Power supply silliness - Is it a bad supply?

[Mark Hammer]

I loaned out my Line 6 M5 to a friend last year.  The M5 comes with a sweet little switching power supply that I like because it is compact and has a status LED on it.

When he eventually returned the unit, it wouldn't power up.  I mean, a menu for adjusting the screen would show up, but I couldn't get the pedal to do anythng more than that.  What up?

I popped the enclosure to see if there was anything obviously damaged inside, or telltale signs of damage, and there was nothing.  I plugged the power supply into a couple of analog pedals, and they powered up just fine.

But there was a conspicuous whine in the background.  None of the analog pedals had a clock of any sort in them (eg, BBD-based or a charge pump), and the whine was absent when battery-powered, so I could only assume the whine came from the power supply.

Brought the M5 to a music store last night, and asked if they could try it with a known good supply.  If the pedal powered up, the I'd buy the supply.  They tried it.  It worked.  I bought the supply.

Not sure what my friend did to corrupt the previous wall-wart, or what might have been corrupted.  But the bottom line is that the PS no longer worked properly, but the M5 did....which made me happy.

[blackieNYC]

Switching supplies often whine when they are dying.  Can be helpful in troubleshooting.  Sometimes the whine (maybe the high crew osc itself) is up around 50kHz, bleeds everywhere, and every piece of gear in the studio thinks it's close enough to 48k and tries to clock to it. A catastrophe of remarkable proportions. Switchers seem like the only game in town now. A little scary to me.

[Transmogrifox]

Audio tones from a switcher usually mean to me that the thing is unstable -- especially if you hear the whine when connected to analog pedals.

Why did it go unstable?  Perhaps it was something as simple as unplugging it from the unit hot created a kickback on a filter inductor which subsequently dumped its energy into feedback components and damaged the feedback circuitry in some manner. 

Might be if you looked at the output with a scope you would find it in a sort of a limit-cycle state.

Maybe the switcher design is flawed from the get-go and you just happened to have the one that combines the correct evil conditions on component tolerances to hit the unstable state.  Shoot freeze spray on it and it might appear to work for a while.

Switchers seem like the only game in town now. A little scary to me.

Not scary to me.  I love them -- low power, low heat, light-weight and ultra-clean with minimal filtering components.  The problem is if they are done cheaply and/or carelessly there's a lot more that can go wrong than with a linear supply. 

DIY SMPS is not really a hard thing to do -- just takes more care with layout considerations.