Is it worth trying to fix a power supply?

[Fama]

I was using my huge pedalboard a little ago and suddenly it started malfunctioning. Just a slow motorboating type sound, some pedals not lighting up, smelled like burning. I finally got around to troubleshooting in more detail, and it looks like my Truetone 1Spot CS12 is broken.

To be more accurate, the changeable 9V/12V outputs are outputting less than 1V each in either mode. Same with the 18V outputs on that same side. The other outputs seem to work fine, although I only tested them one by one with a multimeter and not under load.

What do you think, is it worth opening it up and trying to fix it? Obviously it has 230V running into it, and I wouldn't plug it in while the case is open, but I'm feeling a bit iffy about doing DIY repairs on it.

I bought it used over a year ago, and the date code seems to say it's from 2015 originally, so no chances of warranty.

Edit: I popped it open, each output has it's own voltage regulator (or at least I assume that's what they are, huge heat sinks). I doubt 6 of those just broke at the same time, but I can't see where there could be a split between either side (first 6 do not work, the other 6 presumably work, I didn't test the AC but the other 5 work), or if they are somehow partially chained and the issue could be where the chain is. A ton of caps and diodes, no visible signs of problems.

 

[Feral Feline]

Glad you're keeping the old supply Fama, because as your abilities and confidence grow, you may come back to it and get it working. ( I sure hope so!)

I'm with Chris Knudson on checking to make sure the big caps are fully drained even if it's been a few minutes.
Since you're like me and new to the high-voltage stuff, and even if we were old hands (hacks?) at it...

It's just simply good procedure/practice/habit to ALWAYS make sure everything's drained — it doesn't take long to do so, and if ever something goes awry with whatever equipment you're working on and after a few minutes it doesn't drain itself for whatever reason...

 

[Fama]

Glad you're keeping the old supply Fama, because as your abilities and confidence grow, you may come back to it and get it working. ( I sure hope so!)

I'm with Chris Knudson on checking to make sure the big caps are fully drained even if it's been a few minutes.
Since you're like me and new to the high-voltage stuff, and even if we were old hands (hacks?) at it...

It's just simply good procedure/practice/habit to ALWAYS make sure everything's drained — it doesn't take long to do so, and if ever something goes awry with whatever equipment you're working on and after a few minutes it doesn't drain itself for whatever reason...

That's a good point, at this stage it has been unplugged for weeks I think, so I'm fairly sure it's all drained, but like you said - it's a good habit to develop (and practice with something where I probably won't get a shock even if I mess it up a little).

I probably wouldn't have tried this if I didn't get a really good tip about where the issue is because there's just so many components, and I figured it wouldn't be something as simple as a diode on one of the channels, so that's definitely at least worth a shot.

 

[mzy12]

I am gonna reverse engineer this power supply at some stage, because it is very good, but has some less than perfect design choices. (Power LED runs on the 9VAC transformer and the fact that the 9VAC is derived from a transformer on the mains means you still have to choose the input AC voltage). Obviously, Feral is correct in mentioning you need to discharge the caps, normally I do mention it any time I discuss SMPS stuff, I just forgot this time haha. Only time I've ever got zapped was from one of those bad boys!

 

[Spiff4565]

My Voodoo Labs Pedal Power 2 started humming.

Popped the cover off and found one of the 1000uF caps bulging. Took it out and the dielectric was leaking out the bottom.

So I replaced all the electrolytic caps. Him was greatly reduced but not zero.

Replaced all electrical with low ESR caps. Unfortunately I didn't order the correct quantity of the 10uF's, but those will arrive today.

I also replaced all eight of the power regulators, just because I was already in desoldering mode. Those are now replaced.

The only thing I have not replaced are these things, as I have no idea what the are:




I don't believe there is anything exotic or overtly technical in a typical PS. I'd say crack the beast open and see what's under the hood.

 

[Fama]

My Voodoo Labs Pedal Power 2 started humming.

Popped the cover off and found one of the 1000uF caps bulging. Took it out and the dielectric was leaking out the bottom.

So I replaced all the electrolytic caps. Him was greatly reduced but not zero.

Replaced all electrical with low ESR caps. Unfortunately I didn't order the correct quantity of the 10uF's, but those will arrive today.

I also replaced all eight of the power regulators, just because I was already in desoldering mode. Those are now replaced.

The only thing I have not replaced are these things, as I have no idea what the are:
View attachment 76212

View attachment 76213

I don't believe there is anything exotic or overtly technical in a typical PS. I'd say crack the beast open and see what's under the hood.

Best I could find for AM151 is that it might be a bridge rectifier diode, and here could be an alternative https://www.newark.com/multicomp-pr...h/dp/99AC4581?rpsku=rel3:AM151&isexcsku=false which does look sort of similar. But I could also be very wrong, and I can't make anything of the presumed manufacturer name above the AM151. Do the components have 4 legs (you should be able to check on the underside)?

 

[Spiff4565]

Best I could find for AM151 is that it might be a bridge rectifier diode, and here could be an alternative https://www.newark.com/multicomp-pr...h/dp/99AC4581?rpsku=rel3:AM151&isexcsku=false which does look sort of similar. But I could also be very wrong, and I can't make anything of the presumed manufacturer name above the AM151. Do the components have 4 legs (you should be able to check on the underside)?

Yes, it has four legs. Thank you for researching this. so far after replacing all the caps and power regulators it appears to be cured of its hum.

 

[andrewsrea]

Yes, it has four legs. Thank you for researching this. so far after replacing all the caps and power regulators it appears to be cured of its hum.

The mystery component is a 'bridge rectifier,' consisting of an interval array of four 1N4003 or 1N4004 equivalent diodes. They either work (change AC to DC with a small ripple, hence the filter caps and chokes), or don't they work (typically short out, then fry then measure 'open'), It is unlikely with what you have said so far, that it needs to be replaced.

 

[Fama]

Hello! I was meant to comment on this like a week ago at OP's request. But life happened and I had to look after someone close to me.

I intentionally purchased a CS12 with this defect and in order to fix it. Not a lot of resources online - Truetone seems to be good with warranty repairs, as that is the go to recommendation.

I found that in my case, there was a short circuit across one of the diodes on the underside of the board. I actually emailed Truetone on the hopes I could get schematics - no such luck. Very unfortunate that companies are no longer willing to give out schematics so that people can repair stuff themselves. The support person I was in contact with was actually able to give me a page of their internal troubleshooting guideline. I won't post it here as I doubt I have permission to do so, but the relevant part is as follows:

2.5 Whole output group is NG
2.5.1 Check which output group has no voltage or is abnormal (Outputs 1-6 or 7-11).
2.5.2 When the DC power is off, use a discharge resistor to discharge C1 for 5 seconds and
check whether there is any abnormality (like shortͲcircuit or low resistance) at each
output of the failed group by measuring the resistance of each DC Jack (J101 ~ J1201)
2.5.3 Measure the output diode (D101 ~ D1101) resistance of abnormal output. If its resistance is about 0~100,
replace this diode with a US1G diode (D101-601) or DO-201/HER303 (D701-1101).

(Note that this troubleshooting guide is slightly inaccurate, the AC output jack SHOULD show very low resistance, as it is connected directly to a low voltage transformer with very low DCR windings.)

After I removed the shorted diode, the output came back up, though with lots of ripple. I need to order a replacement part in my next mouser run before I actually plug a pedal into it!

It took me a while to get around to it, but I finally got working on this. And the issue was exactly as you described! D301 was shorted, I desoldered it and the rest of the outputs came back to life. I have no way of checking for ripple, but I won't use it until I get a replacement part. The old part was a US1M and not a US1G, but checking the data sheet (I assume it's this https://www.vishay.com/docs/88768/us1_test_dcicons.pdf, or at least it's close enough) I don't think it matters which one I would use.

Thanks again a lot for your help!

 

[mzy12]

Yep, diode shouldn't matter tooooo much. Keep VF similar, make sure its rated for the current, peak and continuous reverse voltage and make sure that the reverse recovery time is good enough for whatever its rectifying. US1G is good in that position, which I guess is why they recommended it

I'm glad it's back up for you If you have a multimeter that allows you to measure ripple current on DC voltages, I would check that on all the outputs effected by the bad diode before plugging in a pedal. If you have access to an oscilloscope, even better. This will give you a good idea about how 'fixed' the supply actually is. In my case, it was just the diode, but if for whatever reason a capacitor or some voltage regulator or something to do with the switching caused the diode to give up the ghost, you'll want to figure that out before you plug a pedal into it

 

[Fama]

Yep, diode shouldn't matter tooooo much. Keep VF similar, make sure its rated for the current, peak and continuous reverse voltage and make sure that the reverse recovery time is good enough for whatever its rectifying. US1G is good in that position, which I guess is why they recommended it

I'm glad it's back up for you If you have a multimeter that allows you to measure ripple current on DC voltages, I would check that on all the outputs effected by the bad diode before plugging in a pedal. If you have access to an oscilloscope, even better. This will give you a good idea about how 'fixed' the supply actually is. In my case, it was just the diode, but if for whatever reason a capacitor or some voltage regulator or something to do with the switching caused the diode to give up the ghost, you'll want to figure that out before you plug a pedal into it

Hmm, yeah, that sounds like a good idea - but no, I just have the bare minimum of multimeters (no connectivity beeper mode either, I just use the lowest resistance to check for connections) and nothing better.

 

[mzy12]

It is unlikely with what you have said so far, that it needs to be replaced.

Definitely would raise my eyebrow if replacing a bridge rectifier cured the hum. Maybe some heat caused deficiency in the epoxy package that encases the silicon and the diodes started capacitively coupling to each other. Only other guess I'd have is that some electrolyte got to the leads and started creating resistive or capacitive paths for current. If you haven't cleaned the board with iso-propyl alcohol or some other similar solvent appropriate for board work, @Spiff4565, I would be greatly tempted to do so myself. And I imagine you've done so already, but inspect the board for any heat damage or corosive damage from the electrolytes in the cap.

Shotgun replacing components is not necessarily the best way to repair a device, especially one as complicated as a modern powersupply (switching or not) as you may not find the actual root cause of the issue. I am reminded of people repairing old Fender amps finding DC present across coupling caps, then replacing all of them on the board as they beilieve them to be leaky, which will reduce the DC voltage down to zero when they measure it after the repair. This hides the actual cause of the leakage, which is moisture and dust gathering in the wax coating the board. Heat will dry it out, but not fix the actual problem plaguing the board.

 

[andrewsrea]

Definitely would raise my eyebrow if replacing a bridge rectifier cured the hum. Maybe some heat caused deficiency in the epoxy package that encases the silicon and the diodes started capacitively coupling to each other. Only other guess I'd have is that some electrolyte got to the leads and started creating resistive or capacitive paths for current. If you haven't cleaned the board with iso-propyl alcohol or some other similar solvent appropriate for board work, @Spiff4565, I would be greatly tempted to do so myself. And I imagine you've done so already, but inspect the board for any heat damage or corosive damage from the electrolytes in the cap.

Shotgun replacing components is not necessarily the best way to repair a device, especially one as complicated as a modern powersupply (switching or not) as you may not find the actual root cause of the issue. I am reminded of people repairing old Fender amps finding DC present across coupling caps, then replacing all of them on the board as they beilieve them to be leaky, which will reduce the DC voltage down to zero when they measure it after the repair. This hides the actual cause of the leakage, which is moisture and dust gathering in the wax coating the board. Heat will dry it out, but not fix the actual problem plaguing the board.

I've been repairing tube amps since the 70's and another like issue for vintage Fenders is a poor repair. It is easy to overheat the eyelet or try and reflow old solder, causing the wax and solder to flow to surrounding eyelets. It creates the same parasitic effects you are describing.

 

[Fama]

Just to verify for anyone who might run into this issue in the future, after changing the diode I get a steady 9.43 volts or so from all of the outputs (same as the ones that already worked before). Like I said, I don't think I have any way to test for ripple or anything like that, but I'll try it carefully with some pedals first to see if they sound wonky at all. Thanks again for everyone's help, especially @mzy12 !