Faulty lighting desk

R

RichieNorthcott

Member
I have one of these cheap lighting desks, a few years old now:
Screenshot 2023-08-24 at 20.56.05.png

Lastnight when I powered up my garden lights, all the lights flickered (like they were getting bogus DMX data). Today I get nothing. I have plugged a moving head unit directly into the desk, it doesn't recognise the DMX signal. As a sanity check I plugged my garden DMX universe into my USB dongle, and ran the Vixen show. Everything is good. Equally I can also control the lights with Q light controller which is a software lighting desk.

So.... there's something funny about the DMX data coming from the desk. I've checked it in a 2 channel oscilloscope, everything looks very normal (the differential transceiver is doing what it should). The DMX framing and timing looks correct. Each bit of data is 4ms, I see the bits changing correctly when I adjust channel values.

I am totally out of ideas, other than toss it in the trash and buy another one.

Anyone?
Thanks!
 
Put a serial receiver on the output and look at the data chain. I also use a logic analyzer to find the start of frame and then look at the data. The DMX frame includes some command information that must be set properly or the receiving devices dont know how to handle the data.
 
Can you see the impact of the terminator on the logic analyzer? The challenge I get is, the intermittent ringing doesn't seem to register on a stream of 1's and 0's, unless I get very lucky. I have a problem right now where a drastic change in the load introduces noise into the data signal,.and it shouldn't. I have a 12-channel, where if I just go straight 0->255, whoosh, on all 12, then that screws it all up. But if I slowly undim one channel at a time, all is well. So I know the problem--but I can't actually see the noise on any measuring device, and I'd like to.
 
Thanks - yes my thoughts are something has gone screwy with the timing, though it's not jumping out at me from the scope traces. I have a fairly basic 2-channel digital oscilloscope so there'a a limit to what I can do with it. I can detect the start of the frame based on the preceding pulse. TBH I probably can't easily fix the underlying issue anyway, these desks are only about 35GBP on Amazon.

I don't think it's the cause of the ifault - but I'm thinking about including some Isolation circuitry as there's lots that can and does go wrong with the garden installation due to water getting in places it shouldn't. I run a simple splitter and 8 channel decoders (none of it is 'pro' gear), I know none of these have isolation. I also have 4 moving head units, they may/may not, I'm not sure.
 
Did you sanity check whether a simpler RGB flood responds, and not through wireless? You switched out the board for a dongle on the same fixtures, but I didn't see whether you switched the fixtures and removed the wireless, but left the board the same. Another sanity check: did you probe the oscilloscope only right at the output? Or also at the fixture input?

(""sanity check" in my day job is jargon-slang for, "just a check to make sure I'm not going insane")
 
Well it turns out i missed the obvious. There is a DMX polarity switch on the reverse of the desk by the connector. How and why this slider switch got moved in the garden outbuilding I have no idea, it's build into a panel and the back cannot be accessed by anyone (without a scredriiver to hand). Anyway, it is what it is..

1 slide of the switch later and we're back in business. Best return my new purchase :/

Re above comment, I never use wireless since I've had nothing bad bad experiences with it. I used the same cables / fixtures when testing with Vixen DMX output, so the issue had to be with the desk.
 
Thinking about this the more likely cause it corroded contacts on the polarity reversal switch, considering the unit is kept outdoors all year (well, in a summerhouse). Probably some contact cleaner would be a good idea. The main PCB is in good condition having opened it up.
 
I was going to say, not all signal integrity problems are caused by the transmission line. Sometimes the output is logically correct (and thus identical to another, functioning device) but inadequate drive strength. And if you probe both at the output and right at the subsequent input, and see a difference--and a difference when you switch what device is producing the DMX out (or any output, for that matter), then you've isolated the problem. It's a decent sanity check in general. But you said you had solved the problem, so I just lurked.
 
You must log in or register to reply here.
Back
Top