| yup |
I have an old Soundstream amp that I am in process of repairing and I need some help.
I have had this amp in a box for so long that I can't remember exactly how it quit working. Over the past several months I have tried to get it up and running and I still can not get it working correctly.
So far I have checked everything on the board (3 times). I have replaced all the transistors and diodes in both the power supply and the amplifier sections. I have pulled and checked each of the capacitors with a capacitance meter and checked each of the resistors as well. I have check and re-checked the pin out configuration of the transistors. I must say I am getting somewhat frustrated.
I currently have this amp on a bench power supply and when I first tun it on, the current draw on the supply is right at 4 amps (13VDC). It slowly drops to around 3.8 amps as the heat sink warms up (this is a class A amp so it draws alot of current at idle)
The amp idles just fine like this. It gets fairly warm, but not too hot to touch. The problem is that the amplifier section does not work. I can verify that the input signal makes it all the way to the diff pair, but no further (on O-scope)
I have noted some of the measurements on the attached schematic.
The voltage rails have sagged to +17.37VDC & -18.4VDC with visibly more ripple on the plus side to ground. The 1 ohm resistor from the TIP 142 Emittor to ground is hot to the touch, while the one on the TIP147 is not.
Both the left and right channels behave identically
I am not sure if the problem is in the power supply or the amplifier section. Does anyone know what the current draw should be for this amplifier when it is working properly?
Thanks,
Jeff
CA 50 Schematic |
|
|
| Perry Babin |
The current through the 142 and the 147 should be the same so the temperature of R14 and R15 should be the same (assuming that D3 and D4 are not defective).
If you post the DC voltage on the pins of each of the transistors, it may make troubleshooting easier.
Q1
B
C
E
Q2
B
C
E
Q3
B
C
E
Q4
B
C
E
Q5
B
C
E
Q6
B
C
E
Q7
B
C
E
Q8
B
C
E |
|
|
| yup |
Here are the voltages at the transistors (both channels measure equivalent voltages)
(the voltages vary slightly due to my poorly regulated supply that I am using and the change in voltage as the heatsink temp changes between measurements)
Q8
E 6.8
B 5.7
C -18.4
Q7
E 7.4
B 8.5
C 18.3
Q6
E -17.8
B -17.1
C 5.8
Q5
E 5.8
B 6.4
C 8.4
Q4
E 8.5
B 17.7
C 18.2
Q3
E -17.7
B -17.1
C 6.4
Q2
E 6.5
B 7.1
C 18.2
Q1
E 6.5
B 0
C 17.7
Thanks for any help,
Jeff |
|
|
| Perry Babin |
It looks like either Q1 or Q4 is defective. C3 may also be able to cause the fault.
With the base-emitter of Q1 reverse biased, there should be no current flow through R5 which should make the base-emitter voltage of Q4 0v. With Q4 off, the base of Q7 should be near ground.
The schematic is not correct. A PN4250 is a PNP transistor. |
|
|
| yup |
I noticed the error on Q4 recently as well. The problem was I could not remember what I pulled of the board originally, since it had been several months since I had looked at it the first time.
Currently I have a PN4250 (PNP) in the cricuit. Should that be an NPN (MPSA 42) transistor instead?
Thanks,
Jeff |
|
|
| Perry Babin |
The PN4250 is the correct part for Q4.
What happens if you pull Q1 out of the circuit? |
|
|
| yup |
I pulled Q1 and Q9 out and basically nothing changed. Do I have Q4 in backwards? Should the collector or emittor of Q4 connect to the positive voltage rail (collector of Q7)?
Thanks,
Jeff |
|
|
| Perry Babin |
| The emitter of Q4 goes to the positive rail. The collector of Q4 drives the base of Q7. |
|
|
| yup |
Well, that was the problem. I had Q4 & Q12 in backwards. That's what I get for following an incorrect schematic.
I get music out of one channel now. I probably burned something out in the other one. I will take some more measurements tomorrow and post back.
Thanks,
Jeff |
|
|
| yup |
Here are the voltages now that the amp makes music...
The amp draws around 1.8amps at 13.8 volts (25 watts) when cold, and drops slightly as it warms up. So it appears this amp can not be full class A (part of the 25 watts of current draw is lost in the power supply as well)
I assume I would have to pull D3 & D4 out of circuit to measure the bias current through Q7 & Q8 (voltage across R14 & R15). I was thinking of increasing the bias current some more if the heatsink will allow. I could do this via R10 & R11 correct?
Thanks,
Jeff
Q1
E -.65
B -.04
C 18.74
Q2
E -.65
B -.03
C 19.39
Q3
E -18.73
B -18.16
C -.65
Q4
E 19.51
B 18.83
C 1.37
Q5
E -1.37
B -.74
C 1.37
Q6
E -18.84
B -18.2
C -1.37
Q7
E .253
B 1.36
C 19.57
Q8
E -.192
B -1.36
C -19.60
Q9
E -.63
B -.04
C 19.09
Q10
E -.63
B .025
C 19.7
Q11
E -19.0
B -18.38
C -.62
Q12
E 19.69
B 19.05
C 1.36
Q13
E -1.33
B -.71
C 1.37
Q14
E -19.05
B -18.42
C -1.36
Q15
E .28
B 1.36
C 19.7
Q16
E -.19
B -1.36
C -19.7 |
|
|
| Perry Babin |
It appears to be healthy now.
What was wrong with the other channel?
You don't have to remove the diodes at the present level of biasing. The voltage drop across the 1 ohm resistors is less than the forward voltage of the diodes so no current is flowing through the diodes.
If you increase the value of R10, the bias would increase but I doubt that you'd see any improvement in sound quality. |
|
|
| yup |
Thanks for the Help. The other channel turned out to have been working, I had burned out one of my bench test speakers somehow.
Shouldn't I be seeing the same voltage drop across R14 & R15?
Thanks,
Jeff |
|
|
| Perry Babin |
Ideally, you should see the same voltage across them but if they don't have precisely the same resistance, the voltage drop will be different.
How much DC offset to you have at the speaker terminals? It doesn't appear that you have much looking at the feedback to the differential amp. Set the amp to stereo mode when measuring the offset. |
|
|
| yup |
I see 0.030 VDC & 0.040 VDC on the outputs
Thanks,
Jeff |
|
|
| Perry Babin |
| The DC offset is just a bit higher than I'd expect but it's not going to cause any problems. If you want to try to reduce it, replace the transistors in the differential amp of each channel (Q1, Q2, Q9, Q10). Using transistors that are closely matched is important. If you have a meter that will measure HFE of the transistors, select two with nearly identical gain (HFE). If you don't have a meter, use transistors from the same batch (same production/date codes). |
|
|
| yup |
Well, I have replaced the transistors with pairs from the same production lot and now both channels have basically 0.040 VDC on the output. I then swithced Q1 & Q10 (opposite sides of diff pair) and figured that would give 0 volts on each channel, however I still have the same values on both 0.040 VDC.
I put in a variable reisitor under the emittor on Q1 and I can dial the DC offset to 0 V at about 37.5 ohms. Is there a drawback to this? Is there a better place to adjust the DC offset? Should I leave it?
Thanks,
Jeff |
|
|
| Perry Babin |
Inserting the resistor won't cause any problems. Some amps use a similar system to adjust the offset.
If the bypass caps (C4? and C10) are leaking (electrically) that could cause the DC offset to be higher than it should be.
Is there anything on the board that would make Q1 run warmer/cooler than Q2?
I'd leave it as it is. |
|
|
|
