Home Cinema active Subwoofer malfunctioning

[KeepItSimpleStupid]

Now, I'm really confused. How can a 9V battery out out 10V?

I was expecting something like 56 mV across R11 and R12.

0V across Q7 is bad, but it won't cause damage.

How about trying this:

Put R15 and R16 back in. Remove battery. Leave the Q19 jumper in for now.
Remove Q8 and Q9. R20 can stay lifted. No signal. No speaker. Power up.

Most important numbers are:

Voltage across Zd1
Voltage across E-C of Q7
Voltage to ground on both sizes of R15 and R16
Voltage across R4 and R5

 

[lemofus]

Put R15 and R16 back in. Remove battery. Leave the Q19 jumper in for now.
Remove Q8 and Q9. R20 can stay lifted. No signal. No speaker. Power up.

Most important numbers are:

Voltage across Zd1
Voltage across E-C of Q7
Voltage to ground on both sizes of R15 and R16
Voltage across R4 and R5

"Voltage to ground" was measured with one node on the named side of the resistor, and the other linked to the PCB main ground (black cable coming from transformer - I assume this was what was meant here)

WITHOUT Q19 Jumper:
ZD1 = 5.5V (weheeey!)
Q7 E-C = 0V (can't seem to get any voltage from any ECB configuration?)
R5 = 0.07V roughly, then quickly worked its way down to 0?
R4 = 0.8V
R15 = 0.125V
R16 = 0.1V

R15 (R11 side) to ground = 37.7
R15 (Q8 side) to ground = 37.4
R16 (R12 side) to ground = 37.4
R16 (Q9 side) to ground = 37.4


WITH Q19 Jumper:
ZD1 = 0V (Sadly?)
Q7 E-C = 0
R5 = 0
R4 = 0
R15 = 0.025V (surprising - Shorting Q19 take .01V off, but there is a residual 0.025)
R16 = 0

R15 (R11 side) to ground = 37.3
R15 (Q8 side) to ground = 37.4
R16 (R12 side) to ground = 37
R16 (Q9 side) to ground = 37


On another note, I'm upgrading to a better multimeter in the week, should hopefully help in the measurements, the one I currently have is really quite basic.

 

[KeepItSimpleStupid]

Cool:

Let's try another couple of baby steps.

(1) Check R11, R13, R14, R12 in circuit using both resistance polarities?

(2)
Short C4 with a small wire.
Can you inject a DC signal (positive and negative) <= 1.5 V at the junction of C1,R1 to ground?
Look at these two voltages:
Across R4?
Across the R5 D2 combination?

For the DC signal you can even use your 9V gizmo. Short it with a 1K or so resistor and take the voltage across your 1n4148 diode for the injected signal of probably around 0.2 to 0.6 V.

 

[KeepItSimpleStupid]

I don't AT ALL understand the ZD1 reading of zero with the Q19 jumper because +V should flow from R15, Q19E to Q19C through D1 through R3, regulated by ZD1 and filtered by C3 and applied to the base of Q3 as a current 5.6/1500 amps and it returns through R16 to the - supply.

What's going on?

Is it phantom? leakage through Q1,Q2 and Q3?

 

[lemofus]

(1)
WITHOUT Q19 Jumper:

R11 : shot up to 0.4V as I switched the amp on, then went down to .080, and stabilised at .097. -0.098 the other polarity.
R13 = 0V (both polarities)
R14 = 0V (both polarities)
R12 = 0V (both polarities)

Need those readings with the jumper on Q19?

 

[lemofus]

I don't AT ALL understand the ZD1 reading of zero with the Q19 jumper because +V should flow from R15, Q19E to Q19C through D1 through R3, regulated by ZD1 and filtered by C3 and applied to the base of Q3 as a current 5.6/1500 amps and it returns through R16 to the - supply.

What's going on?

Is it phantom? leakage through Q1,Q2 and Q3?

Let me check those readings again...

 

[lemofus]

Shoot - My mistake, I had shorted E-B, that transistor is an E B C configuration apparently. Sorry about that, going to make those measurements again.

 

[lemofus]

Put R15 and R16 back in. Remove battery. Leave the Q19 jumper in for now.
Remove Q8 and Q9. R20 can stay lifted. No signal. No speaker. Power up.

WITH Q19 Jumper at E-C:
ZD1 = 5.5V
Q7 E-C = 0
R5 = 0.025
R4 = 0.77V
R15 = 0.125V
R16 = 0.1V

R15 (R11 side) to ground = 36.3
R15 (Q8 side) to ground = 36.6
R16 (R12 side) to ground = 37.0
R16 (Q9 side) to ground = 36.7

 

[lemofus]

(2)
Short C4 with a small wire.
Can you inject a DC signal (positive and negative) <= 1.5 V at the junction of C1,R1 to ground?
Look at these two voltages:
Across R4?
Across the R5 D2 combination?

For the DC signal you can even use your 9V gizmo. Short it with a 1K or so resistor and take the voltage across your 1n4148 diode for the injected signal of probably around 0.2 to 0.6 V.

- This is without Q19 shorted?
- 1.5 DC without resistor and diode in the circuit?
- AC power is OFF during measurements I assume?

 

[lemofus]

Short C4 with a small wire.
Can you inject a DC signal (positive and negative) <= 1.5 V at the junction of C1,R1 to ground?

AC power ON, no signal, R20 still lifted, Q8 and Q9 still out...

R5 - D2 combo = 2.25V
R5 - D2 combo (1.5 V batt connected, + --> GND) = 2.88V
R5 - D2 combo (1.5 V batt connected, - --> GND) = -0.13V

R4 = 0.6V
R4 = (1.5 V batt connected, + --> GND) = 0V
R4 = (1.5 V batt connected, - --> GND) = 0.8V

While doing the measurements, I also noticed that the row of middle transistors, Q1, Q2, Q19 - somewhere around there - was warm. (I stuck my fingers closer to try and identify which one and got a jolt from the sweaty fingers - and there's an extra small burn mark on my hands! Circuit seems fine though.)

 

[KeepItSimpleStupid]

With C4 shorted, some of the transistors will get warmer. The function of C4 is to "short" when an AC signal is applied. Since we are using a DC signal to test we have to short C4. Consequence of not having a signal generator or scope.

The R4 voltage is bad. Probably should be closer to 2.88 V.

Lift R11 at R15 and lift R12 at R16 and repeat

Kinda suspecting Q1 or Q5 at this point.

 

[lemofus]

Short C4 with a small wire.
Can you inject a DC signal (positive and negative) <= 1.5 V at the junction of C1,R1 to ground?

AC power ON, no signal, R20 still lifted, Q8 and Q9 still out, R11 lifted at R15 and R12 lifted at R16.

I'm feeding 1.5V directly, no diode or resistor in the circuit by the way.

R5 - D2 combo = 2.32V
R5 - D2 combo (1.5 V batt connected, + --> GND) = 2.87
R5 - D2 combo (1.5 V batt connected, - --> GND) = -0.28

R4 = -0.5V
R4 = (1.5 V batt connected, + --> GND) = 0V
R4 = (1.5 V batt connected, - --> GND) = -2.2V

 

[KeepItSimpleStupid]

I lost what I typed. I don't know why that happens sometimes.

Good. This means Q1, Q2, Q3 etc, are working. Since the voltages are different it means that the gains of Q1 and Q2 are not matched. You used to be able to buy matched pairs of transistors from NTE, such as e.g. NTE123MP which would be a matched pair of NTE123's. This means that the positive and negative excursions of the waveform will be amplified with different gains. I could match them for you or you could, but we'll save this issue for a later time.

So, I'm suspecting R11 (somewhat), Q5 (highly), C5(not likely), R13 (somewhat), Q7(maybe), R14 (maybe), Q6 (not likely), C6( not likely), R12 (maybe)

I'm suspecting Q5 is in backwards. So check that. Q5 is a PNP transistor.
Based on earlier tests like the voltage across those resistors would make Q5 in backwards a real possibility

Remove Q7

Measure in circuit:

R11 (already has a lead disconnected)
R13
R14
R12

In circuit using both polarities of the ohmmeter. i.e. switch the leads

 

[lemofus]

I'm suspecting Q5 is in backwards. So check that. Q5 is a PNP transistor.
Based on earlier tests like the voltage across those resistors would make Q5 in backwards a real possibility

Unfortunately, according to the PCB annotations, my diode tests (off-board), and the original PCB pictures I uploaded, it seems that Q5 is the right way in... If need be, I have a spare A935 (Q5, Q16) transistor underhand.

Remove Q7

Measure in circuit:

R11 (already has a lead disconnected)
R13
R14
R12

In circuit using both polarities of the ohmmeter. i.e. switch the leads

This is with R11 and R12 still lifted I assume (and R20, still)?

I lost what I typed. I don't know why that happens sometimes.

I have that at times. I find that I personally click on the wrong button sometimes, I click on "reply" instead of "Post Quick Reply", and find myself having to retype the whole thing - could be the same confusion?

 

[lemofus]

I seem to get 0V for all of those readings with Q7 out (and R11-12 still lifted).

 

[KeepItSimpleStupid]

Not voltage, Resistance. Polarity matters with resistance readings too. One way may turn on a transistor junction whereas the other way will not.

Because of the 0.6 V clamp with R11 connected; that's the reason I'm suspecting something backwards. Because of the 0V readings on R11, R12, R13 and R14 there has to be something wrong with that string.

 

[lemofus]

Not voltage, Resistance. Polarity matters with resistance readings too. One way may turn on a transistor junction whereas the other way will not.

Ha ha ok, makes more sense ! Let me try again...

 

[lemofus]

Remove Q7

Measure in circuit:

R11 (already has a lead disconnected)
R13
R14
R12

In circuit using both polarities of the ohmmeter. i.e. switch the leads

R11 = 46.2 Ohm
R12 = 46.3 Ohm
R13 = 3.25 kOhm
R14 = 4.27 kOhm

 

[lemofus]

Because of the 0.6 V clamp with R11 connected; that's the reason I'm suspecting something backwards. Because of the 0V readings on R11, R12, R13 and R14 there has to be something wrong with that string.

I'm confused. I'm going to re-visit all the steps I have taken so far, see what could have gone wrong here. Could I have swapped Q5 and Q6 around? Is this a possibility?

Q6 has the inscriptions "D667AC" on it, can't seem to find any info online. But I manually checked the orientation of Q5, this is strange...

 

[KeepItSimpleStupid]

Use "2SD667 datasheet". The 4.27K is bad then, if I remember it's suposed to be 3.x K.

Q5 for Q6 is possible too.