Polk Powered Sub PSW150

[MuseChaser]

I picked up this sub for $10, knowing it didn't work. Also knowing that it's a notoriously unreliable piece of equipment that, when working, sounds OK.

Upon closer inspection, the driver itself seems sound.. no rubbing. Amp powers on, meaning red indicator LED lights up, but no sound at all, other than a VERY loud sharp crack when turning off the power. Soooo... removed and disassembled the amp. Now.. questions...

Question #1 - Does anyone have a schematic for the amp in this powered subwoofer? I spoke with Polk today and they informed me they did not have anything available for that model.

#2. I'm assuming this jumper wire I found underneath the control board is not original. Can anyone confirm that this is a bad attempt at fixing something rather than factory?


Also noticed that, on the same board, jumper wire W19 is missing, although there are remnants of that wire existing in the eyelets...



... also notice, in that pic, lots of brown goo around the green caps and slight deformation, although they measured within tolerance of the correct capacitance. I don't have anything that'll measure ESR.

And... on the main amp board, the large main 10,000 mfd 50V caps have brown crap underneath them, although they, too, measure within tolerance on my multimeter's capacitance function. Does this mean they're shot? Brown stuff indicated by yellow arrows...



At this point, I realize that the usual advice is, "scrap the amp, put in a new subwoofer plate amp and call it good," but I'm approaching this project as an opportunity to learn more about the repair process.

Anyone have any experience working on this amp? I've found YouLube videos addressing the "heartbeat" constant noise issue, but nothing specifically for no output at all except for a crack on power-off.

The phase toggle switch was completely frozen and inop, but I've found a replacement for that.

Any ideas/help would be, of course, appreciated.

[McBryce]

That brown gunk is just to stop the caps from flapping around in the breeze as Dave would say. It's meant to be there, although I've heard that it can become slightly conductive over time and cause problems. It's fine as long as it's not bridging any solder joints on the PCB or component pins.

McBryce.

[Jeff eelcr]

The brown glue is usually the type that is both conductive and corrosive the longer/hotter it gets.
It should be removed from areas as McBryce said. And it is still used even though this problem has been known for about 50 years.
Jeff

[CaptDon]

I fixed two Behringer mixer boards, both D.O.A. and both had that plyobond rubberized cement holding a small capacitor in place under the power board. Scraped off the glue with a dental pick and both units immediately powered up!! Kenwood ham transceivers were 100% failure rate due to this very same glue in the PLL area!! The notorius TS-440 series.

[MuseChaser]

Some followup annotated pictures of the top and bottom of the control board, illustrating the jumper wires I was questioning... anyone have any further input... OR... even better... have the schematic or a friend who works for Polk who can get one?  Thanks...

[McBryce]

It's a single sided PCB with a relatively simple circuit, surely you can draw up the circuit yourself in a few minutes!?

McBryce.

[MuseChaser]

In a few minutes? Not even close...sure wish I had the experience to be able to do that. A few hours, maybe, and then another couple hours checking my work and finding all the mistakes I probably made. I HAVE started that process, however...I'm trying to help myself as much as possible, and I'm sure sketching and generating a schematic, if nothing else, is good practice. I've already generated a full component list for the board (except for a few transistors....can't see those well enough on the board), and that sure helped me learn to translate resistor and capacitor codes quicker!

Here's a question for you... how would generating a schematic of the existing circuit tell me whether or not the jumpers in question are as-they-came-from-the-factory or not? I want to compare "what is" with "what was intended" and see if they match. If your response is along the lines of, "well, it'll be obvious once you sketch it out," you're giving me way too much credit. At my level of experience, the only things obvious are charred components, bad solder joints, solder bridges, and other obvious stuff. Hence, my plea for help.

[McBryce]

I would sketch out the entire circuit and then evaluate how the circuit performs with and without the wire to decide which is the intended function for the end user.

McBryce.

[Audiorepair]

The left end of the white wire is connected to ground.

So effectively this wire is shorting pin 12 of the quad op-amp to ground.

[MuseChaser]

That would be true, I think, if W19 was in place, but it's not.  The white wire is not connected in any way to the quad opamp... at least that I can see.

BUT... more in a second!!!

[MuseChaser]

I stared at the circuit a bunch, and the more I though about how the added wire routed a signal meant for the 6080D D+ input to the non-inverting input of one of the 4558 chips, the less sense that made. Same with the removal of W19, which PREVENTED the signal from getting to the 6080 D+ input. All I can figure is that, when the phase switch went defective, someone with really poor eyesight attempted some really bad way of bypassing it... who knows.

Throwing caution to the wind, I replaced W19, removed the added under-board insulated jumper wire, and bridged the center and 0deg tabs as a means of bypassing the faulty phase switch I removed.

Bound the speaker leads to a dummy load and powered it up. Red indicator light, no magic smoke left... so far so good.

Hooked up the scope, fed some test signals... Hmmm... traces look good, frequencies are correct, a little bit of oscillation up around 150K, but otherwise nothing remarkable. Low pass adjustment and volume adjustments work. Sooo..

Tried music input instead of test tones. Traces as expected... Sooooo....

Hooked it up to a cheap speaker, just in case, and tried it at low level...

Works! I need to reapply some damping silicone or hot glue before placing it in the original cabinet with the driver, but I think we're out of the woods.. at least for now.

Here's hoping!!

[MuseChaser]

The left end of the white wire is connected to ground.

So effectively this wire is shorting pin 12 of the quad op-amp to ground.

A kind soul found a copy of the schematic and sent it to me, and I see what you mean now.  Pin 12 of the quad op-amp, the anode of D109, cathode of D108, and pin 3 of45580 all connect to ground in the schematic.  I'm not sure what was gained by removing the jumper W19, which connected the diode node (if that's the correct usage of the term... still learning) to pin 12, then rerouting the diode node to pin 3 of the 45580 which is grounded.  I can't trace path from pin 12 of the 6080D to ground on the board, but it's got to be there somewhere.. maybe I'm missing seeing a jumper wire that doesn't show in the schematic.  In any case, though, you were of course correct about pin 12 needing to be grounded.  Maybe a previous repair person saw the schematic, couldn't trace the path to ground either, and decided to just create one with the wire. 

In any case, the wire is gone, W19 is back in place, and the amp is working fine, at least for short tests so far.  I reinstalled it in the cabinet, only to find that the corrugated spider had come unglued from the basket causing a bunch of buzzing and rattling, so I reglued it and it's drying now. BUT... at least the amp worked well. Tomorrow, I'll know if this all works! 

Thanks again.