DMX challenge

[DJDC]

Kit 4 QTX (shhh!) 250w moving heads, QTX LED PAR 64's (Smart Pars) Martin Mania EFX 600's, Transcienson DMX operator 2

I found DMX easy until 2 weeks ago, the bulb on moving head 2 kept going off so I swapped the unit for a spare unit. However the replacement head 2 would not pick up a DMX signal, and I had to by pass it on DMX for the others to work. It would run in sound with the others receiving a signal from it. So I spent today testing the system I moved the replacement head to the end of the chain Head 4, and shuffled all the other heads along one position (2 - 1, 3 - 2, 4 -3, 2r - 4).

I even tried another spare unit in position 4 but it workedalso as slave but none still received a DMX signal. Leads are fine as the will all work daisy chained in - master - slave - reverse slave - reverse slave

So back to the replacement head it will work in slave mode with the rest, with head 1 acting as master. However connect head one to the DMX and they don't acknowledge a DMX signal any more. I thought the controller was dead, but last week it would control the PAR 64 LEDs, and I've been able to get it to control the EFX's. So I tried each head on it's own on DMX but none respond to a signal ( DMX signal flashes the blue DMX LED ) a head won't even acknowledge a signal off an EFX.

Someone one told me that DMX units do have a form of memory, and if you change a chain (and channel) units drop out as unit (channel) the responded from and sent to signal to has now changed?

So at present I'm a little stumped? Fortunately I can run them in auto or sound mode, and just run the pars and EFX's in DMX, although this does mean I have to run extra DMX leads up onto the overhead, as there is only one daisy chain link on it.

[Jiggles]

DMX doens't know whats on the line, if all 512 channels are used or anything like that. (Add RDM to the mix and the desk knows whats there but thats out of the scope of this forum!) sounds like a duff cable, duff head. try adding a DMX terminator to the end of the chain.

[Vectis]

1. Terminator? Present? Working?

2. Swap all the cables out one by one as there might be a duffer. May well test OK on a multimeter but impedance could be significantly out, distorting the signal. Also as it sounds like you have quite a long line so this means it's important to have "proper" DMX cables and not just audio/mic cables.

3. Least likely but has happened - DMX board in one or other fixture. Try taking one fixture at a time out of the line to see if the problem goes away.

[NKR]

1. Terminator? Present? Working?

2. Swap all the cables out one by one as there might be a duffer. May well test OK on a multimeter but impedance could be significantly out, distorting the signal. Also as it sounds like you have quite a long line so this means it's important to have "proper" DMX cables and not just audio/mic cables.

3. Least likely but has happened - DMX board in one or other fixture. Try taking one fixture at a time out of the line to see if the problem goes away.

This is what is known in the industry as a lot of :Censored: ing about to isolate the fault. DMX cables are sods for throwing out a whole chain when you get a slightly faulty one. Happened to me at a wedding fayre catwalk a few weeks ago. Whole signal dropped out and left me stranded with no control. At least the floods were on white when it dropped out.

[Happy Jack]

Some sound (light?) advice here. How have you patched the units via the controller? And there is output from the desk? I'd also check the XLR input chassis on the troublesome heads to ensure it links back out. If they operate stand-alone then the transceiver is good.

It doesn't sound like a cable issue - EIA485 is very forgiving at this scale - it sound like you may have repatched incorrectly. However, invest in a cable tester if you do not already have one just to ensure continuity and polarity.

http://cpc.farnell.com/tenma/72-8785/cable-tester-av-universal/dp/IN05926

I'd then be inclined to address all heads to channel 1, run a show or control manually to observe how they respond. If you are satisfied, readdress and if issues persist apply the 'split-half' technique with a XLR 'sniffer' to offer some feedback.

http://s262510961.e-shop.info/shop/catalog/details?shop_param=aid%3D1%26

[Jiggles]

It doesn't sound like a cable issue - EIA485 is very forgiving at this scale

Im my experience it is! If the cable aint perfect it will throw many hissyfits!

[DJDC]

Cheers chaps, 2 weeks ago I had swapped out the 2nd head with another unit as the original one had developed a problem with power to an internal board, thereplacement head would not assume the role of head 2, and threw out the rest, I bypassed it and everythign else worked ok.

when I set the rig up yesterday I used another unit in its place but none worked to DMX, but would work either slave or master with it, I swapped it back to the other replacement unit putting it at the end of the chain and they

the bar is DMX pre wired for 5 effects (as per photo) by passing the efx600 the heads will work fine in Master (left unit) and 3 slaves.

run head one from DMX and none detect dmx SIGNAL, I have tried plugging the controller into each head individually with no chain, and they won't pick up a signal, the EFX though would.

I have terminators but never needed them for this particular rig, tried them but it made no difference. When I used the rig last weekend I was not DMXing the EFX, just up to the 4 heads then down to 2 PAR 64 LEDS's, but cannot remember whether they received the signal, but certainly connected without the 4 heads they worked fine.

I believe it's either down to the spare units been dodgy (DMX boards) and killing the signal or a now dodgy DMX desk?

It appears that at present none of the heads will work direct to desk on their own.

My mates should have 3 spare in his unit, so hopefully one of these will work? my logic will be put all back into original sequence and add a new head 2. The only problem I have is the controller needs reprogramming as I have altered the overhead so I could turn the heads round 90 degrees so the stop point in the movement is towards the back of the stage and not left or right increasing the time they light the dance floor, and not the side walls.

More playing tomorrow night I think

[DJDC]

image of rig, please no comments on cables they are only like this to get the lights working, they are much tidier when in use!

[katman]

Looks nicely balanced.

[Happy Jack]

Im my experience it is! If the cable aint perfect it may throw many hissyfits!
While I don't wish to suggest anything other than best practice, EIA485 is a NRZ protocol, meaning the transmission line model isn't valid until the cable run is about 160m in length. This is why many, at this scale, can neglect to terminate a line and use less than 'perfect' mic cable without consequence. That isn't a certainty, and isn't something I'd advise!


...the replacement head would not assume the role of head 2, and threw out the rest...

I'll echo what Callum has said above. The controller does not differentiate; either there is a value associated with a particular channel or not. Whether one particular head is unable to process this signal shouldn't affect the link through to others, unless, of course, that link is intermittent or severed completely. Open up the troublesome units, inspect the XLR chassis input / outputs and prove continuity; they should wired in parallel.

I cannot find much in the way of literature on the units, even via the AVSL website. Without knowing a great deal about them, I'd ensure that all units are engaged in the correct operating mode via the menu if that is applicable, and not in stand-alone / master slave.


It appears that at present none of the heads will work direct to desk on their own.

And with that statement, at this stage I would try to eliminate your controller as the source of the issue. Do you have another means of control to rule that out? If you do suspect your unit, as a last resort, I'd consider a hard reset / clear all channels and scenes, repatch and then start again.

[DJDC]

Right tried each head on channel 1 each on its own nothing, with or without terminator. tried every combination, head 1, 1-2, 1-2-3, 1-2-3-4, 2-3-4, 3-4, 4 (all on channel 1 or otherwise)
Plug the DMX controller into the first head it will continue to work on auto, but the slave heads stop responding until I unplug the controller.

EFX and PAR 64's will work though.

With these units if they don't detect a DMX signal (blue led flashes if any do) they revert to none addressed mode i.e N Slave, N Auto, N Sound.

next option full reset of desk, then check internal XLR links. Cables seem fine as the units work in Master / slave mode (3 slave units blue LED's flash acknowledging signal).

After that run in sound /slave mode , until I can catch up with my mate. Thanks for your help, and the comment for it all looking balanced thanks, I dropped the 4 PAR 64 LEDS from the top due to the weight and allowing me to recude the rig with ease / make it easier to resolve any DMX ploblems

[Jiggles]

Hmm sounds like the desk is outputting the DMX at a speed that the heads cant cope with! Find another desk and see if the fault continues.

[DJDC]

it's strange as the desk used to work fine with them, think it's also under 12 months old, hopefully my mate has a few desks I can try when I can catch up with him. Thanks

[Happy Jack]

Hmm sounds like the desk is outputting the DMX at a speed that the heads cant cope with!
Have you ever known that to happen?


it's strange as the desk used to work fine with them, think it's also under 12 months old, hopefully my mate has a few desks I can try when I can catch up with him. Thanks
The symptoms are curious, but I agree that you should source a known-working controller, repatch and test again so you can at least rule that out.

Regarding the QTX units, I'll hazard a guess that processing is centred around one of the PIC16F8 family of 8-bit MCU's. In fact, I'd be curious to know exactly what it is. That all units respond flawlessly in master / slave and not at all to an output from the desk is rather odd; the introduction of this new unit appears to have had a terminal effect now common across all units, and isn't something I've ever had to debug.

I'd get in touch with the technical department at AVSL, describe the symptoms to them and hope for an informed response. You'd really need to speak to someone who knows the units, or at least has encountered this issue before and is aware of a remedy else you'll soon tire of speculation and the generic stock responses.

Let us know how you get on.

[Happy Jack]

How did you get on Darren? Did you manage to find a remedy?