Debugging a factory MXR Phase 90 with a hot battery

[BobP]

Hi all - I bought a broken Phase 90 (Rev. D) and am trying to debug it but the battery gets crazy hot right away. So far all I've been able to achieve was:

- getting a 9.31v reading at the board, which dropped to .86v with a cable plugged in;
- working LED

I've read about the hot battery issue and since this is a factory unit I don't know where the short would be, unlike say a solder bridge of a home built unit. Some of the components (4 of the 5 transistors, 1 diode) have a clear amber colored 'sealant' looking material (dry) at the solder points (under side of board) that seems to be bleeding onto other traces. It seems factory and yet out of place as well. (Apologies if this is some common thing, I am a noob.)

Also, in the short time I had the (new) battery plugged in, it cooked it to the point that its output was 8.3 when I unplugged it, and 8.5 about 20 mins later. I don't think it will last long enough to get reliable meter readings so, what next?

[Paul Marossy]

In my experience, a hot battery is the result of a direct short of the postive leg of the power supply to ground. Does the thing work at all? Does it work with a wall wart, or same thing with that?

Maybe the battery snap has an internal short or someone that didn't know what they were doing tried to replace the battery snap and wired it backwards.

That "clear amber colored 'sealant' looking material (dry)" is most likely solder rosin. You can carefully scrape it off, it won't hurt anything.

[BobP]

Effect isn't working.

Battery cable checks out ok. It didn't come with a wall wart, but the guy I bought from said it didn't work either way (AC or 9V).

[Paul Marossy]

OK, well that makes sense. If the power supply was shorting out, then it shouldn't be working, and indeed it isn't.

Look for something that is shorting the positive side of the power supply to ground. There's got to be something going on there. Maybe you could try removing the polarity protection diode on the power supply and see what that that does. Maybe it's shorting the power supply directly to ground because it got damaged or defeated somehow. If that fixes your hot battery problem, then the next thing is verifying that the active components on the circuitboard aren't all fried as well.

[Fender3D]

Follow the PCB tracks starting from battery clip or DC socket, you'll arrive to a diode, check if it's shorted...
If the guy said AC because he applied AC, then almost all active parts are dead


Edit:
3 secs. Paul....

[Paul Marossy]

Follow the PCB tracks starting from battery clip or DC socket, you'll arrive to a diode, check if it's shorted...
If the guy said AC because he applied AC, then almost all active parts are dead

Yeah, I suspect fried diode and all dead components. I'm not sure because I never asked the question... if a diode fails, does it fail open or closed?

EDIT: Did a quick search, and apparently diodes can fail closed. That's what my bet is on, it failed closed.

[wavley]

Follow the PCB tracks starting from battery clip or DC socket, you'll arrive to a diode, check if it's shorted...
If the guy said AC because he applied AC, then almost all active parts are dead

Yeah, I suspect fried diode and all dead components. I'm not sure because I never asked the question... if a diode fails, does it fail open or closed?

EDIT: Did a quick search, and apparently diodes can fail closed. That's what my bet is on, it failed closed.

I believe that closed is the most common failure mode for diodes, that's why you see them in things like the hum x so even if they fail you still have a safety ground (but your ground loop is back)

So a big +1 from me too

[R.G.]

This situation has all the hallmarks of Death-By-AC-Adapter.

The most common reverse polarity protection for pedals is the reverse biased shunt diode. When an **AC** adapter is plugged into the socket, the diode shorts the adapter every half cycle, usually overheats after a short time, and the heat then kills the diode. The diode fails shorted in most cases. If you remove the AC adapter before the diode burns open, as it sometimes does, you can save the signal circuits from being killed by the repetitive reverse voltage. This sounds like a random diode failure or a mild case of D-BACA.

[Paul Marossy]

The most common reverse polarity protection for pedals is the reverse biased shunt diode. When an **AC** adapter is plugged into the socket, the diode shorts the adapter every half cycle, usually overheats after a short time, and the heat then kills the diode. The diode fails shorted in most cases.

Good information to know. Never had that happen to me, but I've seen this one come up time and again where someone tries to run their pedal on an AC adapter and kills it.

[BobP]

Thanks everyone-

Sounds like this effect is toast but I'd like to follow the advice to test the diode.

I took a photo of the 9V and AC circuit and I don't see where the diode connects



The 3 horizontal connections top far right are for the 9V connector, which was installed with left to right being -, +, and blank (no wire).
'+' is going to one of three points on the AC connector, the 'blank' connection looks to be going to the sleeve of the input jack.

The 3 components below the 9V are not touching the traces, the leads are poking up. FWIW, 1 and 3 show no continuity, 2 shows .6 ohms.

[Govmnt_Lacky]

The fact that you keep referring to the "AC connector" kinda worries me. 

You do know that this box is supposed to be fed with 9V DC correct?

I suspect that someone may have performed the usual "incorrect wall wart plug-in" and sent 9VAC into the circuit. In this case, most likely the better part of ALL your passive devices are probably shot! 

[Fender3D]

...I don't see where the diode connects

On the other side (components side) just see where the diode IS, first...