DCC++ Motor Shield getting hot/smoking with nothing connected

Jeff Bernardis Oct 26, 2018

  1. Jeff Bernardis

    Jeff Bernardis New Member

    I have an Arduino Mega 2650 and an Arduino Motor Shield. I have flashed the arduino with the DCC++ base station code, and can interact with it through the serial window in the arduino IDE.

    On the Shield, I have cut the traces as instructed - even though I don't think it's necessary until I hook up a voltage supply. I have also jumpered pins 2 and 13. Neither the external voltage connector nor the track outputs have anything connected to them.

    When I mount the shield onto the Arduino and then connect USB to the PC, one of the smaller chips on the shield gets very hot very quickly. My first thought was that maybe it just runs hot, so I watched it for a while. It didn't take long before I noticed a bit of smoke. I quickly disconnected everything.

    I don't know if I damaged the shield or not, but that aside, can anybody think of what I might have done wrong in setting this up? There is nothing connected to the output, so there should be no current flowing, so I don't understand where the heat is coming from.

    Thanks in advance.
  2. NScaleKen

    NScaleKen Permanently dispatched

    I know the shield is damaged from your description of events, completely positive of it if your account is accurate. Whether it was something you did or just a manufacturing issue is debatable.

    My guess is manufacturing issue, or tiny bit of conductive material from cutting traces or some other work got under the chips contacts and caused a short. I would say definitely dont use it, I dont think it can be returned after cutting traces but at least they aren't super expensive like branded DCC components (couldn't hurt to ask though! They are meant to be cut in some cases I guess?). I hope the Arduino is OK. Probably wise to check the output of the arduino's pins with a multimeter before installing a new shield as well, since it might be the cause.

    Bummer about the smoke but that is a very very bad sign and you dont want to damage more components hoping a known bad hardware device might somehow provide function, even though it literally turned part of itself to vapor and ash due to an internal issue.
  3. Jeff Bernardis

    Jeff Bernardis New Member

    Yeah - that's my assessment of the situation too. But it's just baffling. Nothing is connected to the board inputs or outputs. The only voltage on the board should be logic voltage from the arduino - there should be nothing going through the motor circuitry.

    BTW the arduino is fine - I tested it with some other software.
  4. Jeff Bernardis

    Jeff Bernardis New Member

    I did a little more digging. The chip that is overheating is the LMV358 Op Amp. I'm not quite sure what this is for, but it looks like it converts the SNSA and SNSB signals from the motor driver chip to whatever level the arduino is looking for on analog pins 0 and 1.
  5. Xmtrman

    Xmtrman TrainBoard Member

    Solid state devices operate with an internal supply of magic smoke. Once it leaks out, you're done.

    Seriously though, non-brand shields, hats and modules have a dismal reliability rate. To make them so much cheaper than their branded counterparts, they cut ALL the corners. Individual components that don't make test are not discarded. They're sold to 2nd, 3rd & 4th-tier manufacturers.

    If you're lucky, the short-comings don't affect your application.
    If you're not, you could end up with a few hundred dollars worth of models on the floor with a runaway.
  6. Jeff Bernardis

    Jeff Bernardis New Member

    I think I'm going to try the pololu route. I have had good success with pololu components in the past. I like that it has an additional amp available too.
  7. Onizukachan

    Onizukachan TrainBoard Supporter


Share This Page