Control/Dimming DMX issues, 4D cinema permanent installation - First Post!

[adamalltogether]

Hi All,

First post here but I've been doing much reading over the past couple of weeks to try and learn more about DMX and lighting control.
I'm in charge of an facility containing a "4D Cinema" (3D visuals. lighting and environmental effects) rides and other attractions.

The lighting side of my theater is becoming a bit of nightmare, and with my background (mech. engineer, Canadian - no big words, eh?) I'm a bit stumped by the whole DMX side of things...

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

http://youtu.be/cco8q7stJVE

Normal operation during a show is LED PARS off, with moving lights, spots and effects timed with the 3D visuals. LED PARS are used in conjunction with the Source4's for a pre-show, while the guests are entering and exiting.

This YT video demonstrates one of the recurring issues we have been facing. The installation is quite new (< 5 months), designed and spec'd by a German company in conjunction with a US, west coast based co. and installed by local contractors. Not much help in terms of on-site support, so myself and my tech have been working overtime to learn as much as we can to solve this.

Here's my DMX chain, as best as I can determine from the documentation and the brief training supplied:
Show Control computer (running visuals --> Alcorn McBride Lightcue (reading SMPTE) --> 3 Pathway eDIN splitters (using thru ports: 1-->2 for environmental effects, #3 for Lighting) and a Smartpack dimmer. The LED Pars are Chauvet Par38's (x37, addressed in pairs on DMX), using 3-pin XLR connectors (not sure if the cable is DMX or audio at the moment), moving lights are Source4 (x4), with a couple of spots and a strobe. Environmental effects in the chain are fans (for wind) (x4) and Snowmaster snow machines(x4).

Until this week, the "disco mode" shown in the YT link would pop up and disappear at random. Power cycling the entire system (racks and show control) would solve the issue for a time, then it would manifest again, not always as bad as shown in the link. Then, nothing - everything operating fine for weeks at a time and only going full disco when the site is jammed with people and shutting down causes many complaints from guests.

Now our Alcorn has stopped responding and the lights are all down. I'm working with the show company on the west-coast of the US to solve this issue at the moment.

My thoughts are:
- the LED PARS seem to be on the cheap side, based on my research. The theater can get quite humid due to the rain effect (nozzles in the ceiling providing mist) and the cheapo lights might not be suited to this environment
- the cabling between lights seems suspect, not sure if it's proper DMX spec.
- The splitters we have contain a termination switch, 2nd and 3rd are on. The last LED fixture in the chain (with the highest DMX address) is not terminated at the fixture, I've made a couple to see if it helps.
- all of our effects and lights are supposed to be on the same universe, or so I am told by the people who managed the install. I am doubtful of this claim though, as it seems based on our splitter set up that the effects (on 1 and 2) are still functioning as normal and are not affected by the Alcorn Lightcue malfunction.
- The Alcorn unit dumping it's brain may also have caused some of our issues, as it may have been on the outs for sometime or some of the cues are corrupt?
- this is a 364 day a year operation, so our small team of techs and myself don't have the time or energy to devote whole days tearing everything down, but it looks like we might be in for a week or two of all nighters...

From my reading here there are seem to be a lot of knowledgeable people, and many of the DMX related issues also seem to be in line with some of the issues we are facing.
Any questions or thoughts would be greatly appreciated, as we're trying to learn on the fly!

Thanks for reading!!

(TL;DR - 4D cinema, DMX issues out the wazoo, Tech Manager learning as he goes and possibly in over his head.)

Cheers,
Adam

 

[David Ashton]

OK, first response,dmx chips have inputs and output on them, it is quite common for a can to start outputting dmx randomly or as planned in slave mode.so look for groups that are flashing together, dmx splitting the groups down should eliminate various possibilities
second thought, if the dmx plugs have the case connected to pin 1 you can end up with earth loop interference similar to sound system problems.
You absolutely need to buy a dmx tester without one you will be guessing in the dark, this will allow you to check dmx levels and test the cables at working frequency, not at dc like ordinary cable testers
Many cheap led 64s have fans which draw in moisture, outdoor fittings are better, they have no fans but are heavier using fins to cool, even indoors I use outdoor fittings

 

[adamalltogether]

Thanks David,

We have a DMX tester on order, and our LED fixtures do state in the literature "not for outdoor use, do not expose to moisture" ... Too bad the people doing the install didn't read that bit! Ours definitely have fans too.

One thing we did suspect and are investigating is the cabling, as we thought perhaps the moisture had oxidized the pins some what and was causing issues.

Regarding the "garbage" DMX output, is this a common problem across all DMX fixtures or more towards the lower end of the cost spectrum?

We will check the cabling runs and the pins/ connectors when we're up the ladder next.

Cheers!

 

[Rob]

Adam,

Cabling is around 90% of all problems. Check that 3-pin cable. It should be labelled as suitable for EIA485 (or RS485). It may be necessary to take manufacturer information off the jacket to track this down. If the cable is not rated for EIA485, you may have all kinds of signal degradation issues. Oxidized pins can cause problems, but usually those problems show up toward the end of the DMX run, not all over.

Then make sure each line is terminated correctly. Then make sure you don't have fixtures with auto-terminators or other silliness like permanent terminators. I don't think the Chauvets do that, but check the manual. Over-terminating a line is as bad as no terminating.

My second thought involves how the eDIN splitters are wired. (Full disclosure - I work for Pathway.) You have some doubt about whether or not you're running a single universe. Are all three cards daisy-chained - i.e. one line of DMX goes into the first eDIN card, the DMX THRU is wired to the IN on the 2nd card, then the DMX THRU is wired to the IN on the third? If this is true, then there is a DMX merger in your system somewhere (maybe the Alcorn device acts as one). you will then need to determine the merging logic used by the merger. Also, if the wiring description is true, then the DMX terminate on the 2nd card should be turned off.

However, if card number 3 has a separate DMX line landing on its DMX IN, then the termination is correct, and you have two completely independent DMX universes in operation, one from the Show Controller and the second from the Alcorn McBride.

Your video is a bit of a classic. What's happening is the Chauvet lights are losing track of where the DMX data packet begins. They then begin to guess. And strange things begin to occur - in your case disco. Powering down your system will fix the problem - for a while.

The typical answer to this problem is to fix the cabling issues causing it. You need to confirm your cabling is acceptable or the next step may not make any difference - or you may simply exhibit a different problem.

The second step is to change the DMX timings - usually to slow them down. Some Chauvet fixtures are known to misbehave when fed fast DMX - typically faster than 40 frames per second. Slowing down to 32 frames per second will fix them more often then not. To be fair to Chauvet, as a receiver, you're expected to accept and understand an enormously wide variety of DMX timings in order to be truly compliant with the standard. This is not a simple or trivial thing to do. The Alcorn McBride device, on the other hand, needs to produce only one set of timings to be considered compliant, even if those timings are on the edge of acceptable. (I have no idea what timings they use.) A tester will give you this information.

A relatively simple way of checking if you have DMX timing problems is to get another DMX controller (or tester) (preferably a reputable one that allows timing changes) and use it. If the problem remains, it's cable. If it doesn't, then you might still have cable issues, but you can try different timings and hope you never see problems again.

Mr Ashton mentions common mode voltage problems. Yes that's possible, but the two issues above are more likely (Note - I consider shorting of Pin one to the XLR shell a cabling issue), and should be investigated first.

I hope this helps. Please feel free to contact me directly if you have more questions.

Robert.

Robert Armstrong
Manager - Technical Sales and Product Support
Pathway Connectivity

 

[FMEng]

Contact oxidation is an inevitable cancer. In a moist environment like that, I would clean all connections with Caig DeoxIT D-Series and then apply contact preservative Caig Shield S-Series. The stuff works near miracles. I have solved or prevented many gremlins over the years with it. D-Series should be in every electronics tool box and I am not exaggerating.

I also suggest that if your facility is truly a 364 day a year operation, it needs to be fully redundant on the control side. There should be two control systems with one a hot standby at all times. Each one should be fed by its own double conversion UPS. At some interval, like every week or every month, swap to the other system as the main. To do any less guarantees major stress for you and lost revenue for the boss.

 

[adamalltogether]

Hi all, an update on our situation here:

Firstly, the DMX tester recommended by one of our contacts is this one:
DMX 512 Tester Analyzer Preset Scene Store | ELM Video

We're expecting it to arrive soon, so hopefully by end of week we can start ringing things out.

We did investigate the cable further. Each light is connected to the other with... microphone cable (one-pair and shield) :facepalm:
This would seem to indicate that the problem is going to require a lot of re-cabling to work properly and solve this long term. Working to determine who performed the install, but as usual finger pointing is the order of the day.

Regarding our Pathway splitters (thanks for the clear reply Robert) we definitely have two separate universes, one from the show control for the environmental effects, and the other from the Alcorn unit (which is being sent back). The effects are running on two of the splitters via the DMX thru - the second is terminated on the board correctly. The third splitter has the termination active, as it should, and is being fed the Alcorn's output.

I am working with the powers that be to address both the cabling and redundancy issues - it is something that will be a tough fight, as the game around here is build as cheap as you can, let the tech manager figure out the rest.

Thanks for all of the replies, very helpful in lessening the learning curve! Will update this thread with any new information or issues that pop up once we are testing the system.

Cheers,
Adam

 

[Amiers]

That sucks with the Mic cables being installed. The cheapest way to recable is buy the cable in bulk and connectors separate. We bought bulk AccuCable 3pin Cable and the ends to go with them because like your situation the last person that was here figured it had 3 pins it must be DMX. I have a a rubbermade tub full if them needless to say I am not short on MIC cable lol.