Shorted Switchbox

When you’re working under a wind-blown tarp with 48mm of rain coming down in 24 hours, water creeps in where it really shouldn’t, and gear shorts out. Most geophysics equipment is amiable to field repairs, using twisted wire and clamps instead of soldering or including spare parts to pop in and out, and packing a hairdryer in the toolbag can work wonders, but sometimes, it’s all just too much.

Shorted node.

Shorted node.

The good nodes are shiny black, while the now-bad node is matte grey and smells of burnt electronics. A damaged, burnt component of an integrated circuit board is not field-repairable.


The damage was severe enough it wasn’t camp-fixable, either, nor even ‘Bring back to the office machine shop’ fixable, and had to be sent back to the manufacturer. The listed cause of the short was water, and required the replacement of bad relays, ICs, and diodes, and the replacement of a multi-switch.

This entry was posted in Geoscience, Practice of Science and tagged , , , , , , , . Bookmark the permalink.

5 Responses to Shorted Switchbox

  1. steve gough says:

    The gray donut-on-stilts-shaped thing *looks* like a polyfuse; should not burn up, but breaks circuit in overload and auto resets when it cools down. Being burned means big bad overload. If the short is external and you find it and fix it and can take the risk; you could jumper across this thing. Of course then the fireworks will really be big if the short occurs again. I am not a lawyer or doctor, etc…..

    • Mika says:

      Interesting! I have two pieces of broken equipment and I can’t test this one out until I get the other one running. It’s also possible that it has a redundant port next to it.

      I’m fairly certain that with 48mm of rain in 24 hours for a few days, pretty much everything is going to short and keep on shorting. This is the joy of geophysics in the Pacific Northwest!

  2. really says:

    Actually the board looks really dry. I would not dismiss the problem as caused by rain. As you probably got lots of good readings before now in the same rain. And others before you and still more even now.

    To get the necessary current to do a good “cook job” on a part like that would probably require a direct or very very close connection to some serious power and then there would be the need for this power to return to its source as easily. that would require a two way street.

    So the real challenge once the repairs are made is to figure out how the trouble was caused so to be preventable again.

    the part is a tranzorb it is designed to pass excess voltage (and current) to ground it is
    testable in the field with a mulitmeter. It should have very high resistance across the its two pins in both directions. Would go to low resistance at voltage marked as its part number if you limit the current.

    Jumpering the thing as indicated by steve above would probably be 180 degrees in the wrong direction. I doubt a lawyer or a doctor would do much better. Not their “field” of expertise.

    The use of twisted wires and binding posts is fine on small jobs but would be real slow doing many miles and miles four wires at a time. As well as picking up and laying the four wires out each time to get the next reading. Taking a series of readings writing by hand each to average out the values repeat in reverse polarity etc…

    • Mika says:

      The case and boards all had condensation on them when I first opened it; this photo was taken over 48 hours later when I’d finished very slowly and carefully drying it out.

      I used a replacement switchbox to finish the job. I’m looking forward to tearing this one apart and seeing if it is possible to work around the burnt node once it arrives in the office this week. The glorious thing about truly broken equipment is that making it more broken when experimenting with it really isn’t a concern.

  3. Mika says:

    From Facebook, additional comments:

    Seth: looks like there are two ports for redundancy in case one goes bad?
    Mika: I actually have two pieces of broken gear, and this is second in the chain. Once I get the first one going, I have a few ideas on how to bypass this poor, melted contraption…

    Chris: I told you not to lick the terminals, but no one ever listens.
    Mika: Mmmm, polyfuse… it’s irresistibly tasty!
    Chris: The smokey flavor inside diodes and fuses is just so irresistible!

    Josh: the magic smoke escaped, huh? is it a cap or a coil(inductor)? looks like only one dead channel out of 6. If you need that one and not another you can swap components; you remembered your field soldering iron right?
    Mika: It’s actually one out of about 20 components, and I need them all. Camp fixes are thwarting me so far. It looks like I’m on standby until new gear arrives Friday. Argh!

Leave a Reply

Your email address will not be published. Required fields are marked *