stay at home type 69 project

[oongbac]

Reflowed parts of output amp, checked resistor value, swapped all the TRs, it's same...

When I put a scope to emitter and base of Q5, it is okay. But at the connection of collector / r39 / r41, there's a distortion. Is this can be a clue?

 

[TwentyTrees]

When I put a scope to emitter and base of Q5, it is okay. But at the connection of collector / r39 / r41, there's a distortion. Is this can be a clue?

That does reinforce the suggestion of an issue with the output amp. Are Q5-7 the right transistors, and the right way round? Worth checking the datasheets for your particular transistors to make sure the pinouts are as you'd expect.

As Matt says, it's a relatively simple circuit with nice discrete sections (2128 amp > EQ > 2128 amp > buffer) so a systematic approach to checking each part in turn is the way to go. If I was in your shoes, my next step would be to run a calibrated test tone through the unit (say -4dBu 1kHz, gain set to +20, pad engaged, EQ disengaged) and then measure both AC and DC voltages at each pin of each transistor. Any differences in DC voltages between the two 2128 amps would be particularly instructive.

It's also worth checking, with the module unpowered, that you don't somehow have continuity between the ground of the EQ section and the V- in the output amp. I'd imagine that would lead to bigger issues than you're seeing here, but worth excluding any cap failures, solder bridges, or similar.

 

[oongbac]

That does reinforce the suggestion of an issue with the output amp. Are Q5-7 the right transistors, and the right way round? Worth checking the datasheets for your particular transistors to make sure the pinouts are as you'd expect.

As Matt says, it's a relatively simple circuit with nice discrete sections (2128 amp > EQ > 2128 amp > buffer) so a systematic approach to checking each part in turn is the way to go. If I was in your shoes, my next step would be to run a calibrated test tone through the unit (say -4dBu 1kHz, gain set to +20, pad engaged, EQ disengaged) and then measure both AC and DC voltages at each pin of each transistor. Any differences in DC voltages between the two 2128 amps would be particularly instructive.

It's also worth checking, with the module unpowered, that you don't somehow have continuity between the ground of the EQ section and the V- in the output amp. I'd imagine that would lead to bigger issues than you're seeing here, but worth excluding any cap failures, solder bridges, or similar.

Thank you very much for kind reply. I’ll check all your suggestions! I run my TRs to TR tester and pinouts are all good. Thanks again for share your insights!

Edit: There's no continuity between ground of EQ and V- in the output amp. But every pins of TRs in output amp has lower DC voltage I must keep digging...

 

[TwentyTrees]

But every pins of TRs in output amp has lower DC voltage

DC voltages on transistors in the output amp are lower than those in the preamp? Are the main power rail voltages the same between each amp?

 

[oongbac]

DC voltages on transistors in the output amp are lower than those in the preamp? Are the main power rail voltages the same between each amp?

Yes. They're both same. V+ is 15V, V- is -7.9.

 

[dogears]

Hope this is helpful for diagnosis..both 2128 gain blocks are identical, save only the gain set resistor (SW2 selects and R43 for the makeup).

Q1/5 is a 2N5087, 1-2-3 E-B-C
Approx DC voltage
Pin 1 -5.5
Pin 2 -6.1
Pin 3 -9.3

Q2/6 is a BC550C, 1-2-3 C-B-E
Pin 1 +6.9
Pin 2 -9.3
Pin 3 -9.9

Q3/7 is a BC550C, 1-2-3 C-B-E
Pin 1 -4.3
Pin 2 -9.9
Pin 3 -10.6

I also notice few years later that C6 is drawn backward - the higher voltage is at the R49/R50 junction, but the spread is small and probably not enough to damage the cap. Check across that cap and if it is indeed higher voltage at R49/R50, flip it.

 

[oongbac]

Thank you very very much.

Voltage over C6 was about 5V and R49/50 junction was -4V so I flipped C6. Now voltage over C6 is 4.2V and R49/50 junction is -4V. But the problem is still there.

And voltage of Q1, 2, 3 is almost identical except pin1 of Q3. It's -6.9V.

And here's my measurement of Q5, 6, 7.

Q5
-2.9V / -3.5V / -3V

Q6
5.2V / -3V / -3.6V

Q7
-4.3V / -3.6V / -4.4V

Quite lower than the first amp.

Thank you guys for spent your precious time. I have to buy a book about electronics...

 

[dogears]

The two blocks should be identical, so if DC voltages are off I'd start checking resistor values vs the schematic. Maybe I have a BOM error?

 

[oongbac]

At least the resistors in the output amp on BOM are identical to schematic. I checked resistors according to the schematic couple of times, I should do more!

 

[oongbac]

Check resistor on both 2128 with color code and DMM, they're identical. Is there something can cause my problem? And I did some crazy experiment that flip Q5. And then clean sign wave came out, after a minute it change to distorted signal. And Q6 of my unit tend to get hot when I apply sign wave.

 

[Spacecowboy]

Check resistor on both 2128 with color code and DMM, they're identical. Is there something can cause my problem? And I did some crazy experiment that flip Q5. And then clean sign wave came out, after a minute it change to distorted signal. And Q6 of my unit tend to get hot when I apply sign wave.

Try removing R43, that should set it to unity gain

 

[dogears]

Q6 gets hot? Can you confirm R48 is 120R?

 

[oongbac]

Q6 gets hot? Can you confirm R48 is 120R?

Yes. R48 is 120R.

Edit: Oh, sorry it was Q7 not Q6!

 

[oongbac]

I tried removing R43. It's getting really quieter but it sounds normal(still really quiet, not distorted) when gain switch is on 20 position. Turn the switch to upper position, it starts distorting. I tested with mic(without pad) and -4dBu sign wave(with pad).

 

[dogears]

If you remove R43 the output stage is unity gain, or really close to it, and it will get quieter since it isnt making up the gain from the EQ loss.

Gain switch on 20 means you're adding 20 dB of gain. The unit starts to distort around +15 dBu. If you're sending a -4 dBu signal, you're outputting +16 dBu, and it will distort. Keep your output levels around 0 dBu average, which gives you headroom for peaks, and you should be fine. I'm not sure there's anything wrong based only on that.

If the signal is clean on a scope coming out of the preamp or HPF, the loss of the EQ is made up by the output amp with R43 at 8.2k, and it should be the same amplitude coming out of the amp at C52.

One thing to try, perhaps, is removing R35 and isolating the output amp as your input point to send a test signal (something like -4 dBu should be fine). Then check the output signal at C52 to see if it is distorted. With R43 removed you should have unity gain across that amplifier. If not, that confirms there is a problem is that block as we suspect. If you install R43, use a lower input level like -20 dBu.

 

[pahstah]

Is the pad switch soldered correctly to the board?
Is input transformer pin 1 in the right position?
Also, looking at the pictures that you've posted, it sees to me that you have put the Pad switch where the HPF switch should be and vice versa.
Also are you sure R21 680R is correct, I think it should be 68R but maybe I'm wrong

 

[oongbac]

I think pad switch and transformer are Okay. About position of switch, I used front panel design from TwentyTrees so that don't cause any problem. And I flipped HPF switch now.

According to schematic and BOM of this project R21 is 680R. But I'm willing to try that after I test output amp.

Thanks!

 

[oongbac]

I removed R35 and applied sine wave at leg2 of EQ cut switch. And put my scope at C52 it's not distorted and voltage seems almost unity. I applied sine wave to P5 of input transformer and put scope at C5 it was okay only when the gain switch is on 20 position. When it's on 30 position, gain goes down and distorted no matter how loud or quiet is the input signal. And I turn it on 40 and go back to 30, it's normal for a second and back to the symptom I mentioned. Maybe there's a problem in somewhere else.

I'll keep doing debug. Thanks guys!

 

[Spacecowboy]

But what level signal?

 

[oongbac]

But what level signal?

I used signal generator and setting was 1kHz around 300mVAC.