Cambridge Audio Topaz SR20 not turning on from standby

[hexis]

Hello.
I will keep it short and hopefully somebody with more knowledge can help me.
I am fairly new and want to try and fix the amp as i like the sound of it. Worked perfectly for 7-8 months since bought it used, but one morning would not turn on from standby (about two months ago or so did the same and I disconnected the power cable and tried again and it worked ok).
When i try to power on there is a relay (Massuse ME-11 005-2Z40 )on the "power input" board from the wall that clicks 2 times , but the third louder click does not happen anymore. How to start testing if the relay is faulty or there is another fault. Nothing seems burned or defective at first sight.
Any help would be appreciated.
Thank you

 

[mbz]

When i try to power on there is a relay (Massuse ME-11 005-2Z40 )on the "power input" board from the wall that clicks 2 times , but the third louder click does not happen anymore.

Are you certain that this relay is supposed to click 3 times? The amp has at least 2 relays so maybe another relay is not clicking on.
Does the amp stay on or does it always look dead.

Reads like the amp is in protection mode where an excessive voltage is detected and the amp does not connect the speakers (via relay) because damage may be done to speakers. The other possibility is that there is a failure in the protect circuit, ie, amp does not see excessive voltage and orders the speaker relay to close but a fault prevents this.

First step is to test for excessive voltage at the output emitter resistors, somewhat large 3 legged white items near the heatsink. You will need a multimeter, connect black probe to chassis/ground and measure dc voltage at the middle leg of the resistor, repeat for the other resistor, probably safer to use something like minnigrabers. upto 50mV is kinda ok, 1-100V is notok.

Amp may be in protection due to over current, ie a short or over temperature, start with excessive output voltage

 

[hexis]

thank you for your advice.
The Amp is only in standby (the small led on the front near the power button is on)
Ordered a new Omron equivalent relay and the same behaviour so indeed it seems like it is somehow in protection.The third click 'could' be one of the speaker relay (there are 2 relays before the speaker terminals as 4 speakers can be connected) i now think.
I assume the measurement will be at the pins of the 2 big white ressistors from the picture?
I am a newbie but i can follow any step that can help so any advice is really appreciated.Thank you
L.E. I have measured the middle pin of both big white ressistors and the Vdc is 0.4mv (near 0)(red lead on pin , black on chassis metal)

 

[mbz]

L.E. I have measured the middle pin of both big white ressistors and the Vdc is 0.4mv (near 0)(red lead on pin , black on chassis metal)

Suspect you measured the voltage in standby mode, need to capture the voltage in power-on state, ie, hook up the multimeter, set scale to 20Vdc(?) then hit the power on, there will be a flash of numbers on the MM, ballpark what is the highest number? 1-2V or 40-50V or...

Thinking some more, the amp reverts to standby to prevent further damage. The amp can be safely powered up with excessive dc voltage, simply don't engage the speaker relay. Makes me think there is an over current/short issue, will have another look at owners manual/schematic.

 

[mbz]

The unit is microprossor controlled, so something of a PITA. It has a smallish heatsink with
fan forced cooling. Fan ON when heatsink at about 50-70C, thermal protection/shutdown when heatsink
at 110C. The unit also has overload/high current protection detected at the output emitter resistors.

Probable scenario is that the unit thinks it is in thermal or overload protection and shuts down to
prevent further damage.

Check for overload protection.
Attempt power-on while monitoring dc voltage at R62 marked in red. Important to use minigrabbers or similiar.
one slip of probes could cause extensive damage. Anything approaching 0Vdc(-0.5 to 0.8V) would indicate overload.

Check for thermal protection.
Attempt power-on while monitoring dc voltage at R70 marked in yellow.
Appears a digital "high" (greater than 2Vdc) is required

Appears to be test points/a test pad located nearby, ok to test there if safe.

 

[hexis]

measure at R62 on the red markings on the schematic on the Vdc is between 0.3mV - 1.2mV when i try to power on.So it might indicate overload protection.
At R70 measured at the yellow markings the Vdc is 0 when i try to power on.
Any advice where i have to check for the cause of the overload protection?Thank you.

 

[mbz]

Any advice where i have to check for the cause of the overload protection?

Suggest measure again at the emitter resistors to identify faulty L or R channel and also to confirm "real" overload rather than a failed component in the overload protect circuit. Measure the voltage between the two outer legs of the large white 3 legged resistor while powering up. These are the same resistors you previously measured at. Repeat for the other channel. This corresponds to the bias voltage, so maybe 40mV. A fault will be in the 1-2Vdc range.

 

[hexis]

measured the Vdc between the outer "legs" of the 2 white resistors and while i power on it stays at 0V dc.Tried multiple times.Same.

 

[mbz]

measured the Vdc between the outer "legs" of the 2 white resistors and while i power on it stays at 0V dc.Tried multiple times.Same.

Points to a failed component in the protect circuit.

Test Vdc at R96 expect either near 0(off) or about 0.6V(on)
If R96 near 0V,
diode test Q11, in circuit ok, looking for a collector-emitter short

If R96 0.6V (approx)
Measure Vdc across R102

If R102 approx 0V (off) diode test Q13, again looking for a c-e short
If R102 appox 0.6V then diode test D1, in circuit ok.
Also diode test Q10 again looking for c-e fault.
Will get parts locations shortly.

Nice graphic from markthefixer on diode testing. Red/Black indicate meter probe orientation.
2 results expected, either 650mV(approx) or OL/Open Loop/circuit

 

[mbz]

 

[hexis]

" Test Vdc at R96 expect either near 0(off) or about 0.6V(on)
If R96 near 0V,
diode test Q11, in circuit ok, looking for a collector-emitter short "

I have tested R96 at power on and the Vdc is 0.
then tested Q11 ( Which is a C1815 NPN Transistor).As on this transistor the emitter and collector are pins 1 and 2 i could not test with the board on and powering on , so i desoldered the transistor from the board and measured the resistance (checked on youtube how) and the transistor is NOT shorted between the collector and emitter ( no resistence reading).Tested all the pins and all are as expected.
Any advice from here?
Thank you for your help mbz.

 

[mbz]

Prefer that you use the diode test function of your mm (look for diode symbol), it puts out a larger test current than simple resistance measurement.
A quirk of diode testing is that if an item fails testing then it is faulty and that a part that passes testing may still be faulty, ie, the test is not absolute.
Will look at schematic...

EDIT Probably nothing, SM has Q11 as KTC3198, you have KSC1815, a good sub. Perhaps someone has been here before. Do you remember if the solder blobs on Q11 looked original. Or have you owned this amp from new?

 

[mbz]

R62 is at 0Vdc and goes to pin 26 of the micro processor. R96 is at 0Vdc, makes Q11 suspect. Do you have a spare KSC1815, they are like a 12cent part so not worth a mouser order. I do't want to risk power up with Q11 removed. Do you have a Dim Bulb Tester? (google AK)

 

[hexis]

R62 is at 0Vdc and goes to pin 26 of the micro processor. R96 is at 0Vdc, makes Q11 suspect. Do you have a spare KSC1815, they are like a 12cent part so not worth a mouser order. I do't want to risk power up with Q11 removed. Do you have a Dim Bulb Tester? (google AK)

View attachment 2785169

tested again Q11 by taking it out and performing the diode test and there is no short between the collector and the emitter. tested al the other combinations between base, collector and emitter and all is as expected for a NPN transistor.
The soldering was original as there are other transistors like q11 in the area.

 

[mbz]

In that case I need to go over my analysis, maybe some fresh eyes/ideas are needed, hopefully others will chime in.

 

[HPMguy]

Is this a microprocessor controlled circuit with power applied all the time to the low voltage power supply for the microprocessor? Given a few years of "always on" operation the filter caps for that 5V supply go bad -- I've had this happen in any number of remote controlled devices. Check for the presence of 5V on the Vcc of the microprocessor, and for AC on that line, most of the time I've found a bad cap somewhere acting like a resistor pulling the 5V line down or a filter cap gone south allowing significant 120Hz AC on the 5V line, and in either case the processor will not operate and so nothing else does either.

 

[hexis]

This might make sense because if it wasn't for the standby mode i would think this would not happen, because it might have spent a lot of time in standby at the previous owner. any idea where to check for this? Thank you
L.E. Measured the Vdc at the pins (GND to D+5v) of the connector that comes from the power supply board in stand-by and i get 20-25mv , is that ok?
Measured the Vac ? at between the GND and the bottom pin (cannot read well on the board what +5v is that DC OR AC) and get 10.5vac.
Any advice? Thank you

 

[HPMguy]

Ugh, I dislike these modern schematics. Bits and pieces everywhere, so hard to trace things down.

Several possibilities -- check for 5.6V on one side of D22 and 5.0V on the other, if that diode is open the microprocessor will have no voltage and won't work. Also check the caps at C108, C112, and C106 as they are the most likely to be bad and drop the voltage if they have serious leakage.

If i get a chance I'll dig around some more, hard to read sideways on the computer....

You should also check for DC offset on the output transistor emitters, plenty of things can cause enough DC there to keep it in protection.

It looks to me as if the amp is powered up, and least the digital part, any time the power switch on the rear is in the "on" position, and the power button on the front panel just switches the inputs into operation -- might also apply power to the amp board, I'll have to dig around to see.

If the previous owner left it switched into standby, the filter caps at least for the digital power supply have been energized since 2017, quite likely dead. I have three DVD player with this issue, power indicator came on but would not work, replaced the PS caps and they came back to life.

 

[mbz]

hard to read sideways on the computer....

pdf has a rotate function. eg, for firefox hit the ">>" button top right the select rotate CCW.
Trying to get some ideas from your ideas...

 

[hexis]

Checked Vdc at D22 and i got ~150 -180mV on stand-by and ~450mV when i push the power on switch.
Checked C106 , C108, C112 (out of the board) using the resistance test and they all charge and discharge as they should.At visual inspection they look as new.