Speaker pop on shut off Sanyo JCX2900

[mbz]

Little confusing

The relay is closing too fast.

The relay closes sometime after power-on, the delay/mute period is influenced by size of C412 (47uf), increase C412 increases relay operate/close time.

There needs to be a delay before the relay opens at shutoff.

At shutoff the relay should ideally open immediately however in real life there is some delay, C412 needs to discharge for the relay to open.

Why is it not delayed and how can I get the relay to stay open longer?

I think what you are asking is "At power off, how do you increase the delay before the relay opens"
Increasing C412 will inrease the open time

I taped the relay closed and the pop does not happen at shutoff.

Suggests an arcing issue???

Keeping in mind there is no pop when pre/main links removed suggests preamp issue.

Do you still get a pop with Tone Defeat on/off?

 

[curtisw99]

Little confusing

The relay closes sometime after power-on, the delay/mute period is influenced by size of C412 (47uf), increase C412 increases relay operate/close time.


At shutoff the relay should ideally open immediately however in real life there is some delay, C412 needs to discharge for the relay to open.


I think what you are asking is "At power off, how do you increase the delay before the relay opens"
Increasing C412 will inrease the open time


Suggests an arcing issue???

Keeping in mind there is no pop when pre/main links removed suggests preamp issue.

Do you still get a pop with Tone Defeat on/off?

Sorry for the confused post. I went back reread and confused myself. I edited to make correct.

I want to increase the delay at shutoff so the relay stays closed longer.

I get a pop no matter what buttons are pressed

, I don’t think it’s arcing from a bad relay, it happens on the A and B relays. Pop is identical.

So I should increase the capacitance on C412? Then the relay will stay closed longer?

 

[mbz]

I want to increase the delay at shutoff so the relay stays closed longer.

So I should increase the capacitance on C412? Then the relay will stay closed longer?

Yeah increase C412 but doubtful it will give you the needed delay, Probably an easier way would be to lift R420 which is C412 discharge path, or you could do both.

Warning these actions will mean more dc voltage will appear at the speaker terminals so do not connect speakers instead monitor voltage with a MM

 

[curtisw99]

Yeah increase C412 but doubtful it will give you the needed delay, Probably an easier way would be to lift R420 which is C412 discharge path, or you could do both.

Warning these actions will mean more dc voltage will appear at the speaker terminals so do not connect speakers instead monitor voltage with a MM

Do you mean increase the resistance of r420?

 

[mbz]

Do you mean increase the resistance of r420?

Probably an easier way would be to lift R420

My turn for bad comms...

I meant lift one leg or remove R420.
R420 is the bleed path for C412, with R420 removed, C412 will hold its charge and voltage longer keeping the relay closed. This will cause nasty voltages, probably a few volts not 200mV! C412 will discharge more slowly through secondary paths.

edit Working on loading the preamp into microcap12 simulator, probably a waste of time

Another long shot would be to build an audio probe (probe, series cap connected to disposable speaker). Then probe the preamp audio path trying to identify the stage where the pop originates

 

[curtisw99]

My turn for bad comms...

I meant lift one leg or remove R420.
R420 is the bleed path for C412, with R420 removed, C412 will hold its charge and voltage longer keeping the relay closed. This will cause nasty voltages, probably a few volts not 200mV! C412 will discharge more slowly through secondary paths.

edit Working on loading the preamp into microcap12 simulator, probably a waste of time

Another long shot would be to build an audio probe (probe, series cap connected to disposable speaker). Then probe the preamp audio path trying to identify the stage where the pop originates

I’m still developing my schematic reading skills and learning circuit theory.
Am I reading this right?
It looks like Q431 is the switch that runs the relays and it’s hooked straight into R431 and R431 goes to the transformer.

Or is C412 the “C” in a delay RC network?
What keeps C412 from dumping all its charge to the turned off transformer?

 

[mbz]

Credit for trying to understand the schematic

That thick black line that connects -ve of C412 and R431 is ground, 0V... it connects to the centre tap of the transformer, and also the GND connection between the main caps, again 0V

It looks like Q431 is the switch that runs the relays

agree

Or is C412 the “C” in a delay RC network?

Yes C412 is for RC delay, R419(220k) appears to be the R

The way I see it,
At power on C412 charges up through R419, as the voltage increases on C412 +ve, zener D419 goes into breakdown mode and current flows through the zener, this current turns on Q431 which then switches on Q432.
Q432 provides a path to GND for the relay current to flow (from D903, through R401/402, S16, S17 speaker selector switches then through the relay and then Q432c-e to GND)

Working backwards, the expected voltages are
Q432e = 0V (GND connected)
Q432b = +0.6V (standard Vbe = 0.6V abouts)
Q431e = Q432b = +0.6V
Q431b = +1.2V (0.6 + Vbe)
D419 cathode = +1.2 + 4.7(zener) = +5.9V

To speed up the discharge of C412 you could try reducing R420(220) to 150 or even 100ohms but must be 0.5W or higher.

What keeps C412 from dumping all its charge to the turned off transformer?

Ideally capacitors pass AC and block DC, the voltage is on the +ve plate where it remains until it is discharged through R420. In real life there is some leakage through the cap to GND but this is small.

I've got a bit on, will try and have another look later on.

Noted that the RC timer is fed from the regulated output which would be a bit slow to discharge.

 

[curtisw99]

Credit for trying to understand the schematic

That thick black line that connects -ve of C412 and R431 is ground, 0V... it connects to the centre tap of the transformer, and also the GND connection between the main caps, again 0V


agree


Yes C412 is for RC delay, R419(220k) appears to be the R

The way I see it,
At power on C412 charges up through R419, as the voltage increases on C412 +ve, zener D419 goes into breakdown mode and current flows through the zener, this current turns on Q431 which then switches on Q432.
Q432 provides a path to GND for the relay current to flow (from D903, through R401/402, S16, S17 speaker selector switches then through the relay and then Q432c-e to GND)

Working backwards, the expected voltages are
Q432e = 0V (GND connected)
Q432b = +0.6V (standard Vbe = 0.6V abouts)
Q431e = Q432b = +0.6V
Q431b = +1.2V (0.6 + Vbe)
D419 cathode = +1.2 + 4.7(zener) = +5.9V

To speed up the discharge of C412 you could try reducing R420(220) to 150 or even 100ohms but must be 0.5W or higher.


Ideally capacitors pass AC and block DC, the voltage is on the +ve plate where it remains until it is discharged through R420. In real life there is some leakage through the cap to GND but this is small.

I've got a bit on, will try and have another look later on.

Noted that the RC timer is fed from the regulated output which would be a bit slow to discharge.

Your numbers are confirmed by my manual (see attached)
But
D419 on my DMM only reads 1.3v and 4.7v

 

[mbz]

D419 on my DMM only reads 1.3v and 4.7v

Yeah, we noticed this sometime before. It's a bit of a puzzle, maybe a service manual error?
I would think that there should be 4.7V difference between cathode and anode.
Maybe it has something to do with start up? Initially there is 0V at anode so zener starts conduct at 4.7V and gets locked on in that state??? It doesn't appears significant since the relays operate ok.

Did you try replacing R420 with 100ohms 0.5W? It will speed up the discharge of C412

That observation about no pop when relay contacts taped closed is difficult to reconcile

 

[curtisw99]

Yeah, we noticed this sometime before. It's a bit of a puzzle, maybe a service manual error?
I would think that there should be 4.7V difference between cathode and anode.
Maybe it has something to do with start up? Initially there is 0V at anode so zener starts conduct at 4.7V and gets locked on in that state??? It doesn't appears significant since the relays operate ok.

Did you try replacing R420 with 100ohms 0.5W? It will speed up the discharge of C412

That observation about no pop when relay contacts taped closed is difficult to reconcile

My STA 2080 and Kenwood 5030 both have significant delay between switch off and relay open. They make zero noise at shutoff.

I was checking if this JCX2900 would do the same, with the relay locked closed I simulated a long relay open delay. Since it did not pop, this confirmed my hypothesis.
that’s why I would like to chase adding more delay to the relay open.

 

[mbz]

Happy for you to try what you think, I disagree with the suggestion but open to new ideas.
As before, easiest way to add more delay is to remove R420, this would give a significant delay. Minor adjustments can be made by increasing either C412 say 470uf or ... and/or R420 say 1-2-,,,k. Do not run tests with spkrs connected, measure Vdc at speaker terminals

 

[curtisw99]

Happy for you to try what you think, I disagree with the suggestion but open to new ideas.
As before, easiest way to add more delay is to remove R420, this would give a significant delay. Minor adjustments can be made by increasing either C412 say 470uf or ... and/or R420 say 1-2-,,,k. Do not run tests with spkrs connected, measure Vdc at speaker terminals

My turn for bad comms...

I meant lift one leg or remove R420.
R420 is the bleed path for C412, with R420 removed, C412 will hold its charge and voltage longer keeping the relay closed. This will cause nasty voltages, probably a few volts not 200mV! C412 will discharge more slowly through secondary paths.

edit Working on loading the preamp into microcap12 simulator, probably a waste of time

Another long shot would be to build an audio probe (probe, series cap connected to disposable speaker). Then probe the preamp audio path trying to identify the stage where the pop originates

The service manual has R420 as 1000ohm in the wiring diagram and 220Ohm in the schematic.
I wonder if they were also playing with this resistor for the same reason

 

[mbz]

The service manual has R420 as 1000ohm in the wiring diagram and 220Ohm in the schematic.
I wonder if they were also playing with this resistor for the same reason

Try it, takes a few minutes... then you know.

 

[curtisw99]

Try it, takes a few minutes... then you know.

Nope, didn’t work, only delayed turn on.
I think next I will put a cap across the relay diode to see if I can keep the relay energized

 

[mbz]

Nope, didn’t work, only delayed turn on.

Not clear what you tried.
Certainly increasing C412 will delay the power on/relay click time. I'd expect a very small delay at power off, yeah probably not noticeable.

If you increased R420 then hard to see it increasing the delay at turn on since R420 only in play when power switch is off. Likewise if you removed R420.

 

[curtisw99]

Not clear what you tried.
Certainly increasing C412 will delay the power on/relay click time. I'd expect a very small delay at power off, yeah probably not noticeable.

If you increased R420 then hard to see it increasing the delay at turn on since R420 only in play when power switch is off. Likewise if you removed R420.

I changed the following
D419 to a 1W 4.7v zenner
C412 to 470uf 63v
R420 back to 1000Ohm per the wiring diagram

 

[mbz]

It is worth trying the removal of R420 however do not connect speakers, fully expect a dc spike, use MM. However focus should be on when the spike occurs, yours happens immediately at power off. Any delay would confirm your theory.

 

[curtisw99]

It is worth trying the removal of R420 however do not connect speakers, fully expect a dc spike, use MM. However focus should be on when the spike occurs, yours happens immediately at power off. Any delay would confirm your theory.

I went back to the preamp and measured voltage at shutoff . I measured at the pre-out jacks on the back. My multimeter isn’t fast, but I get at least a 1600mv DC spike out of the preamp at shutoff.

Is this normal?

And is there a dedicated set of filter caps for the preamp? I’m having trouble finding the preamp supply on the boards and in the schematic

I can’t try the R420, but the relay doesn’t close if I only have a multimeter in the speaker outs.

 

[mbz]

I get at least a 1600mv DC spike out of the preamp at shutoff.

Yeah, that's significant, bit different to the 200mV abouts measured at speakers.

You would really need someone with a JCX2900 to say if that is normal.
Generally I would be thinking that's a little high, certainly pops from preamps occur and that's the reason/purpose of the protect circuit, to break the speaker connection for the pop occurs.

The thing that gets me is the pop occurs so quickly/immediately at power off.

I need to have another look at the preamp. A common cause is the +ve/-ve power rails collapse at different rate. In this case it looks like a transistor is "immediately" shutting down causing the amp to be flooded with +ve? voltage

 

[mbz]

Have you tried removing audio path coupling caps to help identify area where pop is injected into audio.
The caps of interest are C551,557,562,565,591,592.
Suggest start with C591, need to do both L&R channel, expect no pop, then try C562.