Patching problem?

[HPCTechnical]

We're having issues with our desk, in that various of our fixtures stop responding intermittently. In the rig, we have 4 Chauvet Intimidator 250 spots, and 2 Showtec Phantom 50s. So far, we've had 1 of the Chauvets and both of the Showtecs stop responding. They're both receiving good data: when we then set them to the same address as the neighbouring fixture, they behave perfectly. It appears to us to be an issue with the desk's patching: chaing the DMX address by as little as 1 seems to fix the problem. But the problem has reoccurred this morning, and we can't afford to be getting the Genie out every morning before the first show of the day!

 

Any thoughts?

 

We're running 7.8.5 installed yesterday, but we have had this problem in earlier shows.

[Jon Hole]

Hi,

 

If this is only happening on some of the fixtures and not the whole rig, I would suggest we might need to be looking at the fixtures, rather than the console.

 

Have you got a DMX tester that allows you to see the channels changing in the DMX stream?

 

Alternatively, if a fixture is meant to be addressed to "150" for example, and stops working the following day, try changing the address of one of the working fixtures to 150, and see if that responds correctly.

 

Jon

[HPCTechnical]

Hi Jon, the difficulty is that it's not the same fixtures. One of the Showtecs which didn't respond yesterday is behaving perfectly today: its partner, which behaved perfectly yesterday failed today. Ditto with the Chauvet that failed today, we had no issue with it yesterday. And they fix immediately by changing the DMX.

 

It's worth mentioning that we also have some iSolutions Splendor 36 moving head LEDs, with which we havent had any issues so far.

 

Changing to the faulty DMX isn't something we've tried (d'oh!) but will do next time we encounter this. We're just running through a sequence of restarts to see if the problem reoccurs, and how often. As far as I'm aware, we haven't yet had an instance of a fixture ceasing to respond once it has been working properly, it seems to be at start-up

 

Afraid we don't have a DMX tester...

[HPCTechnical]

Current show.isf

 

Hi Jon, we set the DMX address of the Showtec unit which failed this morning back to the where it failed - and it failed again. We then reset its partner to the same address and it failed also.

 

I don't know if it's connected, but the desk did crash once while we were rebooting. I've attached a showfile and the debug report from that crash.

 

*EDIT* re the debug report, it's telling me "You aren't permitted to upload this kind of file"

??

 

[HPCTechnical]

Further update: the problem seems to be within a range of DMX channels from 207 to 210. I'll explain my fault-finding and see if you agree with me. Both the Chauvets and the Showtecs are in 13- channel mode. In any start address between 194 and 210, they misbehave, although the nature of the misbehaviour changes according to the start address. When we started from 193 or 211, all was fine. We could export that misbehaviour to every one of the spot fixtures by setting it to a start address within the range 194 - 210.

[Jon Hole]

Try addressing your dimmers to those DMX addresses. Then, when you increase the Pan / Tilt, you should be controlling dimmer channels instead.

 

That will tell you if it's the console or the fixture that's causing the problems.

[HPCTechnical]

Hi Jon, yep, have done that too, and the result was highly erratic, to say the least. Also to add that I took the output direct from the back of the desk to the Showtec fixtures, and the behaviour was the same, so I don't think it's a DMX distribution problem.

Any thoughts on why I'm not being allowed to upload the debug report?

[Jon Hole]

How do you mean by "erratic"? If the Pan / Tilt is a 16bit parameter, when moving either of these I'd expect one dimmer to flicker a lot, whilst the other raised and lowered. I'd also expect a lot of dimmers to be at random levels, matching the default values of all the parameters of the fixture patched.

Any thoughts on why I'm not being allowed to upload the debug report?

Drop it into an email to me - jonhole@eaton.com

[HPCTechnical]

How do you mean by "erratic"? If the Pan / Tilt is a 16bit parameter, when moving either of these I'd expect one dimmer to flicker a lot, whilst the other raised and lowered. I'd also expect a lot of dimmers to be at random levels, matching the default values of all the parameters of the fixture patched.

 

Hi Jon, there was indeed a lot of flickering but nothing that looked like a clean fade at all.

[Jon Hole]

Hi Jon, there was indeed a lot of flickering but nothing that looked like a clean fade at all.

Try patching just normal standard dimmers on these addresses, and see if that works with your dimmer addressed to 200? I'm not sure what is causing the issue. I'm willing to be proven wrong, but with what you've described so far I very much doubt it's the console .

[HPCTechnical]

Sorry, Jon, wasn't clear, that last post was indeed having patched a standard dimmer, and that was me attempting to describe the results.

 

I could be missing something, but if this summary is accurate-

 

the fault follows the addressing, and the fixtures only misbehave within a certain number range

 

Any of the fixtures can me made to misbehave by numbering them to that range

 

All of them behave fine outside of that range

 

 

- I can't see how that would be a fixture fault - particularly as we're talking fixtures from 2 different manufacturers?

[Jon Hole]

Hi Ian,

 

OK, let's chat through this over the phone... I've just tried to call you but got through to answerphone. Maybe you can give me a call when convienent?

[kgallen]

Hopefully a daft question, but do you have a DMX terminator plugged into the last fixture in the DMX chain? Didn't see this mentioned above but could have missed it...

[HPCTechnical]

Yep, terminators all round, apols for omitting that. Jon, think we have a viable work round now, and am in tech week, so unless it reoccurs will have to leave it until I can next draw breath - probably in early 2016. Will hope to be able to revisit this under less pressured circumstances. Thanks for your help and patience so far.