DIYstompboxes.com

DIY Stompboxes => Building your own stompbox => Topic started by: cd on August 12, 2004, 05:15:26 PM

Title: How many different ways can you kill a circuit/pedal?
Post by: cd on August 12, 2004, 05:15:26 PM
How many things/ways can you think of that a circuit (typical guitar pedal form) can be destroyed/toasted/killed?  A friend of mine brought me a fairly expenive boutqiue pedal that didn't work anymore.  LED didn't light, no effect, bypass worked (naturally, since it's true bypass), so it must be a power problem.  I open up the pedal, the circuit is a regular opamp based cookbook circuit, nothing special.  Power protection is through a series diode, that checked out fine.  DC jack is fine, battery snap and solder joints are all OK.  Opamp is socketed so I replaced it with a known good one, voila, pedal works fine, except the LED is burned out.

So what could have blown the opamp (and LED) and left the series diode intact?  Friend says he didn't go anything unusual, just left it on his pedalboard.  The interesting thing about the circuit is that besides the opamp and gain setting resistors (and other stuff for bias & housekeeping), there are no other parts in the signal path!!!!  That's right, no caps or anything.  Is it good practice (or even advisable) to continually plug/unplug a guitar and other FX directly into/out of an opamp?
Title: How many different ways can you kill a circuit/pedal?
Post by: Peter Snowberg on August 12, 2004, 06:25:38 PM
No coupling caps at all???? :shock:

Not very many opamps are content to have their outputs shorted as you plug and unplug. An NE5532 will be fine with it, but the odd RC4558 won't be quite as happy.

The only thing that would blow the LED would be too much voltage. As that could cook the opamp too maybe that's what happened. You usually need LOTS of voltage to cook an opamp though.  :?

Maybe the failure was in two parts?
Title: How many different ways can you kill a circuit/pedal?
Post by: Paul Perry (Frostwave) on August 12, 2004, 09:30:40 PM
If there are NO coupling capsd (!!) then, you could plug the input signal lead (or the output) to a high voltage source & blow the op amp (depending on any input resistors) and then, depending on circuit, maybe take out other parts. Without damaging the 'protection' diode.
A leaky guitar amp input cap might rise up to a voltagfe that would damage stuff that was DC-coupled to it.
Someone once plugged the control voltage output of my midi to CV converter into a high power organ amp output, whihc melted some of the PCB traces (one curled up like a watchspring!) fortunately a cheap locic chip acted as a fuse & protected the expensive parts  :)
Title: How many different ways can you kill a circuit/pedal?
Post by: smoguzbenjamin on August 13, 2004, 07:00:55 AM
If that's a boutique pedal and you payed much $$$$$ for it, I'd go back and demand that the guy puts coupling caps in there! Back practice to have no caps in it. Might be good from an audiophile viewpoint but then we're guitarists, not HiFi wacko's.
Title: How many different ways can you kill a circuit/pedal?
Post by: cd on August 13, 2004, 08:25:56 AM
Yup, no coupling caps.  Pretty nutty, huh?  Luckily I didn't pay for it  - I don't know WHY there are no caps, aside from the audiophile wacko-ness (which isn't a part of the pedal's marketing hype).  I don't know what kind of opamp is in there either, the top has been sanded off, so it could be anything.  Strange.

Thanks for all the thoughts, guys.
Title: How many different ways can you kill a circuit/pedal?
Post by: gez on August 13, 2004, 12:42:33 PM
Is it dual-supply?  Or does it have a JFET buffer at the input?
Title: How many different ways can you kill a circuit/pedal?
Post by: R.G. on August 13, 2004, 12:51:53 PM
You might be interested in the article "When Good Opamps Go Bad" at GEO.

An opamp input connected directly to an input jack, no explicit protection on it, will sooner or later be destroyed by the electrical wasteland that can be connected to it. Things like leaky amplifier AC power can be thrashing the opamp's ground around by 30Vac easily, and then something as simple as *grounding* the tip of a cord connected to it will exceed the common mode range of the opamp, send it into latchup or cause damaging currents to flow inside it because the diffusion wells in the chip are forward biased.

I'm not surprised that it's dead.

And to answer your question about how many ways a circuit can be killed - it's infinite.
Title: .
Post by: petemoore on August 13, 2004, 01:16:39 PM
You can tell if it's a common dual OA or single OA by the power supply. Ca3080's and 386's etc. notwithstanding.
 Dual gets V+ at pin 8, single at Pin 7, you may need to trace past any small resistors or PS protection diodes etc.?] between the battery clip+  and the chips V+ pin to use DMM beep mode, for finding the chip type.