Mackie CR4 amplifier no sound

[UserXP]

An update regarding that R22 resistor.

On the working unit, this R22 instantly reads at a steady 10.02kOhm value no matter which way I connect the probes to its contacts, it simply remains consistent.

On the faulty unit, when I place the black probe on the contact connected to the the ground and also to the emitter of Q1, and the red probe on the other contact, I get a steday ~12kOhm reading. If I switch the probes I get around 8.4kOhm slowly increasing to around 9kOhm and then the increment becomes very slow but it's still there. All this is far different than on the working unit. And remember I desoldered and tested that R22 and it was within its specs ~10kOhm.

Does this indicate anything? Can a faulty transistor induce these results? I still haven't desoldered/replaced the transistor as I'd like to be sure I dont't have to deal with the unsoldering of that SMD unless we can conclude it is a definite cause.
I also tested Q1 and Q2 of the workng unit in the diode mode for NPN, both show equal results between themselves, and I get those same results between the Q1 and Q2 of the faulty unit, kind of makes me postpone the transistor replacement as all four of them show almost identical readings in diode test mode.

[UserXP]

If you remember that image I posted a while ago sent by the Mackie guy, he marked blue and red voltage values. Blue is normal operation (sound on), red voltages are when the amplifier enters the mute/standby mode. I don't understand what switches those voltages, maybe you can.

I will test again, the difficult part is I have to dismantle the working unit every time I need to re-measure something.

[UserXP]

He said that trace A comes from the headphone jack. In normal mode it is around 12V. When phones are plugged in, it drops to 0.250V and TDA switches to +20V and mutes.
Trace B is for the mute pin of TDA7265, again blue voltage is for the play mode, red is for the mute mode.
Trace C is what he called a "soft start". I suppose what he meant is that it mutes the amplifier on turning the speakers on as there really is a ~2 seconds silent delay from the moment you turn the power on until the sound starts (on the working unit, that is). He mentioned that "cutting the copper trace at line C will disable the soft start and enable the amplifier". Sounds to me like he suggested a "cut pin number 5" kind of a workaround, which is a no-no.

What are your thoughts on all this?

[petehall347]

q2 should be pulling the voltage down at the mute pin .
check voltages again at q2 .

[momaka]

Originally Posted by UserXP

Blue is normal operation (sound on), red voltages are when the amplifier enters the mute/standby mode. I don't understand what switches those voltages, maybe you can.

Originally Posted by UserXP

He said that trace A comes from the headphone jack. In normal mode it is around 12V. When phones are plugged in, it drops to 0.250V and TDA switches to +20V and mutes.
Trace B is for the mute pin of TDA7265, again blue voltage is for the play mode, red is for the mute mode.
Trace C is what he called a "soft start". I suppose what he meant is that it mutes the amplifier on turning the speakers on as there really is a ~2 seconds silent delay from the moment you turn the power on until the sound starts (on the working unit, that is).

Going by that description, I think the way it works is this:

Upon power up, capacitor C74 (previously discharged) acts momentarily like a "short-circuit" while it is charging up. This pulls the voltage high on the Base of Q1. As such, Q1 is -conducting-. With this, regardless if headphones are plugged in or not, the collector of Q1 will be low (i.e. near 0V), which means the Base of Q2 will also be at near 0V. Because of that, Q2 will NOT be conducting and so VCC is left unchecked to supply whatever voltage (probably close to 19V as the schem suggest) to the Mute pin on the amp IC (pin 5)... which means the amp IC should be in standby.

Once C74 charges up, it essentially acts as if it's "disappeared" from the circuit, so now the voltage on the Base of Q1 is determined by the resistive divider formed by R86 and R77, which will come out to around -2.2V. This negative voltage makes Q1 to STOP conducting (and in fact makes it look as if Q1 is removed from the circuit - it will be high-impedance.) With that, now the voltage on the Base of Q2 will depend on if headphones are plugged in or not. If headphones are plugged in, Q2 base will be near 0V, so Q2 will not be conducting and again we are left with VCC to dump whatever voltage it is on the Mute pin of the amp IC.

Now once headphones are unplugged and C74 startup has been charged up (and thus Q1 still not conducting), then the voltage on the Base of Q2 will be pulled "high" through resistor R88 by that +12.99V line A. When the Base of Q2 is pulled high, Q2 will start conducting current. And with that, now you have a voltage divider formed (more or less) with resistors R91 and R92, dividing the VCC voltage according to their resistance... and hopefully in the range that is needed for the amp IC to come out of Mute/STBY and turn On. C39 is there to provide a "smooth" transition from STBY/Mute to On state.

So with this information, first measure your voltage on line A on the non-working unit when you have the headphones plugged in and when they are not plugged in. Do this test by actually plugging in headphones in the headphones jack, not just shorting the switches. Do you get close to that +13V mentioned by the Mackie guy when headphones are unplugged? If yes, continue below...

Now check the voltage on the Base of Q1 after the unit has been turned On for a few seconds. To avoid confusion, measure the voltage on the negative (-) lead of cap C74 relative to ground. Do you get around -2.2 or -2.3V after a few seconds? If NO, I suspect either C74 or Q1 may be faulty. Since C74 is an electrolytic cap with only two leads, try removing it and replace it with another cap. Note: you don't need to match the capacitance value of C74 exactly to the original. If you use a smaller capacitance, the soft-start will be shorter, and if you use a bigger cap, soft-start will be longer. The reason I suspect C74 is because electrolytic caps can get electrically leaky over time. If C74 is leaky, that could be the reason you're not seeing -2.2 to 2.3V on the base of Q1... or it could also be Q1 is faulty. However, if you do replace C74 with a new cap and the voltage on the base of Q1 is still not -2.2 to -2.3V, try removing the cap altogether and then power On the unit. This will remove the soft-start feature, but you should now get -2.2V to -2.3V on the base of Q1. (Also note that removing C74 may cause the speaker to "pop" very loudly during startup, so if you want, you can disconnect the speaker for this test.) If you still don't get -2.2V to -2.3V on the Base of Q1, then very likely Q1 is faulty. Before removing it, though, have one last check on the voltages on each side of R86 and R77 relative to ground. One side of R86 should have 13V going to it and one side of R77 should have about -13V going to it. The middle point where they meet should be -2.2V to -2.3V... but if it's not, then Q1 is definitely bad.

Report back what results you get for all of the above items I asked about and then we can continue from there.

[UserXP]

Momaka, thank you for the detailed instructions.

Ok, here it goes.

While the headphones are PLUGGED:

Line A voltage reads around 0.100mV (fluctuating).
R88 reads the same voltage of ~0.100mV on both its ends.
Q1 base reads -2.1V.
C74 reads -2.1V on its negative lead.
R86 reads +12.0V on one end and -2.10V on the other.
R77 reads -11.9V on one end and -2.10V on the other.
Q2 Collector reads +21V.

While headphones are UNPLUGGED:

Line A voltage reads around +12V.
R88 reads +12V on one end and +2.9V on the other.
Q1 base reads -2.1V.
C74 reads -2.1V on its negative lead.
R86 reads +12.0 on one end and -2.10V on the other.
R77 reads -11.9V on one end and-2.10V on the other.
Q2 collector reads +14V.

It appears that Q1 is fine or not? I didn't get exactly the same values like you suggested, but I'd say most of them are close. I didn't remove C74 because it measured close to what you expected, so it looks like its OK. But please have a say about this.

This is the first test after I resoldered back that R22 and tested the unit inside the speaker with all the cables attached. I don't get it. The readings now show that the amplifier will play. And it does. But I can't say what's different now because I haven't replaced any component. I have a feeling it's just waiting to start to fade out again and problem not solved. And that R22 still doesn't measure instantly at 10kOhms, it's stil fluctuating and giving uneven readings on switched probes.

[UserXP]

I managed to "capture" the voltage on Pin5 when it started to fade. The threshold seems to be at around 15.25V. As it starts to fade, the voltage increases in increments of 0.01. Each additional 0.01V increment further fades the sound. As it gets close to ~15.45V the volume is very low and disapeears completely at around 15.5V.
I tried to measure the voltages on contacts you suggested and noticed that on certain contacts when I place the probe, the volume instantly returns and goes away again.
So, as I suspected, it worked for a minute or two and went silent again.
I measured the voltage on pin 5 when the amplifired first worked and it was around 14.1V. Then something obviously started to increase it. It really is like something is "leaking".

And I don't know what to do to reproduce the working again. The only common thing between those working moments I can think of is that the unit was left unpowered for a couple of days.

[UserXP]

Update: I switched off, waited a couple of minutes and turned it on again. Before that, I attached the probes on the ground and pin 5, so that when I turn on the amplifier I can watch the voltages immediately.
Upon power on, it displays +21V and after ~1.5 seconds it drops to +14.1V and then starts to slowly increase by 0.01, but the increment is not constant; at times it is steady, then it will increase more rapidly, then it will drop by a few 0.01 units. But it gradually increases overall. The amplifier then works until it reaches that threshold and then the volume starts to fluctuate and fade out. Does this new information shed some light on the possible cause? What should I check while it' still working?

[UserXP]

In circuit, C74 measures ~134uF. But so does the one in the working unit, it's not ~100uF.

[UserXP]

And, momaka, you are right about C74. While the unit was playing, I placed another empty 16V100uF capacitor on the contacts of the soldered one - the unit immediately went silent for about 1,5 seconds while the capacitor was charging and then resumed. This, however, did not influence the voltage on Pin 5, it didn't go back to 14.1V, it just left off where it was. Once it goes over 15.1x volts the volume starts to fade or fluctuate. I guess we have to find what is leaking and adding those 0.01 increments. Any idas on that?
Hopefully, I provided all the measurements you suggested.

[UserXP]

On the working unit without phones, these are the values and all are quite constant:

Line A: +12.12V
R88: +12.12V and +2.9V.
Q1 Base: -1.93V.
C74 (negative lead): -1.93V.
R86: +12.12v and -1.93V.
R77: -11.89V and -1.93V.
Q2 (collector): +13.60-82V and does not go over 14V, let alone 15V.

[budm]

If C9 has bad leakage resistance it will affect the Mute pin Voltage, you can remove just for testing to see if the sound will stay on.
BTW, when test capacitance, ESR, etc., you need to remove the cap from circuit since it has other component connected to the cap.

[UserXP]

Originally Posted by budm

If C9 has bad leakage resistance it will affect the Mute pin Voltage, you can remove just for testing to see if the sound will stay on.
BTW, when test capacitance, ESR, etc., you need to remove the cap from circuit since it has other component connected to the cap.

C9? The small SMD capacitor above the U1 chip?

[budm]

Sorry, C39 1uF Lytics Cap connected in parallel with R92.

[UserXP]

Originally Posted by momaka

Report back what results you get for all of the about items I asked about and then we can continue from there.

I did the measurements and some more/other observations above.
This looks like it's probably a power issue as the sound is superb when the unit is working, no hiss, no crackling or buzzing whatsoever.

[UserXP]

Originally Posted by budm

Sorry, C39 1uF Lytics Cap connected in parallel with R92.

Oh, you meant that one. OK, got it. I will check that as well. I am also waiting for the input on those measurements.

[budm]

Originally Posted by UserXP

Oh, you meant that one. OK, got it. I will check that as well. I am also waiting for the input on those measurements.

You can just lift one leg off the board and see if the unit will function.

[petehall347]

looking at the schematic it looks like there should be 10v B to E on Q2 .. seems a bit much to me .

[momaka]

Originally Posted by UserXP

In circuit, C74 measures ~134uF. But so does the one in the working unit, it's not ~100uF.

Never measure caps in circuit - it won't give you accurate results.
In the case above, the cap is over 20% of nominal capacitance, which means the cap could be bad. But it may not be. We simply can't know, because the cap is in circuit, and that extra capacitance could be caused by something else. So if you have to measure electrolytic caps in the future, just desolder them and then test. I know it's an extra step, but it gives accurate results. Otherwise, you just have to ASSUME that C74 is OK. And when you assume, that's when you might miss an issue somewhere.

With that said, it does look like C74 is OK, since you do get around -2.1 to -2.3V on the Base of Q1. Really any voltage below 0.4V should keep Q1 off / not conducting, so I think the soft-start circuit with Q1 is working normally... and moreover, you verified that when you added another 100 uF cap in parallel with C74, which immediately muted the speaker for around 1.5 seconds again.

Originally Posted by UserXP

While the headphones are PLUGGED:

Line A voltage reads around 0.100mV (fluctuating).
R88 reads the same voltage of ~0.100mV on both its ends.
Q1 base reads -2.1V.
C74 reads -2.1V on its negative lead.
R86 reads +12.0V on one end and -2.10V on the other.
R77 reads -11.9V on one end and -2.10V on the other.
Q2 Collector reads +21V.

While headphones are UNPLUGGED:

Line A voltage reads around +12V.
R88 reads +12V on one end and +2.9V on the other.
Q1 base reads -2.1V.
C74 reads -2.1V on its negative lead.
R86 reads +12.0 on one end and -2.10V on the other.
R77 reads -11.9V on one end and-2.10V on the other.
Q2 collector reads +14V.

It appears that Q1 is fine or not? I didn't get exactly the same values like you suggested, but I'd say most of them are close.

Well, I can't say with 100% confidence - I believe I also asked for the voltage at the Base of Q2 (which would also be the voltage at the Collector of Q1)... but I think we can give Q1 a pass for now, because it looks like the voltage at the Collector of Q2 switches between Mute and Play mode normally... so don't worry about measuring the Base of Q2 / Collector of Q1 right now.

Originally Posted by UserXP

Update: I switched off, waited a couple of minutes and turned it on again. Before that, I attached the probes on the ground and pin 5, so that when I turn on the amplifier I can watch the voltages immediately.
Upon power on, it displays +21V and after ~1.5 seconds it drops to +14.1V and then starts to slowly increase by 0.01, but the increment is not constant; at times it is steady, then it will increase more rapidly, then it will drop by a few 0.01 units. But it gradually increases overall. The amplifier then works until it reaches that threshold and then the volume starts to fluctuate and fade out. Does this new information shed some light on the possible cause? What should I check while it' still working?

Yes, this info does help, actually.

Since the voltage on the Mute pin (and also Collector of Q2) is slowly increasing, it looks like there is a leakage current / drift from somewhere. It may be Q2, but it may also be C39 as budm suggested.

Overall, I concur with budm - remove C39 and see if the voltage creeps up again and whether or not it goes over the threshold. While at it, also check capacitance and ESR on C39 (out of circuit) and report back what you get.

Originally Posted by UserXP

I did the measurements and some more/other observations above.
This looks like it's probably a power issue as the sound is superb when the unit is working, no hiss, no crackling or buzzing whatsoever.

No, I don't think it's a power issue, unless you see both VCC voltage and the +/-12V rails on the faulty unit to be somewhat far off from the working unit... in which case, the small voltage differences could be what's causing this issue, and that means Mackie designer(s) didn't factor in circuit voltage drift over time (and due to mfg. tolerances) when they designed the STB/Mute/Play circuit. So if that's the case, we may be able to correct the circuit by modifying R92 or R91 resistance values. But try the test with C39 first as budm suggested (i.e. remove from circuit), and report back if voltage is still drifting up. Then we can go from there.

Originally Posted by petehall347

looking at the schematic it looks like there should be 10v B to E on Q2 .. seems a bit much to me .

I don't think that's possible. Z1 should clamp the voltage to ~3.9V max.
On the other hand, maybe Z1 could also be leaky, causing Q2 Base voltage to slowly decrease over time and make Q2 conduct less, thus Q2 Collector voltage increasing. IDK, just throwing that out there as a possibility too, though I think it's less likely. I think the main suspects here are still C39 and Q2.

[petehall347]

whoops i meant C to E .
anyway if Q2 base was monitored and it stays the same then it should be staying out of mute .
i would have added wires to Q2 C and E and shorted them when it went back into mute . then i saw that 10v difference . it should be hard on .. maybe i not looking at it right and the resistors are throwing me off .