DC detection circuit failure on Subwoofer

[Durenth]

Hello,

First of all sorry for my English and for my poor knowledge in transistor behavior, I'll do my best to be clear…

I have a problem with the DC detection circuit of a Panasonic 40W Subwoofer, model SW-P270.
The relay in the DC detection circuit never turns on thought there is no DC in the audio output signal.
Output signal is ok, controlled with an oscilloscope.

I found the service manual (full board schematic attached to this post) and measured the voltage as shown below, pink is the expected value indicated in the manual, red is my reading:



First of all, I suspected a failure on Q13 (NPN Transistor), so I removed it and got those new readings:



Getting closer but still doesn't trigger the 24V relay and created a problem with the R19 resistor which began to overheat.

I have multiple questions to ask you:

1. About the overheating resistor:

Voltage at B+ being +35V, if Q10 is on then, if I'm not mistaken, there should be a +35V voltage drop at R19.
R19 being a 330 Ohm / 2W resistor it cannot dissipate U²/R = 35²/330 =3.7W so I must be missing something…

Could you tell me what I got wrong? There is no voltage drop in the relay right?

2. A curiosity question:

In the schematics I see that Q11 is supposed to be on (Vce = 0V and Ve=0.8V) with only Vbe = 0.5V when I understand it should be at least 0.6V to turn on a transistor… Maybe there's a tolerance but then why creating a circuit so close to the limits.

Again, I must get something wrong, do you have any explanation for me?

3. About my main problem:

The schematics says I should get Vbe = 0.7V on Q10, in my circuit I have reading of 0.6-0.1= 0.5V that, I believe is not enough to fully trigger the transitor and explains the 10.6 V voltage drop on Vce.
I also get lower voltage on many other rails as shown on the second schematics.
I thought of replacing Q11 and Q10 but I fear there are other problems in my circuit…
I double checked my measurements with to different multimeter one Fulke and one ChauvinArnoux.

Do you think changing Q11 and Q10 will do? Then I fear for the over heating R19 resistor…

Thanks for your time and for your help, it was a long post!

Bye Bye

[budm]

Re: DC detection circuit failure on Subwoofer

Just for testing, you should be able to short the E and C pin of Q11 to force on Q10.

"Voltage at B+ being +35V, if Q10 is on then, if I'm not mistaken, there should be a +35V voltage drop at R19.
R19 being a 330 Ohm / 2W resistor it cannot dissipate U²/R = 35²/330 =3.7W so I must be missing something…"

You will not have Vdrops of 35V on R19 because you also have 24V relay coil resistance + Vdrops on Q10.
What are Vdrops of R19 and the Vdrops of the relay right now?

[Durenth]

Re: DC detection circuit failure on Subwoofer

Hi Budm,

I shorted E and C pin of Q11 and Q10 beavior didn't change for what I can understand, I got the same readings as before Vce=10.6V on Q10 and the relay didn't trigger.

I know have :

R19:
24.5V voltage drop
Q10:
Vce=10.5V
Vbe=0.6V (0.7V when Q11 is shorted)
Vcb=9.9V
Relay:
0V voltage drop

Just for you to know, I did try the relay offboard, it's operative and starts triggering at 20V.
Do you think D4 could be the cause of the problem?

Thanks for your help!

[budm]

Re: DC detection circuit failure on Subwoofer

"Relay: 0V voltage drop"

You need to check the DC resistance of the relay coil and the dc resistance of snubber diode D4, suspecting shorted D4.

[redwire]

Re: DC detection circuit failure on Subwoofer

These circuits can be hairy because they're trying to do so much with a few transistors - DC detect, power on delay, AC fail, over temperature etc. I see maybe two problems.

With Q13 removed, Q10 is not fully on (low beta) at VCE 10.6V, OR it is overloaded. I second the notion of a shorted zener D4 or relay coil. OSA-SS-224DM3 24VDC relay should have a coil resistance of 1,100 ohms needing at least 18V to pull in. R19 should never smoke, it drops the 35V down to something the relay is comfy with.
Because of the large voltage drop across R21, you know there is base current flowing into Q10 so it should be on. Using the VBE measurements i.e. 0.5V or 0.7V does not tell you the full picture. Some transistors are naturally 0.55V to 0.75V it does vary, especially with temperature.

Q12, Q13 are special "muting" transistors KTC2874 which have a reverse-beta and can actually conduct a bit backwards. So a + or -ve DC offset can activate Q12.
Q13 is for fast off when AC is switched off, the -31.6V rail falls faster than the +12V rail.
Q13 also has high reverse E-B voltage spec 25V, which it does not seem to meet so it could be bad.

[petehall347]

Re: DC detection circuit failure on Subwoofer

also r19 resistance .

[redwire]

Re: DC detection circuit failure on Subwoofer

Odd that R19 350R is a 2W part but my math says it dissipates 0.2W for the 24mA and 8.5V it would drop.

[budm]

Re: DC detection circuit failure on Subwoofer

Since the B+ is 35VDC , relay is 24VDC coil, so the Vdrops on the R19 should be around 10 ~ 11VDC.

[Durenth]

Re: DC detection circuit failure on Subwoofer

Thanks to you all for your replies and your help.

So D4 is short dead as Budum and Redwire suspected!

I took D4 off the board and I now have 1,050 ohms on the relay, good
I don't want to damage anything without what I understand to be a flyback diode, so I'll try to find a replacement diode on some other board before turning on the device again.

Redwire, thanks for your detailed explanations. You guys do also believe Q13 is bad if I understood everything well.
Just got back from work so I'll look into your explanations in details tomorrow, as I told you in my first post, I'm really not familiar with transistors so I'll work on it!

Bye and Thanks again

[redwire]

Re: DC detection circuit failure on Subwoofer

Good to find a problem With the shorted zener, R19 and Q10 would be overloaded and may be damaged from the heat. I'd check R19 and then run it to see if Q10 is doing OK.
Q13 seems to be fishy. You should be seeing -10V at Q13 base but are seeing -5V. It's spec is min. -25V withstand VEB so it looks leaky. Could swap it with Q12.

[Durenth]

Re: DC detection circuit failure on Subwoofer

Hi!

First of all R19 is ok: 330 Ohms.
Unfortunately I can't manage to find a equivalent Zener diode for my dead MTZJ24 so I still can't power up the board and try Q10.

I believe the D4 is a Zener in order to maintain a stable Voltage drop on the coil and protect it from overvoltage.
If I'm correct with R19 being 330ohm, the voltage on the relay shouldn't exceed Rrelay*(Ub/(R19+Rrelay)) = 1100*(35/(330+1100)) = 27V.
The relay datasheet states that it can handle up to 30V, so I thought I could put a regular diode for testing and maybe replace it later with a correct 24V Zener Diode.

Do you think I'm right and I can try it that way without damaging anything?

Then for the choice of the regular diode I should put in my circuit...
D4 (MTZJ24) datasheet says :



I looked for 500mW diodes in my spare parts supplies and I found myself with a few options:

1N4001:



1N4148:



1N4448:



BAV20:



I was going for 1N4001 500mW with Vrrms = 35V, but not really sure it's the best choice.. What do you think?
I "usually" use donor boards or order the exact same part when fixing my stuff so as you can see I'm not really practical with choosing...

Bye Bye!

[redwire]

Re: DC detection circuit failure on Subwoofer

The zener circuit is a bit weird, I think to prevent mains sags from cutting out the relay. Nobody likes a sub cutting out when the bad guy explodes lol.

A diode is used to suppress a relay coil's back EMF, the -ve voltage spike due to coil inductance that will damage the driver transistor.
But in this circuit, the zener acts as an ordinary diode for coil back EMF, and in the reverse direction it regulates the relay coil's voltage to around 24VDC.
Actually a relay is not finicky about the voltage it gets, so a zener is not typically used there.

MTZJ24B is a 1/2W 24V zener, I would upsize to 24V 1W zener 1N4749A there, for long life. Your old zener is running over spec 0.6W and why it died.
Just a plain 1N4001 or better would work- assuming the relay coil gets around 22-26V. If you have high mains voltage, instead of 33V you have 35V so a bit extra the zener gets rid of. But the relay will just run a bit warmer and not be an issue.
A diode's peak and RMS voltage ratings - here it can see around 35V max. so a 1N4001 (50V) or better is fine. A 1N4148 is OK too because the coil current is around 25mA, the back EMF is also 25mA max.
I thought R19 could even be higher in value, even 400 ohms.

[budm]

Re: DC detection circuit failure on Subwoofer

Do you have 1N4007 to use sincee back EMF can be 100's of Volt from 24V coil? The snubber diode will try to clamp it down to Vf of the diode.

[redwire]

Re: DC detection circuit failure on Subwoofer

The design is a 24V zener regulator for the relay coil when it is on, and a diode 0.7V for clamping the back-EMF. Not the usual circuit.
Zeners are used for a back-EMF clamp when you want fast relay switching. The coil's magnetic field collapses very slow with a diode at 0.7V compared to a high voltage (zener) clamp.

[budm]

Re: DC detection circuit failure on Subwoofer

I do not think in this case the relay release time (speakers disconnected when error occur) is much of a factor.
The zener being used in this case will also clamp to 0.6V (compared to your figure 3 which is not the case), but when relay is energized, the zener will clamp the relay col Voltage to 24V, they may have done that because the 35V power supply is not regulated power supply so the if the incoming AC can go up and down.

[Durenth]

Re: DC detection circuit failure on Subwoofer

Hooray It works!!

I've replace the Zener with a regular 1N4001, just for testing (I also I ordered a 1N4749A diode for the long term fix).

I'm kind of busy lately so didn't have the time to go probing around the circuit, I can only say that voltage drops around the relay are now back to normal and that I can here the beautiful sound of the relay turning on when I power up the board! I also tried with audio input and the signal reaches the speaker connector.

So tomorrow I'll try Q13 again

Again thanks for your explanations and for your time, I've also read some articles related to your answers and began to understand much better how this circuit works.. Exciting subject!

[budm]

Re: DC detection circuit failure on Subwoofer

I doubt that there is any problem with Q13.

[Durenth]

Re: DC detection circuit failure on Subwoofer

Unfortunately when Q13 is back on the board, it doesn't work anymore..
Q13 seems to leak between Base and Emitter and drops the voltage to a negative -0.5V at the base of Q10, thus not allowing the relay to turn on.
Voltage measurements are almost the same as the red ones in my first post..

After a lot of research, I found what I believe to be a good replacement for Q13, an other muting transistor with almost the same characteristics except for Collector Power Dissipation that is 400mW instead of 625mW on the existing transistor.
For what I understand Q13 is almost never, or let say for a very brief amount of time, conducting (when fast-off), for that reason I thought it might not be a big deal if Pc is lower on my replacement part..

What do you think?

Q13 / C2874 :



My best replacement (actually, I found other but tehrer not available online) / 2SC2878:



Bye!

[petehall347]

Re: DC detection circuit failure on Subwoofer

you might be better with a matched pair for q12 q13 ... not 100 percent on that though .

[budm]

Re: DC detection circuit failure on Subwoofer

"Q13 seems to leak between Base and Emitter and drops the voltage to a negative -0.5V at the base of Q10, thus not allowing the relay to turn on."

Emitter of Q13 is connected to circuit ground so leakage between Base and Emitter will cause low negative reading.
So if Q13 is not in place what is the Voltage on the Base of Q11? To have negative on Q11 the leakage will be from Base - Collector junction of Q13.
I also cannot see how you can have negative Voltage on the Base of Q10 unless you have negative on the Base of Q11 and it also has bad leakage on the Emitter-Base of Q11 (Negative Voltage on the Base of Q11 = Reverse Bias of Q11 E-B junction which means there will be no current flow)) to let the negative Voltage get all the way to the Base of Q10.
I think you need to report the Voltage readings of all the Base Voltages (with ref to circuit ground) of the Transistors when Q13 is in place.