 1.56W for Q4. That's a little much for a TO-220 without a heat sink. If you have any kind of bolt on or slip on heat sink, I'd use it. I wonder if D5 is one of those dual diodes. What is the voltage drop across D5?

[The Dude]

The original transistors were germanium which is why the BE voltage is low. The voltage readings numbers in black are are printed on the schematic from Ampeg. The original transistors have been replaced with silicon. I have not taken voltage readings at that point since the switch but they both get too hot. The original Q4 transistor diode checked as good.

?????? You can't just replace germanium transistors with silicon without some circuit modification. They have different operating specs and voltage drops.

[The Dude]

If I may add: all of the joints seem to be lacking in that it appears that the pads themselves where not properly heated.
The pad itself & the component lead have to be above the melting point of the solder or you will not get a proper solder joint.
Moving the iron tip back & forth between the pad & the component lead, while adding in the solder, will go a long way to ensuring a good solder joint.
2 cents.

I agree. Aside from the obvious one I pointed out, the solder in general looks "iffy" on this board. If I were working on it, I'd probably start at one end and resolder the entire board.

[Jetbat]

I was assuming they were germanium because of the low base, emitter voltage. In doing searches for the original part numbers I could not find any datasheets and only one website had equivalent listings for them. That was weisd.com.

610226-1 NTE199
610083-5 NTE159
610153-1 NTE152

The only info I could find for the output transistors was on forums like this with two replacement suggestions that people said worked for them.

610289-2 NTE331 TIP41C
610289-1 NTE332 TIP42C


There is a .762V drop across D5 and a .491V with the diode test.

Jazz P Bass, The amp worked fine and was silent with no hiss or buzz before I worked on it. I only noticed Q4's heat because of the hiss and buzz from the speaker caused me to look around inside the amp. I did a test with an IR thermometer. The temp of Q4 went over 60C in less then 5 minutes so I turned it off. The cabinet does not have any ventilation holes so it is not designed to get hot.

[52 Bill]

It's been a long time since I've worked on one of these, but my recollection is that the two output transistors were TO-220 darlingtons.

I have re-read your original post and it sounds like there really is only one problem, the constant hum or buzz. The additional hum/buzzing that you describe when probing around the base of Q2 sounds like normal hum pickup from the input of the power amp. You stated that the hum was the same with the volume turned up or down, and that the problem was after the volume control. This is true, but the problem could also be power supply related.

I would suggest that you stop replacing transistors, as they are probably not the source of your hum. In fact if you still have all of the original ones, I would meter test them and carefully reinstall them. The more changes that you make, the better the chance that you are going to cause additional problems.

Exactly what parts did you change and what have you changed back? Did you change any of the ground connections when you did the first set of mods? Have you checked to see if any damage was done to the pc board causing bad connections or bad grounds?

When you meter test the transistors, you can tell if they are Silicon or Germanium by the voltage reading that you get on the meter. Germaniums will normally turn on at around 0.3 volts, while Silicons will turn on somewhere around 0.6 volts.

[Jetbat]

I took readings of the original transistors with a multimeter using diode check and Hfe. Numbers in parentheses are transistor numbers on case.

Q1 NPN (226-1) BE .629V, BC .625V, Hfe 268.
Q2 PNP (083-1) EB .644V, CB .635V, Hfe 240.
Q3 NPN (266-1) BE .631V, BC .627V, Hfe 152.
Q4 NPN (153NIC) BE .474V, BC .446V, Hfe 62?
Q5 NPN (289-2) BE .444V, BC .443V, Hfe 329.
Q6 PNP (289-2) EB .511V, CB .506V Hfe 87.

For some reason I could not get a Hfe reading from Q4 with the multimeter so I connected it to a old B&K 501A curve tracer that I am learning. The tracer needs some work but it shows me the transistor is working and I calculated the Hfe 62. That may not be right but Q4 is functioning.

Also Q5 gave me a reading using the diode test with the + on the emitter and - on the collector of 1.52V. I don't know if that is a problem but all the others did not give me any other readings then what I wrote above.

[Jetbat]

The parts changed were resistors- R3, R4, R6, R8, R9, R23, R25, R26, R28, R29 and capacitors- C1, C2, C3, C4, C5, C10, C11, C12, C15, C17, C18. I changed all capacitors back and all the resistors past the volume pot R15. There was a spot where the trace on the PCB was lifting and was glued down possibly in factory. I checked that connection and it is still good. No other parts of the PCB look damaged.

I installed a grounded power cord which is bolted to the case. First thing I did when I heard a buzz was to lift that connection and listen again. Did not change. The continuity is good from the PCB to the ground prong on the cord.

[g1]

As far as your transistor readings go, I would expect closer readings between Q5 & Q6 as they are supposed to be complimentary.
Especially with the hfe.

[J M Fahey]

1) STOP replacing parts at random and with no clue

2) to start, let's concentrate on Q4 overheating.

"The other" current path to it is clearly visible, it's Q6 BE junction.

R29 voltage drop shows some 120 mA passing through it.

If Q6 BE is shorted or Q6 collector is open (or simply disconnected from ground with all this shuffling around) those 120 mA will add to biasing 60mA at Q4 and turn it into a very hot fellow.

So check Q6 Vbe, straight at the transistor legs (I don't trust wiring/tracks/pads/soldering) and voltage at Q6 collector leg (relative to ground) , also at the transistor legs.

Be careful because it's easy to slip a probe and short something

Side note: those output transistors are not Darlingtons, that's why Q4 load resistors are so low (180+180 r) and it works so hot.

[Jetbat]

I would not say that I am replacing parts randomly or that I have no clue what I am doing. All resistors and capacitors originally changed were chosen to lower noise and increase sound quality.

Measurements for Q6 are BE .598V and collector to ground 1.1mV. (taken at the transistor legs)

I had thought of base current from Q6 (being a PNP) but I felt that the majority of the current was coming from R24.

Q4 had been get very hot with the original transistors. Now with all new transistors the heat issue has not changed.

I installed a heat sink on Q4 so I could run the amp longer without fear of damaging it. This allowed me to hook up a signal generator and scope to probe around. I must admit I made a mistake with the ripple voltages when I checked last week. It seems my probe was set to 10x. The ripple voltages are actually B+ 1.2V and .240V at the output. Just to be sure I replaced C18 with a new capacitor and checked again. Still 1.2V ripple on the rail.

A 40mV sine wave on the input gave a 7.6V output with the bass and treble pots set in the middle and volume up full. The sine wave looked good in all sections of the circuit.

Removing the sine wave I started to follow the output ripple around the circuit and marked it in red on the attached picture. Outside of the red areas the scope was flat. Also wherever the ripple was there was also noise (hiss) superimposed on the ripple. I included a picture of a section of the ripple with noise. (100mV divisions) The rising part of the wave has multiple lines. The B+ ripple has this also but lacks the noise. (the lines are smooth) Don't know if this will be helpful or not.

With the heat sink on Q4 I decided to put the amp back in its cabinet and play bass through it. The buzz increases dramatically with the speaker in the cabinet compared to out on the floor do to the lack of cancelation. The amp sound quite good except for the buzz and hiss. Can get loud too but that is maybe due to the more efficient speaker that was installed. (the old one had a shredded cone)

[Jazz P Bass]

Was the speaker hooked up when you did these tests?

If so, try it again without one.

C17 may be leaking.

[Jetbat]

No the speaker was not hooked up for the tests.

I did what the Dude said and re-soldered all the connections past the volume pot. I pull out and tested all the resistors while I was at it. No change.

Then I did a test and I am fairly certain that Q5 and Q6 are oscillating and that is why I have had such a hard time tracking down the problem. My oscilloscope is a 100 Mhz. The oscillation would have to be higher than that since I do not see it on there.

Anyone have any ideas about this?

[The Dude]

......Then I did a test and I am fairly certain that Q5 and Q6 are oscillating and that is why I have had such a hard time tracking down the problem. My oscilloscope is a 100 Mhz. The oscillation would have to be higher than that since I do not see it on there.

Anyone have any ideas about this?

I have doubts about any oscillation above 100Mhz. How did you arrive at the theory that Q5 and Q6 are oscillating?