Help me put the smoke back in to a set of DMX controllers

[Factotum]

Hi there everyone. Thought I might throw this one out to the hive mind.

Scenario :

I have three sets of 30 channel DMX controllers :

https://www.aliexpress.com/item/32993494983.html?spm=a2g0s.9042311.0.0.27424c4dpow2er

I had them all wired up to a DMX shield on an Arduino.

I wasn't happy with the quality of the DMX signal so I hooked up my scope to pin 3 (maybe 2) of the DMX line. I then managed to briefly short the ground clip of the scope probe to the +12v supply line while heading for the ground terminal (well done Bumblefork!).

Unsurprisingly they don't work any more (*).

When I say don't work I mean they don't respond to DMX data. They *do* work if I use the DIP switches to put them in demo mode.

So I have the DMX units connected to a known good DMX device (an Enttec DMX USB Pro2) and have begun probing away.


The DMX data on pins 2 + 3 go in to a SN75176B differential bus transceiver (8 pin SOIC on low LHS near DIP switch). I've had a look at the output (pin 1) [see image] and it doesn't look out of place.



The signal feeds pin 4 of the "mystery" 28 pin IC. There's a crystal (24.000 MHz) on pins 6,7 and the output is on pin 13 (when in demo mode). There's no markings on the IC. There is a logo / marking in the listing photo but not clear enough to make out.

I'm at a loss as to what has failed here - the transceiver looks OK, but I can't see how the over voltage would have made it by the transceiver and damaged what I presume is a MCU. Could always swap out the transceiver. Happy to do some DaveCAD drawings if required...

Any thoughts on how to proceed or what to look at? I know it's only $150 of gear but I'd like to redeem myself

[coromonadalix]

I have dismantled quite a few  dmx smoke generator,  all of them where china stuff

All the models i saw where working with STC  micro controllers,  tried to read a few,   but they where all protected from read ??

The STC cpus i saw where serial interfaced for the programmation ...

good luck

EDIT

Check in the images of your link the cpu is : MA803AS2   they did not buff it  loll

http://www.datasheetcafe.com/ma803as2-datasheet-pdf-microprocessor/

[cozza]

Check dmx polarity is correct then replace a transceiver and see if that fixes it. Tiny expense vs a big time sink.

[Factotum]

Interestingly while I wait for replacement transceivers I had a further poke around.

Turns out the driver chips further upstream from the MCU are TM1812 LED drivers.

The Arduino running this whole project using the FastLED library can drive these chips directly with much faster refresh rates (and would mean I could drop the whole DMX shield / step altogether).

So if I can't fix the DMX stage I can mod it to run directly...

[Factotum]

Replaced transceiver with no change

Still miffed at how the VCC transient damaged things given I can't see a path to the MCU.

Anyhow I did the mod to drive the TM1812's directly and it's refresh rate is at least 4 times faster!

If I ever get another driver I'll do an A/B compare to see if it sheds any light on the point of failure...

Test of unit

Thanks everyone.