Muff's Modules & More

[wsy]

I was in the middle of a patch (taiko drums - it worked out pretty well for the low
registers at least, but that's another story) and found that one of my new Dotcom
christmas-green cables wasn't working right.

Not only wasn't it passing signal, but anything I multed it with didn't pass signal either!

Out with the ohmmeter- tip conducts to tip. Sleeve conducts to sleeve. But -
tip and sleeve conduct to each other!!! Uh-oh.

I take apart the ends- they're perfect. Nicely soldered, with the Dotcom heatshrink
reinforcement. But worse- no sign of the short.

I carefully inspect the coax itself (Dotcom uses nice 20-gauge stranded center
conductor coax for cables - better than I have access to at work!). No pinches,
no pins, no flat spots. Nothing to show where the problem is.

So I figure, what the ****. Sending the cable back to Roger will take a couple weeks
and fifteen bucks postage, and the cable is only $7 anyway. I may as well try blowing
the short clear with a nice high-current pulse. At least the fireworks will show me
where the problem *WAS*.

So, I go out to the garage, put on safety glasses, hold the sleeve of one
end against the (-) post of an old truck battery, and stab the tip against the +
terminal. There's a nice spark - and then nothing. Ohm-meter shows I've
successfully cleared the short.

The cable now works fine. In fact, it's in the taiko patch.

So, keep this in mind if you're ever stuck out with a shorted cable and in a "fly or
die" situation... car batteries can be your friend!

- Bill

[obscurerobot]

[wsy]

Quite true - but in this _particular_ case, it wasn't a hazard, because the battery in
question was a sealed-case AGM battery for my old 4WD. It has recombiners on each
cell.

But in the general case, you are correct.

- Bill

[daverj]

You're lucky that whatever was causing the short was fairly small and melted when you did this. It is definitely a dangerous thing to do and not a good idea.

[wsy]

Actually, hitting a piece of signal-shorted piece of electronics with a wallop of
low voltage hi-amperage current is a well-estabished procedure.

Googling shows up "zapping" for nicads to remove shorts:

http://www.instructables.com/id/Revive-Nicad-Batteries-by-Zapping-with -a-Welder/
http://www.ka7oei.com/nicds.html
http://www.repairfaq.org/REPAIR/F_audiofaq9.html

and many others as well (though till I googled just now, I didn't know about the
idea of using an arc welder as the power supply!

I first learned about it many years ago from an old DEC engineer. The DEC PDP-10
printed circuit boards came through with about a 50% "dead short" rate - that is,
the + and - power planes shorted together somewhere - and this was fresh
out of the multilayer press, no chips or solder yet. And even then, an empty
circuit board of PDP-10 size was $$$.

So, effectively those boards were total losses. But some engineer got smart:
he said "the short is probably a much smaller "whisker" of metal than the main board
conductors, so if we just burn the whisker off, the board will be fine."

"But what if it destroys the board?"

"So what? We were going to toss the board anyway."

So he pulled two of the biggest power supplies the stockroom had (something
like 500 amps each), and wired up a test set with all of the powers connected,
all of the grounds connected, and a big contactor to the dual 500's in parallel.

He put a shorted board in, hit the foot switch and a thousand amperes of linearly
regulated DC goodness surged into the board. The power supplies growled
and groaned. The bad board lit up in a half dozen places as the shorting
whiskers fused and vaporized. Then the shorts winked out, in rapid succession.

The board tested out "perfect", including a point-to-point test for signal integrity.

"Blasting" the board became a standard procedure at DEC. And yes, he
got a double promotion for this - one for the fix, and one for showing that the
shorted boards weren't just shorted in one place, but many (much to the
consternation of the board fab people). No, it was not me, but yes, I've known
him for almost thirty years now.

Since then, I've seen "blasting" or "zapping" in several other contexts (including
regenerating shorted NiCd cells; same problem with whiskering).

Sure, wear your safety glasses, maybe leather gloves, and do it outdoors if you
are at all unsure. But it's not particularly dangerous compared to, say, riding a motorcycle.

- Bill

[pulse_divider]

Amazing!

[daverj]

With a big fat cable there's no way of knowing it was a whisker. And shorting a battery (with something that isn't a whisker) is much more dangerous than shorting a regulated power supply with current limiting.

[denim dan]

just wanted to add to the above conversation that this is "standard procedure" in welding/fabrication shops. ground the welder with a clamp to the table and fire away. testing/fixing all sorts electronic of crap this way. just not welder to welder obviously

if you said "uh, isn't that dangerous?" you'd get laughed out the door.

[pugix]

Nice. Can I send you all my shorted dot com cables to repair?

[Juxwl]

[wsy]

[pugix]

[wsy]

[revtor]

Someone make a machine that does this, that looks sleek, with some gague or two on it, and someone else write up a semi plausible theory "white paper" and Ill make the website and we'll sell it for 10,000 to audiophiles who will use it to blast their speaker cables and power cords..

~Steve

[daverj]

To sell for that much, it should have a lot of brass (steam punk style) on it, and have a silver plated knife switch on the front.

[CJ Miller]

+1

I want a set of your PDP-10 PCB clones when they're ready

[Graham Hinton]