Grrrr... darn glass diodes

[angrykoko]

So I'm happily building my uproar, one of the 1n34a's came pre-broken but tested ok.. put it in.. next ,next, next..snap!...   I hate those glass diodes.

So I head down to what must be the only supply store in Atlanta and pay 1.85 ea for 2 of them (robbery!)

But these new ones are different.. they are the 2 band variety and I wouldn't have thought anything of it until I saw midwayfairs post on the one's he has for sale..

So the question is..  should I put them in as a pair?  D1 & D2 = 2 band version, D3 & D4 single band.
Or mix em up D1 & D3 = 2 band version D2 & D4 = single band.

It probably wont make a hill of difference but thought I'd ask the collective knowledge before I break out the soldering iron.

[midwayfair]

Man, if you think clear glass diodes are fragile, don't go anywhere near black glass ones!

First, ALWAYS bend the leads with pliers. Second, if a glass diode is broken or cracked, you might as well trash it. Chances are it'll fail before you finish boxing up the circuit, and then you'll spend days debugging a big complicated board for a boxing mistake only to find it was a diode. (I've done that.)

Anyway. 1N34a will sound very slightly different from each other even in the same series, but for the most part, because their actual performance characteristics are similar or identical, the main differences will be the average forward voltage of any given series/package, which can vary from .3v to .35v (and no, that's not a big difference). I would say that if you're really curious whether the sound is different, use a socket. If you've got more than two diodes left, you might actually want to use the two with the closest forward voltage. But we're talking about really subtle changes in sound among diodes, and it's not like you can't mix different diode series to begin with. Personally I do it all the time.

[angrykoko]

Thanks midwayfair, that confirms what I was thinking.

Yeah, I have no game plan with bending leads on the glass diodes.  This one broke when I was inserting it into the pcb.. must have just stressed it in just the right way.   

Got any pics of how you bend and mount yours?  I was thinking of taking a minimalist approach and start mounting these standing on end in hopes that only one side could ever possibly become stressed.

Cheers and happy 4th!

[nzCdog]

I've broke a couple myself
+1 to Jon's suggestion, use needlenose pliers and be gentle

[JMart]

Hello guys! I think I may have a bad diode, and was wondering if any of you that have debugged a build to find diodes the culprit could help. Everything seems to be correct with the wiring (how many times has that been said and solder joints good - no shorts, etc.  But I'm getting a constant buzz with the pedal engaged. Its an Ibanez standard fuzz build Btw. Signal passes fine when effect is bypassed, but when engaged it just buzzes - pots, tone switch and clip switch have no effect on the buzz. I was just wondering if a cracked glass diode could potentially cause this type of problem? The reason i think its a diode is because two of these broke when I bent the leads down 90 degrees (with pliers). So I trashed them and bent two more.  After closer inspection, these two I installed look like they have a hairline crack on the underside- and like a dumb-a$$ I was too impatient to wait for the sockets to arrive in the mail (two days later I might add) and I soldered them in!!!  Any help is greatly appreciated! Thanks.

[angrykoko]

I have several with hairline cracks (or what look like hairline cracks) that were fine.
I've gotten in the habit of checking the diode.. bending it.. then checking it again before I solder it in, only adds a little extra time.

Have you built and tried an audio probe to trace where the problem starts?  Besides visual inspections for solder bridges, bad wiring etc, an audio probe is a really fast way to find where the problem starts and what it could be (like I cannot tell a bad solder joint by just looking at it).

Also take a picture of the board and components and upload it, there are many really smart folks here that can sometimes spot things really quick.

[JMart]

Thanks for the tips guys. I'll see if I can snap some picks and post them soon. I'm a newb to this so my knowledge is limited on the debugging with a probe. I'm slowly getting better and learning new things constantly. I've gotten 26 builds so far. I've had to debug quite a few of them, but 5 of them are still sitting in a drawer because it's something i believe is component related and not just a hidden short or wiring issue. I learn a good deal during the building process so I figured I'd get back around to those after I learn more about the circuits and what the components are actually "doing" to the audio signal, and not just "painting by number."  I've been reading up on how to build an audio probe to start that endeavor. I'm just waiting until I complete the little headphone amp + oscillator test rig I saw (on this forum I believe), and then I'll start trying to debug those 5 problematic builds.