Stuck on troubleshooting sub-amp plate

[kasumyku]

Hello guys,


I'm stuck on repairing a Boston Acoustics CS Sub10 subwoofer , pdf schematic attached.

Symptom: Weak sound output even at high volume.

A few years back I blew up the amp plate after cranking it up at a party.
When powering it up the only thing I could hear was just a faint output. I took it to a repair shop and they fixed, but also set me back about $100.
 
Last week happened again but this time at normal volume. Now I decided to try fix it myself...

What I tried so far:

-looked for obvious and found that when the technician fixed it, the only thing appeared to be resoldered/repaced  was U203. So, I replaced U203(quad op-amp on pre-amp board, page 9 on pdf file) but with no improvement
-pulled Q5 (D718) and Q6 (B668)  on power amp board (pg. 3)- both tested OK
-pulled Q21 (D669A) and Q22 (B649A)  power transistors  on power supply board (pg. 6) and found that Q21 had a voltage drop of 1.2-1.5v across Collecto-Emmiter on both ways. 

After replacing Q21 I've gained back volume but is still weak.
It looks like the amp outputs only 1/4 of power (comparing when it worked fine)

Any support would be much appreciated!

Thank you!

[Armadillo]

You should start measuring voltages like the power supply voltages instead of just changing components.

Separate the pre-amplifying with the main amplifying to troubleshoot, that way you cut the circuit into half. Make life easier.

Last page connector P03A, isolate pin 1 e.g. take out the capacitor one of the leg of C236. This is a suggestion to isolate the two.

There are a lot of electrolytic coupling capacitors, any one of them can be the likely culprits.

What measuring equipment do you have? If you don't have a oscilloscope, do you have a signal tracer?

If both don't have, do you have PC or laptop external speaker amplifier. You can use it as a tracer.

Hope it provides some insights to you.

 

Edit: wild shot - temporary take out [desolder] Q10 [page 3] from the board, if no improvement, solder it back in in.

[kasumyku]

Thanks a lot for replying Armadillo.

I do need some guidance on how to trace the signal as I am still a newbie...

I got my hands on a scope not too long ago so I’m still learning how to take measurements.
After separating pre-amp from the power amp do I start by applying at the LOW LEVEL(left and right) input a signal (like a 1Khz), then scope ground clip to chassis ground and use the probe before and after each op amp?Also, can I use my laptop or my iPod as a signal generator?

What should I look for?

1.Does the amplitude of the wave should increase after each op amp?
2.What should the signal look like at the input and output of limiter MD1 with VR201(volume knob) at minimum and then at max?
3.Do I look for distortion?

Another question that I have might be a rhetorical one

Can I also use LFE input for my 1Khz signal or not because it’s a low freq input?

[Armadillo]

It is good that you have a oscilloscope. You should repair yourself instead of paying us$100 the last time.

Use the low level signal because the LFE bypassed a few of the modules, so it is not suitable to find the faults or do the repair.

Apply low level signal from the usual source as you normally would [phono, cassette deck, radio etc], then put the oscilloscope at the Pin 1 of connector P03A. You should see larger signal than the input moving around. If you cannot see it, then the preamplifier is faulty. If you can see it, then the amplifier side is faulty which comprises, protection circuit, amplifier, mute circuit etc...

Then you divide which ever faulty side by 2 again and continue to probe. Let me know which side and we can continue;

But firstly as always, make sure power supply voltages are correct, speaker is not broken etc...

Edit: Turn preamp volume control to mid level, use ac coupling, probe to 10:1, set say somewhere 200mv/Div, 1ms/div [adjust to suit].

[Armadillo]

By the way, have you tried to remove the transistor Q10 to eliminate mute problem?

Don't need to worry about distortion because you are troubleshooting loss of amplification problem now.

The voltage at Pin 1 should be around ~ 1vac.

[kasumyku]

Hi Armadillo,

I've tried to remove Q10 but no luck because its a small SMT transistor. BTW, most of the components are SMT except a hand full of caps and transistors...

Power supply looks like like its in good shape . I've got the following readings from chassis ground to:

+VCC       +37.22Vdc
-VCC        -37.22Vdc
VB           +27.25Vdc
+15V       +15.00Vdc
-15V        -14.91Vdc

On the other hand the preamp board doesn't look so good. I removed one leg of C236 and used the scope ground clip on chassis ground and the probe on the leg of C236.
As you can see in the pictures, my input is a 50Hz with a Vpp of 824mV and the output is 454Khz with Vpp of 432mV, please see pictures.

Any idea how to go from here?

Also, what +VCC, -VCC and VB stand for?

Thank you!

[Armadillo]

Hi, nice work there;

The power supply voltages look good indeed.

The VCC, VB, just take it as legends to identify the types of power supply voltages used for this particular amplifier, VCC, plus and minus supplied for the power amplifier, VB is used for protection and the regulated plus minus 15 used for those op-amp.

Firstly, if you supplied 50 Hertz signal input, you cannot be getting 454khz output. So please tell me how did you inject in the signal and to what input port did you used.

From your oscilloscope graphs, I can see that you are using DC coupling. Please switch oscilloscope to AC coupling input. Also please tell me did you calibrate the probe to 10:1 attenuation [there is a sliding switch on the probe and you also need to set the oscilloscope accordingly otherwise voltage will be read wrongly].

454Khz is above human hearing ability, it may suggest that the circuit is self-oscillating but which I doubt so. Probe at location shown attached.

[kasumyku]

Thank you for the encouragement and support,

For injecting the signal into the low level input I used my cell's headphones output.

I'll be probing at the shown locations and post the updates with the readings.
Probe and scope attenuation were set to 10X ; on the other hand coupling was indeed set to DC, great for pointing that out 

[Armadillo]

on the other hand coupling was indeed set to DC,:-+

Something I need to point out is that, even you used DC oscilloscope coupling, because the pre-amplifier sections are mostly AC coupled by the coupling capacitors and the ops-amp is using balanced power supply, so we still should be seeing the waves somewhere in the centre of the ground / centre of scope. But it sat above the ground potential.

So it suggested 3 things if the measurement is correct. One, the IC is faulty, two failure of the -15v supply, three, coupling capacitor faulty.
But we know that -15v is ok. So please measure the ICs NJM2060 Pin 11 with respect to ground to eliminate this concern. it should read -15v. Check all the NJM2060.

When all your oscilloscope measurement results are available, we can diagnose further.

There are 2 anomalies that forbid further diagnosis, first the 454khz out and second, the above as described.

[Armadillo]

Also measure the voltage here;

[kasumyku]

Regarding the 454Khz....
I realized that  I pulled out the wrong capacitor leg, I desoldered on the side of the Limiter MD1 pin 3 and measured at pin 1 of P03A . That's why I was getting high freq

After probing again it measured 50Hz 4.16Vpp

I've attached a pic with rest of the readings.

Also, all op-amps at pin 11 with respect to ground are reading around -15Vdc.

Thanks!

[kasumyku]

I forgot to ask one thing...

Why there's such a low reading at pin 1 of Limiter MD1 ?

[Armadillo]

I forgot to ask one thing...

Why there's such a low reading at pin 1 of Limiter MD1 ?

You should adjust VR201 left and measure, then right and measure to see the difference. The limiter will limit this signal regardless to max 200mvpp at In of limiter [if vccd is 15v and -vccd is -15V. [Edit: if so, output should not be over 4.166vpp you measured??, additional measurement points attached]

You need to at least show me the scope trace at Pin 1 again [and the 380mv location]. This time use DC coupling, but first switch the oscilloscope to Ground and adjust to centre the trace then switch to DC coupling to measure the signal. I need to know if the trace is Centred or Loped to the side. [as those last measurements].
Also you didn't measure the voltage requested. Can you measure it. We need to ensure that the mute is not "secretly" activated.
I think the measurement looks Ok and we should we half completed [upon confirmation of the scope trace and the voltage].
We can proceed to the amplifier side. Solder the leg of the capacitor back to the circuit [and adjust the vol control back to mid level].
I will draw the amplifier side check location.
I think you are mastering the amplifier circuit by now.   

Also measure the voltage here;

Edit: Additional Measurement Points attached to confirm output at 4.166vpp and check voltage of vccd plus and minus.
Edit: Max Out = ~2.4 Vpp

[kasumyku]

Hi armadillo,

I'm kind of lost... 

As mentioned , I'm still learning how to use all the scope functions and what they mean...

All evening I was on the net looking for info so I can make sense of your last directions.
I understand that AC coupling means only AC signals will pass thru due to DC blocking capacitor but for DC and GND coupling I'm still scratching my head...

I'll be watching some more videos regarding the AC/DC/GND Coupling so I can better grasp the idea and hopefully I'll get back to you with the measurements tomorrow after work.
 
Thanks for your patience.

[Armadillo]

No worry. No rush;

Suppose you want to measure a unknown signal. The actual signal is AC but with DC bias. How would it be represented in the scope.



You can see that the AC signal is loped above the ground potential. So in another case if the bias voltage is very large, you wouldn't be able to see anything on the screen. That is the problem with large DC bias and small AC signal. That is also why I ask you to use AC coupling initially for troubleshooting because of unknowns.
And so if I ask you what is the bias voltage in the AC signal and where is the Ground reference of the scope?



So you would, to avoid stray voltage interfering, you would put the scope to Ground Coupling. Actually internally the input is shorted to Ground by the Scope so that the Ground Reference is shown on the Scope. You can also touch the probe to ground if you want do, but let the scope do it internally is better.
The trace show the Ground potential and you can therefore calculate up from the grid to see what voltage the bias is and know where Ground potential is represented by the scope in the first trace above. You can adjust the ground reference by adjusting the vertical knob lower if you want but essentially it let you know where the ground reference is.

Because your amplifier signal is complementary both positive and negative on ~ZERO ground bias, so the signal will be centred on the ground reference. In fact your amplifier do have a DC voltage protection to protect the speaker, that means if the bias is above ~zero, the relay will not engage the speaker.

I did browse the rigol user manual and found short cut crappy explanation on this subject actually.

[kasumyku]

Much appreaciated for your explanation and directions.

Hope I got this right...

I know you have mentioned that at In of the limiter should be max 200mVpp, but with the volume knob VR201 turned to the left is 0Vpp and to the right is 1.62Vpp.

Attached are the rest of the measurements with the volume knob turned 1/2 way.

p.s.: I don't know if it matters or not but no speaker was connected during the measurements.

[Armadillo]

I know you have mentioned that at In of the limiter should be max 200mVpp, but with the volume knob VR201 turned to the left is 0Vpp and to the right is 1.62Vpp.

Attached are the rest of the measurements with the volume knob turned 1/2 way.

p.s.: I don't know if it matters or not but no speaker was connected during the measurements.

The preamplifier including the limited appears to be functioning. Because the legend of the supply "VCCD" is not shown in the schematic, I have made the assumption that it was +/- 15V but instead you measured +/-36.0V. So the limiter should be functioning there at that voltage.

So we should move on to the power amplifier section.

But you need to hook onto the 100W speaker as load for the measurement to be done. Use jumper cables to the speaker if not long enough.

Please measure the amplifier section voltages again.  thanks.   

[kasumyku]

Hi Armadillo,

I have attached bellow the new measurements with load connected.
And if its fine with you I will have a few questions after your input.

Thanks!!!

[Armadillo]

Hi Armadillo,

I have attached bellow the new measurements with load connected.
And if its fine with you I will have a few questions after your input.

Thanks!!!

Long time ago in the beggining, I did say to remove Q10.... LOL. [how did I have suspected it?].
Because the collector cannot be 0v and the base cannot be -ve voltage [measure again]. It appears that this transistor could be shorted. You have to remove it to check.

The opamp output [voltage amplification] is too too LOW. It could be Q10 causing it.
Replace these capacitors C01, C05, C06. [not expensive] or ESR it. Replace it is preferred.
With Q10 taken out do new measurement as shown in attached.

We need to confirm these before proceed any further.
After Q10 is taken out, If voltage at output of opamp still so low, confirm Pin 8 voltage with respect to Pin 4 then change JRC4558D.

Edit: BTW what were your question?
Edit: If the amplifier power is restored, no need any measurement.

[pamarakos]

Hi to both of you and the rest
I've just registered for this thread
The only thing that I don't understand is how I got exactly the same failure at exactly the same date~!?~?~?~?!?!?
I've given my CS10 with exactly the same behavior to a friend and he told me that he tested everything and it's the amp board, so I've googled d718 subwoofer and voila...here I am...so shall I get to a repair shop or shall I wait for your results to see what will come up? I don't have an oscilloscope but I do have some technical-electronic knowledge....
Thanks a lot
Regards to all

[Armadillo]

Hi to both of you and the rest
I've just registered for this thread
The only thing that I don't understand is how I got exactly the same failure at exactly the same date~!?~?~?~?!?!?
I've given my CS10 with exactly the same behavior to a friend and he told me that he tested everything and it's the amp board, so I've googled d718 subwoofer and voila...here I am...so shall I get to a repair shop or shall I wait for your results to see what will come up? I don't have an oscilloscope but I do have some technical-electronic knowledge....
Thanks a lot
Regards to all

Welcome!. You found the site.

These days, modern electronics are built with self-destruct timers [MLCC timers] inside and yeah, it's no surprise they would likely fail just about the same time, though I am not certain if the CS10 fits the same category.
But since your friend is repairing it for you, why the need to bring it to a shop?

Our friend kasumyku did had a problem in removing the Q10 SMT transistor before, this type of device is quite small, so he may take some time to acquaint with SMT soldering and de-soldering. Yeah, first time always take just a little bid longer.
Without proper tools, it may take a bit more time.
But its a journey that everyone [electronic] will take, including me, once before.
So it will be awhile.   

[pamarakos]

My friend just tested everything else, he tried to find some parts for the amp but he didn't manage it after all.
So today I went and found a store that repairs amps and in general circuits, so maybe I'll have a luck. I wouldn't be able to de-solder such SMD with my equipment. I'll guess I'll have some news following days so I'll let you know. I hope till then we'll have some updates from kasumyku  as well.

Thanks for responding

C'ya

[kasumyku]

Hi Pamarakos,

Whit the great help of Armadillo it looks like Q10 is shorted and few other caps are faulty.
Yesterday I finally managed to remove it and I’m hoping tonite to post the new reading with Q10 out of the circuit.

[pamarakos]

Q10 is this KTD1304?

https://www.google.gr/search?client=firefox-b&dcr=0&biw=1920&bih=955&tbm=isch&sa=1&ei=4HAAWsOPPMiusAfsg77gDA&q=ktd1304+&oq=ktd1304+&gs_l=psy-ab.3...989.989.0.1125.0.0.0.0.0.0.0.0..0.0....0...1.1.64.psy-ab..0.0.0....0._kYpmf6NK9Q

[kasumyku]

Hi Armadillo,

After removing Q10 I'm getting all kind of weird readings, please see attachments.
All its left now is to order C01, C05, C06 and U01.
I'll post new measurement readings after swapping the parts.

Thanks!