ART SLA1 - Sticking relay?

[wayward bluesman]

I recently repaired the power supply on an ART SLA1 studio amplifier (shorted power transformer, of all things) and am having trouble with the protection relay on Channel 2.

SLA1.pdfSLA1.pdf

When I power the amp on, the relays are supposed to click "on" after a second or two. Channel 1 works as it should. Channel 2's relay will not click on unless I flick it or tap it. It takes almost no force at all. But once it's "on," Channel 2 works normally as well. What's throwing me off is that all the voltages I'm reading seem to be right (and match those on Channel 1).

Voltages:
28 VDC @ cathode D6
14 VDC @ anode D6
0.72 VDC @ pin 6 U4

I have replaced the relay RY2 as well as D6. R61 reads 14.27k (close enough to 15k). Should I suspect U4 or Q20? Is it possible that, despite the voltages being correct, U4 isn't sourcing the current the relay needs? Any other suggestions?

[Enzo]

Some thoughts:

It seems you are measuring the voltage across the diode, and inferring the relay coil voltage. Normally we do this as a matter of convenience, but in a case where voltage is present but no relay action, we would want to check directly at the relay coil pins. HOWEVER, the fact there is 14v dropped across the diode, implies the relay coil resistance is in the circuit, otherwise there would be 28v across the diode probably.

The circuit starts out with protect LEDs lit, then turns then off when the relays should turn on. SO does the ch2 LED light at first and go out as it should?

Can I assume those voltage readings are stable? And that they are taken while the amp is in the failure mode? Monitoring the D6 anode, does the voltage there change at all when you flick the thing into operation?

Just to be on the same page, when the relay does not close, and you flick it on. Are we talking the actual mechanical click not happening? Or does it click, but no circuit closure occurs? If ther is a cracked trace on the board in the output wiring, the coil operating would not insure signal passing.


Ah, here is one. R61 is called out as 1.5k, not 15k. SO if you have 14k of resistance, that might explain it.

And when a resistor reads way off like that, I always reverse my leads and re-measure it. If I get 14k both directions then I believe it. If I get 14k one way but "negative ohms" the other way, then I know there is a little residual voltage in the circuit confusing my meter. If you lifted the resistor then that is moot.

You can exercise the relay externally. Ground a 1.5k resistor at one end and use a clip wire and touch the free end of it to the anode of D6. That should complete the circuit and enegize the relay. WIll it consistently do so? Or does it behave the same way?

For that matter just ground the low end of R61. That would normally fire the relay consistently. If that works but U4 can;t do it, then I might suspect U4. But remember that the same pin of U4 also runs the LED.

[wayward bluesman]

To answer your questions:

The voltage readings are stable and are taken while the amp is in "normal mode," and the voltage at D6 does not change when I flick the relay. This is the sequence of events at power on:
First 1-2 seconds of power on
28 VDC @ cathode D6 (and D3)
27.2 VDC @ anode D6 (and D3)

After 1-2 seconds, Relay 1 clicks and simultaneously...
28 VDC @ cathode D6 (and D3)
14 VDC @ anode D6 (and D3)

Specifically, when Relay 2 clicks (after a tap), the circuit closes and Channel 2 operates normally.

Also, R61 actually reads 1.427k (NOT 14.27k). Major typo.

Something else you pointed out that I hadn't even noticed is that the protect LEDs (for both Channels 1 and 2) NEVER come on, which I assume means that Q8 and Q9 aren't turning on, which could also mean that U2 and U4 are faulty. The B+ supply voltage for the LEDs is there (about 14V) and they'll both work if I clip in a 2.7k resistor to ground at either cathode.

Those LEDs not lighting is pretty big; I don't know how I didn't notice. I suppose I'll go ahead and replace Q8, Q9, U2, and U4 just to be safe. I had to replace the main filter caps after the transformer transplant. I wonder if that uPC1237 chip (U2 and U4) is pretty sensitive to high supply voltages or poorly filter DC.

[Enzo]

No that sounds like too much. WHy would all those things fail at the same time? If the other relay works, and this one does sorta, then I have to think the ICs are working. WHat do the LEDs have in common? Power supply missing? I don't have the diagram up at the moment.

[wayward bluesman]

Unlikely, sure. But both LEDs aren't lighting and the only thing they have in common is the B+ (nominally 15V). For whatever reason, neither Q8 nor Q9 is turning on and providing the path to ground. I only figured the fault was either with the transistors or the UPCs. If I ground either R26 or R28, the LEDs light right up.

[Enzo]

You could be right, it just doesn;t sit right on my guts. They may run off 15v, but they also have to connect places. And a broken connection or cracked pc trace can do that.

Those LEDs are to me a clue. Use them for diagnostic purposes. You got power supply on to of them? Got voltage flowing through them when their driver transistors are off?

Are the bottom of the drive transistors making it to ground?

Watch the bases, are turn on levels happening there?

If we get the LEDs going as they should, it is possible that whatever is ailing them is the underlying cause of the relay problem. Multiple problems in units have a way of winding up related. And sorting one often sorts the other. And even if they don;t fix the relay, they may illuminate its problem some.

[wayward bluesman]

Update:
As far as I can tell, all connections are good. The LEDs have power to both of them, each drops about 1.5 volts, they're each connected to their respective resistor (R26, R28), which are in turn connected to the collectors of each 2sc945. The emitters both have solid ground connections. However, the base voltage never goes above 0.5ish volts - definitely not enough to turn the transistor on. After the delay, the base voltage drops to something negligible. R39, R40, R30, and R61 all read the correct resistance and are connected as they should be. The only thing I can figure is that both UPCs are faulty OR the V+ from the power supply is lower than it actually should be...

[Enzo]

The bases are probably the limiting factor. The diode formed by the B-E junction sets that .5v. Pull the transistor and measure the voltage that now appears at the base connection pad, that will be the real voltage.

COme to think of it, you have a 1k to 47k voltage divider off pin 6 of the IC down to those bases. WHen the IC is off, there is about 28v there, right? SO we are talking a little over .5v from that alone. Maybe it is a trifle shy of turn on.

SInce the one channel works the relay, how do voltages compare?

Test the leds, Short across the transistor E-C, that should light the LED. If they light that way, then the transistor is either bad or not being driven.

[dmeek]

Could someone re-post the schematic for this amp?

Thanks

Doug

[g1]

This should be it:

[dmeek]

That's it. Thanks g-one.

[jmartin919]

Having similar issues with mine and looking for help. Power LED comes on but the Protect LEDs do not and the relays do not click on (with or without source connected). If I short the E and C of Q8 and Q9 the LEDs come on. If I short the high side of R40 to ground the relay clicks on and plays the source on Channel 1. Same with R61 on Channel 2. At turn on Pin 6 of U2 and U4 is 28v, and then after a couple seconds drops to .7V. The base of Q8 is .5v at turn on and then drops to 0V. Voltages on U2 are as follows (steady state): pins 1-3 0V, 4 1.1V, 5 0V, 6 .7V, 7 2.2V, 8 3V. Same on U4. Relays have 28V on both pins at power up and then drops to 14V on one side but do not click on.
If I touch one end of lead on high side of R40 (14V) and touch the other end to the anode of D2 the channel 1 relay clicks on and STAYS on (even after removing the jumper) and provides output. Same thing happens on channel 2 but the relay does not stay on if I remove the jumper. When the relays are on one side is 28V and the other is 14 - same as when they are not on.

[jmartin919]

Well I never did figure out exactly what was wrong with it but I came up with a "fix". Even though the voltages at the relays and Protect LEDs seemed right I didn't seem to be getting the required current flow through the transistors (Q8 and Q9 ). My solution was to shunt R30/R39 with a 47K resistor - that made the LEDs work - and R40/R61 with a 1K resistor - that made the relays work. (Original resistor values were correct. Shunt resistors were just what I had on hand). It's been running for a couple days like that and works great. Anybody see a problem with what I did? Could I be over driving something (Q8/9, U2/4)? I did notice that the relays click on at slightly different times but don't know if this is normal since I never had a functioning unit before the fix.