DMX flicker

[edinc90]

I'm having an issue with flickering LED fixtures. I get flickering and strobing on one type of fixture in particular (Elation ELED Fresnel 150) whenever I'm sending lots of DMX data to another fixture, like animating a mover or flashing RGB LEDs. I thought it was a termination issue so I added a terminator, but it didn't work. So then I started pulling cables while the board (ETC Element) sent a move effect to one of our Design Spot 250 Pros, until I got down to just one fresnel at the beginning of my chain, with nothing else. It was still flickering! I have no other 5 pin DMX cable to test it with, otherwise I would switch it out, but the first fixture is only about 30 cable feet away from the board.

The interesting thing is that I have 8 of these fresnels in the universe, but only the 4 at the beginning of the loop exhibit this behavior. None of the other fixtures, which include Elation Tri White Pars, TVL2000, and DLED RGBAWs, exhibit this behavior. This also happened with another board, the Elation Show Designer 2, but I chalked it up to a board issue. Since it's happening with the Element, too, I think something else is going on.

 

[len]

Can you adjust the data refresh rate?

 

[derekleffew]

... The interesting thing is that I have 8 of these fresnels in the universe, but only the 4 at the beginning of the loop exhibit this behavior. None of the other fixtures, which include Elation Tri White Pars, TVL2000, and DLED RGBAWs, exhibit this behavior. ...

Symptoms (and likely solution) sound remarkably similar to this thread: http://www.controlbooth.com/forums/...ue-ive-encountered-my-8-years-experience.html , or this one: http://www.controlbooth.com/forums/...-dled-elation-rgb-108-pars-ghost-machine.html .

Can you post the entire patch sheet for that universe? You might also try re-addressing the offending LED Fresnels either higher or lower (probably lower). Or, easier, just as a test, address the offending ones to the same addresses as the non-offending ones, and see if the problem persists.

I'm fairly certain the location of the fixtures in the chain matters not. Are there any opto-splitters involved?
.

 

[TheaterEd]

I see that in 2 channel mode one the fixture will strobe, I'm sure you've checked but just throwing that out there to make sure it's in the right mode.

 

[edinc90]

Here's the patch list. I have them in as dimmers, since I am running them in 1 channel mode and there's no fixture preset. I'll try to change the address and see what happens. No opto-splitters or anything else.

 

[edinc90]

Changing the address of the fresnels to match the ones that aren't misbehaving fixed the symptoms. Then I went further, and tried other addresses. It seems once the address is above the address of the movers, then the flickering starts.

I tested this by fading some fixtures that are addressed before the fresnels that were behaving, and sure enough they started flickering!

I changed the output speed from Maximum (the default) to Slow, and I have the same issue.

 

[derekleffew]

Changing the address of the fresnels to match the ones that aren't misbehaving fixed the symptoms. ...

Okay then. Some re-addressing of several fixtures, or perhaps redoing the entire universe, and you're golden. I'll leave it up to @DavidNorth, @MiltonDavis, @eloader, @ELASERV to explain why the fixtures exhibit this unusual behavior. I'm not sure I've ever heard of the manipulation of higher DMX channels affecting lower ones.
.

 

[edinc90]

If only it were that easy. It seems any big move, like fading up more than one light, causes any Elation fresnel to flicker if it is addressed above the fixtures being changed. I read something on here about the DMX Decelerator, and how it could solve issues like this. Does anyone think it's worth a shot?

 

[JD]

I'm not sure I've ever heard of the manipulation of higher DMX channels affecting lower ones.

It shouldn't, but if something is out of spec..... well, the higher the number, the further that byte is away from the sync bit.

If only it were that easy. It seems any big move, like fading up more than one light, causes any Elation fresnel to flicker if it is addressed above the fixtures being changed. I read something on here about the DMX Decelerator, and how it could solve issues like this. Does anyone think it's worth a shot?

The DFD decelerator may be the answer as it smooths out the timing and takes care of some common error spots. Still, it is an additional expense.

 

[DavidNorth]

If only it were that easy. It seems any big move, like fading up more than one light, causes any Elation fresnel to flicker if it is addressed above the fixtures being changed. I read something on here about the DMX Decelerator, and how it could solve issues like this. Does anyone think it's worth a shot?

Can you check to see what happens if you only have one fresnel on the DMX line? I am curious of changing the addresses still shows the same problem or if the nature of the problem is now different.

If the same porblem occurs at higher DMX adresses no matter how many or how few fixtures are on the DMX link, then this is an internal software problem in the Elation fresnels.

This is unusual.

David

 

[edinc90]

I have tried this. What I did was send an effect to a mover that was addressed below the fresnel (fresnel was 214 and the mover was 65). Even with the fresnel being the only fixture on the entire DMX chain, it still flickered.

Another interesting note, the fresnels flicker whenever someone is standing near them and uses their walkie talkie (Motorola CP200.) None of the other lights do this.

 

[DavidNorth]

As long as you are using quality DMX cable, and not mic cable, then this is a fixture issue. Are you using all 3 pin cable? Did you make adapters? Have you checked to make sure shield is not touching case in the connectors or that shield and data are not swapped?

You could try to put the fresnel on the exact same power feed as the cosnole [share the outlet] to see if the issue goes away. If it solves it, then if you can run that way, good, but it would mean that there is either a miswired or bad DMX cable, poorly designed fixture electronics or the DMX wiring in the fixture was assembled wrong. Having an identical ground reference might place the cosole and fixture within the same DMX common reference based on the above [or data signal reference for D+ or D- is miswired]. This is obviously a longshot and is only meant to expose the possible issue.

With susceptibility to RF from a radio, again, it could be a shield/data wire swap somewhere as mentioned above, or a fixture design weakness.

Elation has been known to say that their fxiures may not work with DMX from other manufacturers consoles but are certainly designed to work with theirs, which you have proven does not work. I think a quick call to them might be in order here as the next step to see if they know something that we don't in regards to the fresnel. I've looked through the manual and checked pricing. These are not inexpensive fixtures and you have set the fixtures in the correct mode.

I hope this helps. Perhaps someone has a different thought?

David

 

[JD]

Elation has been known to say that their fxiures may not work with DMX from other manufacturers consoles but are certainly designed to work with theirs, which you have proven does not work.

Then can it really be called "DMX?"

Have to say, I have a bunch of Elation stuff and it works fine, but it really does sound like a fixture issue. To me, as long as the cables are good and there are no massive ground loops, then what we have is a DMX transceiver chip that is hearing things outside of the normal DMX bandwidth. (Such as the walkie talkie!) RF noise is just a part of life. This has to be taken into account when a circuit is designed. Since the DMX signal is fairly low in frequency when compared to something like gigahertz Ethernet, there is no reason for a DMX chip to be listening to those frequencies! It would be interesting to take a good quality opto-splitter and patch it's output to a fixture in question with a 3 foot DMX cable and properly terminate it (in other words, the "perfect" clean signal.) If the problem disappeared, it would be a good indicator that the fixture is a bit too sensitive to noise outside the DMX bandwidth.

What may be serving as the "antenna" for this noise is not limited to the DMX cable alone. Noise in the power, or picked up on the ground line may be entering the fixture as well. Again, the DMX receiver in the fixture needs to be designed to reject a reasonable amount of noise.

EDIT: Here's a "poke in the dark"- Ever see those lumps molded into computer monitor cables? They are basically a small toroid with the wire looping through it once or twice. The purpose is to stop RF from being picked up by the cable, or transmitted by it. If you can rip a small toroid out of an old power supply or surge suppressor, you may want to try making up the DMX equivalent. (Simply a short DMX cable that loops through the core once or twice.)

One last thought- On the walkie talkie issue, will the fixture do it if both DMX cables are removed? How about with no DMX cables but a terminator plugged into the fixture? May give you an idea as to how oversensitive it is.

 

[edinc90]

JD, you're on the right track. Elation is talking to their overseas manufacturer and they think they will need to ship me some new ICs. The batch we have may have a defect. The communication has been slow, and my director is getting impatient.