Peavey XR8600D power supply cycling

[Delta362]

A Peavey XR8600D came in with the complaint that it is no longer passing sound. I opened the unit up and found one obvious issue on the power supply board. R233 and R234 (10ohm 1W resistors) were split into two and no longer soldered to the board. I temporarily replaced the R233, R234, R236, R260 network with a 10 ohm 5W axial resistor as I continue troubleshooting.

With my LBL (light bulb limiter) in place, I powered the unit on and found the front panel LEDs operating strangely so I check the power supplies going to the mixer board. The +18, -18 and +48V supplies dropping really low. Pulling the ribbon connector, those voltages stayed up so I then pulled the mixer board.

On the mixer board, I found U26 bad (4565 opamp) as well as C300 (470uF+ 6.3V) swollen. I replaced U26 and C300, plus U25 and C301. C236 (47uF+ 35V) was shorted and replaced. R460 (100 ohm 1/10W) on the PowerAmp2 jack was also failed open and replaced. I tested the mixer board by itself using a dedicated +/-15V power supply and it all seems to be good for the mixing board now.

I then reconnected the mixer ribbon to the power amp board and powered the unit on. Without the LBL, the relay was clicking in and out almost immediately. I removed the ribbon connector and confirmed the power amp does the same thing without the LBL. With no mixer connected, and using the LBL, the power supply comes on and stays up for about 2 minutes before the relay begins cycling. I am trying to use this time to troubleshoot the power supply.

Using the LBL, and the Speakon jacks board disconnected, I measure +7.8VDC on the speaker Red and Blue Faston connectors. The +/-18V regulators read correct. My +/-30V rails are around 26VDC, and the +/-80V rails are around 68V. Q103, Q104, Q203 and Q204 are not shorted. The +330V rail reads +142V to chassis, and the negative half of that rail reads -142V to chassis. I have not found any shorted diodes or transistors as of yet.

I'm looking for some guidance on how to troubleshoot this power supply. Thanks!

[Delta362]

Made some progress on it, but still not entirely solved. I found that I had to have the Speakon connector board plugged in, though I don't think it's truly an accurate statement of the amp's design. The +7.75VDC offset that was on both sets of outputs reduced down to +4.8VDC with that board plugged in as it provided a 10k load per side.

The mixer board ribbon cable was reconnected. The unit powers on and stays on for several minutes (at least). I didn't have it powered on super long as I continue to take measurements and see if I can find the anomaly. The RVBPWR pin reads +13.1VDC, while the +/-30V rails are pretty much spot on, as are the +/-18V and +/-80V rails.

Given that both the left and right outputs have the same positive DC offset on them, while the +80VNL and -80VNL rails are within (absolute) 0.25V of each other, I suspect it's something common to the two. The bias voltage (relative to chassis) measured -50VDC (pin 1), -64.0V (pin 2/4), -65.0V (pin 3). Looking at the voltages across the pins (as opposed to chassis), they seem reasonable for the IC8 functioning properly.

What should I be looking for in regards to the same DC offset on both outputs?

[Axtman]

I am familiar with Peavey amps but not this model. I have found that sometimes amplifier circuits do not like light bulb limiters. The reduced current screws with them somehow. (SWR amps come to mind.)

I also had an issue with a voltage regulator in a PV amp where it worked fine for a little bit then started cycling.

I hope this helps.

[Delta362]

Thanks, Axtman! I'm currently able to power it on without the LBL and no speakers. But I still have the 4.8VDC offset on each speaker that I'm having trouble figuring out what may be causing that.

[x-pro]

На что мне следует обращать внимание в отношении одинакового смещения постоянного тока на обоих выходах?

Is it the same on both outputs?
So there's zero between the outputs?
Then you don't have to worry too much. Especially since the supply voltages are all normal.
You didn't specify whether there is a signal at the inputs of the power amplifier.
Is there an audio signal on POST_VR_A and POST_VR_B?
What is the state of HV_SHUT_DOWN ?

[Delta362]

Output A has +4.8VDC between terminals Red_1 and White_3. Output B has +4.8VDC between terminals Blue_1 and White_1.

HV_Shutdown has 0.08V on it, with +28.9V on D46K.

If I remove the 10k loads by unplugging the Speakon board, then the amp goes back to cycling the power relay. That also coincides with +7.7VDC on each output and the D46A triggering.

So while the amp doesn't cycle with the Speakon board connected, there's still almost 5VDC that would be going to each speaker. My gut instinct told me to expect near 0 before hooking up any load or speaker load.

I then just tried putting an 8 ohm load on one of the Speakon connections. The +4.8VDC for that particular channel dropped to 0, while the channel without the 8 ohm load remained at +4.8VDC. What a revelation that was! Thank you!

I have not tried to inject a signal, but will do that tomorrow. Frankly, I'm surprised as I would not have considered putting any load on the output with a DC level that high. But with your comments, and thinking back to how the 10k load brought it down some, I finally tried a dummy load (successfully). Tomorrow will be the real test.

[x-pro]

Initially, I thought the circuit had the load in a bridge connection, considering that the signals arriving at U6 are inverted with respect to each other after it.
But there are class D amplifier circuits where the output without load can be even half of the supply voltage of the same polarity.
Before connecting the speakers, check the second amplifier as well as the first. On a resistive load.

[Delta362]

Some updates. I had signal on both channels passing thru IC1 pins 1 and 7 and on one side of the resistor (R105, R205) on their way to the IRS2092S (IC102, IC202), but nothing out of the outputs. Then, unfortunately had a probe slip on Q103 and now the power supply won't kick start. An unexpected setback. I will be troubleshooting that next.