Home Cinema active Subwoofer malfunctioning

[lemofus]

Recap it and get it over with especially C16

Done, fully recapped. R41 and Rbonus were replaced with 8k resistors, all I had underhand was 8 or 20k...

Measure V across R38 in the OFF position.

V = 0V

With no signal. Short C16 and measure V across R41

V = 0V, in both ON and OFF positions. interestingly, the LED did not turn green when switched to ON, with C16 shorted...

Check delay to LED change from OFF to ON.

Hard to estimate, I'd say 0.5-1 sec.

Move switch to AUTO.
Apply music. Check delay to LED change.

I'd say a good couple of seconds here, longer than OFF-ON change.

R43 and the internal leakage current of C16 allows the timing circuit to discharge when the mains are cut. At 5RC, the cap is about 99% discharged. I can fix that with a J174 FET and a resistor. The resistor across the cut must be used too.

I'm just confused as to why ON wouldn't show a green LED with C16 shorted?! Expected?

That ought to keep you busy for a while.

I'm off to Paris on sat, south of france on wednesday, spain on Friday.... Then I have a few days of work, and I'm off to Italy for a few days... Next few weeks will be a bit more on and off so I'm trying to speed things up, or at least stop somewhere reasonable! Can't complain about holidays though...

 

[KeepItSimpleStupid]

Recap it and get it over with especially C16
Done, fully recapped. R41 and Rbonus were replaced with 8k resistors, all I had underhand was 8 or 20k...

Perfect

Measure V across R38 in the OFF position.
V = 0V

Good. Hopefully you switched to the Millivolt scale

With no signal. Short C16 and measure V across R41
V = 0V, in both ON and OFF positions. interestingly, the LED did not turn green when switched to ON, with C16 shorted...

Can't. Only happens when C16 gets charged above D5's voltage

Check delay to LED change from OFF to ON.
Hard to estimate, I'd say 0.5-1 sec.

Better number - don't you think?

Move switch to AUTO.
Apply music. Check delay to LED change.
I'd say a good couple of seconds here, longer than OFF-ON change.

Lot's of reasons for a slight change. I thnk it's a better delay than stock?



R43 and the internal leakage current of C16 allows the timing circuit to discharge when the mains are cut. At 5RC, the cap is about 99% discharged. I can fix that with a J174 FET and a resistor. The resistor across the cut must be used too.
I'm just confused as to why ON wouldn't show a green LED with C16 shorted?! Expected?

Totally. If C16 can't charge, then delay will be infinate.

That ought to keep you busy for a while.
I'm off to Paris on sat, south of france on wednesday, spain on Friday.... Then I have a few days of work, and I'm off to Italy for a few days... Next few weeks will be a bit more on and off so I'm trying to speed things up, or at least stop somewhere reasonable! Can't complain about holidays though...

Luck you. Hiding from the big wedding?

Would you happen to have another amp lying around? If you did then I would input a signal on Low level in or Speaker in and see if there is hum present on Line out.

Check for the presence of any AC between SG of the connector and ground (No signal applied). If you have a frequency counter mode, check for the abcense of 2x the main frequency.

Let's do something a little bit different than before.

Remove Q8, Q9, Q10 and Q11
Put a 1K or so resistor between where E and C go on Q8 and Q9. Estimate the current as 80/((2 * R)+220) to size the resistors.

Tell me what you currently have installed for R13 and R14

Look at voltage between EC of Q7
Look at voltage between junction of R15/R11 and base of Q7
Look at voltage between junction of R16/R12 and the base of Q7

Consider installing the bias pot modification.

Remember as R14 gets bigger the higher voltage is applied between the B and E of Q7. The resistance should be highest in the CW position of the bias potentiometer.

 

[lemofus]

Would you happen to have another amp lying around? If you did then I would input a signal on Low level in or Speaker in and see if there is hum present on Line out.

Yep - but an amp specifically, and not simple an mp3 player or something else like that? Should be the same for testing?

Check for the presence of any AC between SG of the connector and ground (No signal applied)

0.003V AC

If you have a frequency counter mode, check for the abcense of 2x the main frequency.

DMM has a "Frequency measurements" mode, I assume that's what I'm after? If so, nothing is showing.

Tell me what you currently have installed for R13 and R14

R13 = 3.3k (unchanged)
R14 = 3.3k (1/8W, a bit larger, I reverted back to this one after frying things by letting the extra pcb touch the transformer...

Consider installing the bias pot modification.

Yep, would be a good idea, back with a 3.3k resistor (so 3.3 + .5, potentially). This time I'll glue gun the ends so that I don't short anything with the extra PCB.

Put a 1K or so resistor between where E and C go on Q8 and Q9. Estimate the current as 80/((2 * R)+220) to size the resistors.

Let me roam through my growing draw of electronic components, see if I have a 1k resistor underhand. R being Resistance? If so, current will be around 0.036A, so that means with a 1/4 W resistor I shouldn't exceed 7V. Should be ample, I recall Q7 being around 2.5V roughly.

EDIT: Just realised we are talking about Q8 and Q9 voltages, no idea what to expect there. Good chance they will be higher, in which case I'd have to go and get some resistors in town tomorrow, current ones would be too flimsy...

 

[lemofus]

I have two 1K resistors measuring slightly short of 1cm long, and 3mm in diameter. No idea what wattage that would be, probably 1/4W, or at best, 1/2W...

 

[KeepItSimpleStupid]

(0.036^2) * 1K, so a 2W resistor will be fine. I know the leads won't fit, but it's not permanent. 3mm sounds like 1/4W.

Based on the frequency meter and some of the tests we did WAY back, I don't expect hum the pre-amp section. It's just an additional check. You can also extract a signal right before it's applied to the sub's amp too on the connector. Use the MP3 player as the signal source; the output being the LINE out and the SG points separately.

Warning, the glue gun may not be a good fastening method. Been there. Done that.

 

[lemofus]

(0.036^2) * 1K, so a 2W resistor will be fine.

I'll go and pick up a couple of 2W 1K resistors this afternoon, 1/4W would have been a little flimsy if the voltages are anywhere close to those of Q10-Q11...

Warning, the glue gun may not be a good fastening method. Been there. Done that.

Thanks. Will take your word for that. I'll isolate using electrical tape, and once everything is setup permanently, I can glue the PCB add-on properly with an epoxy; will be cleaner.

 

[lemofus]

Put a 1K or so resistor between where E and C go on Q8 and Q9.

Done. Q10-11 still out of course.

Look at voltage between EC of Q7
Look at voltage between junction of R15/R11 and base of Q7
Look at voltage between junction of R16/R12 and the base of Q7

LED on RED (standby):
Q7 EC = 0V
R15/R11 and base of Q7 = 0.05V
R16/R12 and base of Q7 = 0.15 to 0.30V

LED on GREEN (ON):
Q7 EC = 0V
R15/R11 and base of Q7 = 0.05V
R16/R12 and base of Q7 = 0.15V

A few observations:
- 2W resistors in EC of Q8 and Q9 got very warm in a short period of time; made sure I didn't leave the thing on too long whilst doing measurements!
- Obvious discrepancy between R11/R15 and R16/12... Why?
- R15/R11 and R16/R12 measurements were done using a fixed range, the DVM went crazy in autorange... Not reassuring.
Both R13 and R14 are 3.3k resistors at the moment, although R14 should apparently be 3.6k.

 

[KeepItSimpleStupid]

Stupid me. That's what a migraine does. the Q8/Q9 E-C resistor does absolutely nothing except get hot. +V to 1K to 220 to 1K to -V. Oops!.

Go for broke. Leave out Q10 and Q11 and use the 25 W lamp.

Get the R14 combination going and set to 3.3K or less. Monitor voltage across Q7 E-C. Don't let it get above 2.4V

If successful. Reduce Q7 E-C to the minimum value you can get. Install Q10, Q11 and the heatsinks. Again, monitor Q7 E-C.

No light bulb. Do a brief sound test. There could be cross-over distortion.

FINAL BIAS ADJUSTMENT.

No signal. Monitor voltage across R18 or R19. Increase your bias POT until 10 mV or so appears. Continue monitoring for at least 1/2 hour. If it starts to creap up rapidly, it too high.

MUSIC test. No light bulb.

 

[lemofus]

the Q8/Q9 E-C resistor does absolutely nothing except get hot

I wonder why we bother with E-C then

+V to 1K to 220 to 1K to -V. Oops!.

More seriously, not sure I follow here...

Get the R14 combination going and set to 3.3K or less

OK, so something like 3k + 0.5k pot, and then we can fine tune.

Monitor voltage across Q7 E-C. Don't let it get above 2.4V

You bet; will be keeping an eye on that one for sure!

Reduce Q7 E-C to the minimum value you can get

Sounds good to me; this could be mean either increasing, or decreasing R14 resistance? Or will it necessarily go one way or the other?

I hope nothing else is burned somewhere... Got through a few too many Q10 and Q11 so far!

Will give those things a go tonight or tomorrow if I can, before going to Paris... Hope the migraine is fixed. I bet a leaky transistor is responsible for that too.

 

[KeepItSimpleStupid]

More seriously, not sure I follow here...

It turned out to be nothing but a voltage divider with 80 V across it. Basically a heater. It wasn't useful at all.

OK, so something like 3k + 0.5k pot, and then we can fine tune.

3.2K or 3.3K might be the ideal value, but go for 3K. You can always change it.

You bet; will be keeping an eye on that one for sure!

That's the initial key voltage. At 0 to maybe 1.5 V+ volts there is no way to overheat the outputs. It's always essential to monitor this voltage first. You should be able to use the light bulb to at least determine if you can vary it. But don't have a high voltage here without heatsinks on the outputs for very long.

Sounds good to me; this could be mean either increasing, or decreasing R14 resistance? Or will it necessarily go one way or the other?

Ideally, what you want is when the potentiometer is turned clockwise, the voltage goes up which means the bias goes up, which, I think, means R4 goes up.

Got through a few too many Q10 and Q11 so far!

Tuition is expensive!

I've had about a week of daily migraines. In the southern US there was 300+ killed because of tornadoes. Even that's way to far from me, my area was still bothered by the weather system. The day before it hit in the south, I had my worst one.

Have a fun trip.

 

[KeepItSimpleStupid]

If you think this repair is fun, what about this one:

Dog Story

A Kansas farm wife called the local phone company to report her telephone failed to ring when her friends called - and that on the few occasions, when it did ring, her dog always moaned right before the phone rang.

The telephone repairman proceeded to the scene, curious to see this psychic dog or senile lady. He climbed a telephone pole, hooked in his test set, and dialed the subscriber's house.

The phone didn't ring right away, but then the dog moaned and the telephone began to ring.

Climbing down from the pole, the telephone repairman found:

1 . The dog was tied to the telephone system's ground wire with a steel chain and collar.

2. The wire connection to the ground rod was loose.

3. The dog was receiving 90 volts of signaling current when the number was called.

4.. After a couple of jolts, the dog would start moaning and then urinate.

5. The wet ground would complete the circuit, thus causing the phone to ring.

Which demonstrates that some problems CAN be fixed by
pissing and moaning.


Thought you'd like to know.

 

[lemofus]

Ah aha ! Love it !

 

[KeepItSimpleStupid]

Next time you go shopping.

Shopper: I need a 3 Volt 2 Watt light bulb?
Salesman: For what?
Shopper: No, two.

 

[lemofus]

Hello there! Back for a few days... Before heading to Milan this time. Good traveling !

So, where were we?

Go for broke. Leave out Q10 and Q11 and use the 25 W lamp.

Get the R14 combination going and set to 3.3K or less. Monitor voltage across Q7 E-C. Don't let it get above 2.4V

That means sticking Q8 and Q9 back in? I hope I won't burn them....

 

[KeepItSimpleStupid]

Welcome back. Yep. go for broke. The moment you see decent adjustability, back it off. Put the heatsinks back. Do a brief sound test and final adjustments.

You could. but be VERY careful make some small heatsinks out of a cooking aluminum throughoway container or multiple folds of aluminum foil. Clip it into place with a clothespin. Watch for shorts.

 

[lemofus]

Welcome back. Yep. go for broke. The moment you see decent adjustability, back it off. Put the heatsinks back. Do a brief sound test and final adjustments.

You could. but be VERY careful make some small heatsinks out of a cooking aluminum throughoway container or multiple folds of aluminum foil. Clip it into place with a clothespin. Watch for shorts.

With Q10 and Q11 still out, Q7 EC voltage is 0, on or off.... expected? No apparent heat... I need Q10 and Q11 back in? R14 is set to 3.28k at the moment.

 

[KeepItSimpleStupid]

I don't think Q10 and Q11 have to be back in. Remember that the "proper value" is 3.6K and a higher value i=of R should translate into a greater voltage across Q7.

You need to adjust it while monitoring E-C. My guess is it will "suddenly" start to work while turning it.

If you can get to 2V and still have some travel left your probably OK. I'd turn off the amp and measure the total R. Set ot back to 3.3K and re-assemble with a heat sink.

For the final adjustment you need to be concerned about Q7 (E-C) and the voltage across the low value emitter resistor which will probably be near 10 mV.

If that low voltage starts to creep up rapidly, then the bias is too high. With no signal, monitor it continuously for the first 5 minutes and then check every 5 minutes for an hour.

 

[lemofus]

You need to adjust it while monitoring E-C. My guess is it will "suddenly" start to work while turning it.

Unfortunately, nothing yet. This is the situation now : R14 replaced with 3.3k + 0.5k pot; Q10 and Q11 are out, everything else is back in (including additional resistor to add delay).

Power on is just fine, bulb acts normal, LED gows both RED and GREEN depending on switch position. But regardless of R14 value, Q7 E-C remains at 0V (I varied between 3.3 and 3.8k). At least nothing is burning, I was worried about killing another Q8 or Q9, but no cause for worry if voltage remains null !

I'd love to think I'm close to fixing this

 

[KeepItSimpleStupid]

Voltages across R4 and R9?
and
Voltages across R11 and R12?

 

[lemofus]

Voltages across R4 and R9?
and
Voltages across R11 and R12?

R4 = 2.2V
R9 = 1mV...

R11 = 0
R12 = 0