Learning project: repair a junked audio receiver!

[garrettm]

Looks like the 5V stby circuit goes to Vdd on IC801. So we definately need to get this rail working.

Just to confirm, you never powered up the amp without D753 in place right? Cause that would be bad. I don't think IC801 would like roughly 12V going into it.

To test where the loading is coming from, remove diodes D714, D710, and D709. Check to make sure each diode is okay. Then check the output of the zener D753 with them removed. If it's okay, reinstall D709 and then check the output of D753. If okay, reinstall D710 and repeat. And again with D714 until we find which trace is causing the loading.

We may also want to check C707 (main board) and C803 (display board) capacitors. While unlikey to be the problem, it would be good to make sure they are healthy.

[de_light]

All the diodes have been pulled and are testing OK. The C707 supercap(? Terminology) I pulled as well.

I'll run the rest of the tests in the next few days but I'll report back - work calls!

[garrettm]

All the diodes have been pulled and are testing OK. The C707 supercap(? Terminology) I pulled as well.

I'll run the rest of the tests in the next few days but I'll report back - work calls!

Good to hear the diodes are okay. They are used to split the 5V standby into two separate 5V rails: "Back Up" and "Main". The Back Up rail is then OR-ed with D714 and D709 using two different 5V sources (Zeners D712 and D753). I'm not entirely sure why they did this. I suppose, it might be to power IC801 so the remote control can turn the unit on from stby, and once on, the power requirements are such that the second source picks up the slack.

D753 should measure 5V with D714, D710, and D709 removed. D710 and D709 will let us know which rail (Back Up or Main) is lugging D753 down.

[de_light]

Super interesting. I have removed D714, D709, D710 and Q752 (well, the emitter lead for this broke due to my cack-handed removal technique so I have to source a replacement transistor....MPSA06 should work OK from my local supplier).

D753 measuring 6.3VDC with all the above removed! Super interested to find out what is pulling it down.

I'm also going to replace C707, I think. It behaves like it is totally open. I can't get seem to charge it whatsoever - 5.5VDC with a 22 Ohm resistor should give a time constant of 1.034s. 5RC to get the 99.5% charge ratio gives approx 5.2s. I've waited over a minute and never get the voltage above 2V across the terminals. Multimeter shows no continuity whatsoever and capacitance is measured at 1.26nF on the meter.

[garrettm]

D753 measuring 6.3VDC with all the above removed! Super interested to find out what is pulling it down.

That seems a tad high. The datasheet shows we should see 5.3 to 5.6 volts at D753. But maybe it just needs some loading? I guess there is the forward voltage drop of D709 and D710 to help reduce that.

I'm also going to replace C707, I think. It behaves like it is totally open. I can't get seem to charge it whatsoever - 5.5VDC with a 22 Ohm resistor should give a time constant of 1.034s. 5RC to get the 99.5% charge ratio gives approx 5.2s. I've waited over a minute and never get the voltage above 2V across the terminals. Multimeter shows no continuity whatsoever and capacitance is measured at 1.26nF on the meter.

That definitely sounds dead. And it looks like it failed open circuit...so we still have the short circuit to find. In the meantime, I think any 10V electrolytic that fits will do as a substitute.

[de_light]

Great day for repairs. I'm fairly confident I've found the fault on the rail.

First things first, I replaced the C707 supercap. New one charged as a cap normally should, so I was happy

I then replaced D753 just in case with a 5.6V zener - read 5.6V with D709, D710 and D714 out.

I then inserted D709 and the voltage across D753 dropped to 3.3V. This is the 5V backup rail. I looked at the schematic at possible culprits shorting to ground and identified C803, C804 and C805.

C803 had continuity to ground on both leads and a short across the leads - hurrah! As I was desoldering it, one of the leads just disintegrated - the bottom of the cap is bulging and appears to have corroded? Just waiting on a replacement now.

You did say to check C803, Garrett!

[garrettm]

Excellent work de_light! Looks like C803 ruptured at some point...not a good sign. Hopefully the traces aren't corroded.

While rare, electrolytics can fail short circuit--usually while rupturing from excessive internal pressure. Otherwise they just tend to dry out and die gently in their sleep. In the early to mid 2000s many Japanese capacitors were plagued with bad electrolyte compositions that caused them to rupture violently or slowly leak out and die. If your amp dates from that time period, you may want to test the remaining capacitors to be safe. If they haven't failed by now they are probably okay to leave if the ESR and capacitance are still reasonably close to spec.

https://en.wikipedia.org/wiki/Capacitor_plague

Have you gotten a chance to fire up the amp with the new components? Assuming those were the only bad capacitors on the 5V rails, the display should be working now. You may want to check the voltages on the other rails too, as there could be more bad capacitors hiding around.

[de_light]

Hmm.

I put it the new cap and inserted the remaining components (having replaced D714 with a 1N4148 since I lost it).

I also discovered C802 had corroded away - it is a tiny 0.1uF/50V cap that I haven't replaced. There is a tiny amount of green corrosion around one of the legs of Q101 but it is intact.

Still having issues! The front panel equaliser controls etc light up, but the LCD has an intermittent fault. It either displays nothing, or displays random stuff- for example, a single segment lit up. Or a random 'AM' will appear.

Measuring the voltages with all the components replaced:

6.2VDC backup
3.8VDC 5V main rail

(These measurements are taken coming off the J4 pins)


1. From the previous replies, the low voltage suggested a short (as the resistance was lower due to the short to ground) - this suggests this may be the cause on the 5V main rail now?
2. I've tried to take a voltage measurement off the VDC pin - I seem to be getting around 1.4V to that pin?

[garrettm]

If shorted, C802 could be holding down the reset line on IC801. This may be part of the problem with the display acting up. I'd also check C801.

The reset circuit itself looks a bit odd, but is pretty simple. Since the stand by circuit appears to always be on, Q801 will only conduct when the amplifier is first pluged into the wall. While the 5V standby rails stabilize, C801 charges up and turns on Q801--pulling the reset line low. After the standby rails reach a certain point, Q801 turns back off. So the reset pin should be held high when the 5V standby rails are stable. I attached a TINA simulation of this. The transistor and biasing resistors might not be exact, but the principle is the same.

1. From the previous replies, the low voltage suggested a short (as the resistance was lower due to the short to ground) - this suggests this may be the cause on the 5V main rail now?

Yeah. Likely another capacitor, since these all seem to be failing shorted.

2. I've tried to take a voltage measurement off the VDC pin - I seem to be getting around 1.4V to that pin?

I'm not sure where this is. Is this on IC801 or somewhere else?

[de_light]

Quick update: collateral damage from that ruptured cap is more than I though. Q801 is faulty, open on all leads and testing configurations. Switch 5805 was nonfunctional, 5807 is intermittent. I need to source some spares now.

Like that TINA simulation a lot - very cool!

Edit: Realised that Q801 is a digital transisor/internally biased. I've read that testing these with a DMM is difficult:

"
The addition of R1 makes testing with a multimeter other than for shorts more difficult. With a VOM, you should see a difference in the B-E and B-C junctions in the forward and reverse directions. However, a DMM will probably read open across all pairs of terminals. "

Might hold off replacing it until I've excluded the other components as the problem.

[greenpossum]

The 100uF cap on C802, do you think I'd be okay to replace that with a simple ceramic 104 cap?

You realise that 104 means 100nF?

[de_light]

Sorry - post edited. Meant 0.1uF.
I've ordered the appropriate electrolytic from RS Online now so problem solved.

[garrettm]

Like that TINA simulation a lot - very cool!

TINA-TI is free to download from Texas Instruments. There is also LTspice from Analog Devices that's free to download as well. They both look like they came straight out of a time machine from 1995, but are really helpful when testing out simple circuits and ideas.

Realised that Q801 is a digital transisor/internally biased. I've read that testing these with a DMM is difficult

Should be pretty easy to test this type, as they are intended to be used as switches. Simply hook your DMM up as an ohmmeter to the collector and emitter and then connect your DC power supply between base and emitter. Then slowly increase the applied voltage till it turns on. The datasheet shows 1 to 10 volts for the I-V plot, so 1 or 2 volts should be enough to turn it on and get a reading at the DMM. You can also test in circuit, the collector to gnd voltage should be about +0.6V for maybe 100-200ms after the amplifier is plugged into the wall and then rise to about +5V and stay there. But this assumes both capacitors are in good health.