Control/Dimming LED DMX Anomolies HELP PLEASE

[kendal69]

I need some LED / DMX help.

On several occasions I have had strange LED anomalies and was wondering if others have the same and maybe some answers.

I string a dozen LED Paps ( a no name Chinese brand ) - and it had happened with Mega bars also - anyway I string them out zero out all the setting RGB to 0000, set them all to the same DMX channel connect DMX in and DMX out connect to the board and start to adjust and the lights are all working fine until they hit lets say the 7th or 8th light and from then on they can not be controlled. Cables signal tests OK all the way to the end. Terminator on the end. Now I switch out the working 7th light for the non working 8th light and now the working light stops working and non working light starts working. I switch out to bran new cables no fix. I switch out lights till the cows come home and can't seen to get a signal pat the 7th light.

I spent huge money on a Doug Fleenor splitter thinking this would clean up the signal - nothing.

These anomalies continue on with different boards, different cables, different lights.

The length of the runs are only about 200 ft. Like i said I was blaming the Chinese lights but it has happened with the Mega bars today.

Seems LED lights at the tail end of the string just will not cooperate.

Is this just an LED thing or what??

Man will I be grateful if anyone has an answer.

 

[JD]

DMX should be able to chain up to 32 devices.

Any chance these have built in terminators that are switched on?

On the splitter, have you tried running units 1 - 4 on output #1, and 5 - 8 on output #2?

Where is the splitter located?

With a two output splitter, you should be able to run 64 devices post split (32 on each leg) as well as 30 before the split. In other words, something is wrong. There is some kind of signal degradation.

Do all these fixtures share a common ground off the same source?

 

[MNicolai]

It may be an isolated problem, but last week I ordered a DMX terminator from a usually-reputable source that ended up just being a 5-pin XLR connector with a heat shrink label "DMX Terminator" on it. (Better yet, when sent a photo of the "terminator," the company claimed they don't use those labels and the product wasn't made by them, even though it shipped directly from their factory with additional purchases from them.)

Maybe twist the end off and see if there's actually a resistor soldered on the pins.

Also, are you using 3-pin mic cables or 5-pin DMX cables for data?

 

[kendal69]

Hi John, Thanks for the quick reply.
Interesting, had not thought of a built in terminators that are switched on - is this a software thing with the manufacturer? There is no user control for this. The LED light had (9) channels, nothing for termination.

The splitter was put at the console to try and boost power down the entire line, and that was useless. Now you other idea was absolutely BRILLIANT putting the splitter 1/2 way, interesting.

No common ground, various outlets, well unless you consider the building as a common ground - then yes, but this has happened in three separate locations.

MNicolaihttp://www.controlbooth.com/forums/members/mnicolai.html

Hi MNicolai.
I'm converting from 5 pin DMX from the board to three pin DMX to the light. What's weird is everything works PERFECT until it hits a point then it will not go any further.

The DMX test good with a DMX tester before and after the "light DAM "

The weirdest is if I replace the working light with the next NON working light, the non working light begins ot work in the working light position and the working light STOPS working in the stopped light position - just one light apart.

Thanks guys

 

[MNicolai]

Built-in terminators are only found in a handful of devices. Chances are if you don't know that it exists, it doesn't.

Are they data-quality 3-pin cables or audio-quality mic cables? The difference is that DMX quality cables are low-capacitance while mic cables are higher capacitance and can degrade the signal.

If I were to guess, I'd say the reason you're having problems is that after x amount of connectors in the daisy chain, reflections caused by those connectors corrupt the signal by the time you get to the last fixture. It's also questionable as to what your signal tester is testing for. Does it test for the presence of a signal or for the quality of a signal that is present? (there's a difference, and it's very important. don't automatically assume that because it's a tester, it's testing for quality.) It could be testing purely for the presence of a signal, which means it won't tell you if you're getting a signal that's absolute garbage.

 

[kendal69]

Hi MNicolai.

Hear that sound? It's the light going on above my head. GREAT POINTS, yes it's a cheapie tester just showing a signal.

I also think ( because I didn't do the run ) it was a mixed batch of cables, so you just gave me something to ponder and actually set up again.

Any recommendation on cable and a great tester that gives me signal strength.

What are you thoughts on what John suggested, putting the splitter 1/2 way between the last light, giving it a small boost?

Either way I wish I could buy you guys a beer. REALLY!

 

[millamber]

This has to be a cable problem. You've eliminated the fixtures as the problem by swapping them around. I'd try to swap the cables around. Start with the cable between the last working fixture and the first non working one. Put that cable in the spot between the first two fixtures in line and see what you get. If it works, go to the next cable and repeat. It could be one of the cables at the end of the line reflecting back.

 

[JD]

DMX splitters don't really boost the signal, they simply duplicate the signal and provide isolation between different branches and the board. In this way, they can be though of as a repeater. If the signal is weak from running through a long cable, the splitter brings it back to 100%. If you have a long run from the booth, then the best place is at the stage end. If they are at the console end, they will still provide ground isolation, but that's about it.

When placed at the stage end, your cable footage count starts at 0 again. Also, does the splitter have a terminator? If it is designed for the DMX input to loop through pre-split, it may not, or may have a switchable one. Worth checking, although not related to the problem you are having.

As noted above, if there is not a visible switch, dip-switch, or setting for termination, then the fixture probably does not have one. Double and triple termination can cause a lot of problems as the signal line loads down, so worth checking.

Now, on to the cabling. This does sound like your problem. Plenty of people use mic cords and claim no problem. In all cases, they are causing a problem but they don't know it because the DMX is not trashed enough to effect operation. The cable MUST be DMX cable as mic cable operates at a different impedance. The result of mixing cable is that the signal echos at every point where there is a change of impedance. This makes the signal hard to decode. (Think of someone shouting in a gymnasium.)

 

[kendal69]

Millamber,

Yep did the cable switch with brand spanking new cable before and after the troubled light and that did nothing, light still worked to the light but not beyond.

Cable is Hoya, but thinking I need to get some Gepo?

JD and Millamber,

You guys are making me perhaps wonder if the cable coming from the light may be the issue. The Chinese light has about two feet o DMX in and two feet of DMX cable out and that stuff look real lame. So with what you guys are thinking may hit their junk cable on the units and react. In other words as the signal goes from light to light hitting their cable the signal degrades by the time it hits light 8.

The only caveat was when I was having the same problem the video guys were running a video projector and my lights were going nuts, THEN their video projector took a dump and the second they went out my lights started working?

Why don't these boards have a BUY THIS GUY A BEER BUTTON? You can trust if I ever meet you gust the beers are on me all night.

 

[mstaylor]

Try doing the same length cable run to the point where it stops working. See if the distance is doing it or the in and outs of the lights. In case that wasn't very clear, eliminate the 1 through 6 lights and make seven the first light but at the same distance of cable run as before. I would even try both longer cables to make the run, then shorter pieces to simulate having the first six lights.
As an aside, don't what effect the projector may ahve had without knowing the set-up but just shoot the video guys on general principles.

 

[millamber]

You say it happens with the mega bars as well? Are they on the same DMX run? And if they are, are they also addressed the same dmx channel?

Regarding the DMX cables on the fixtures; have you tried eliminating the interim cable altogether and just plug the out of one light into the in of the next? At least in testing this will tell you if you have a bad third party cable somewhere.

Try to pare it down to the absolute essentials; console/fixture/cable. Start with one fixture on the console, then add a second, and so on until it stops working. Distance may come into play, depending on the cable, so if you can plug one fixture into the next you can eliminate that problem as well.

 

[JD]

One day, just for fun, I patched together about 1000 feet of DMX to see what would happen and it still ran fine. (Not recommended.) I don't think length is the issue. Outside of a ground loop issue, the video should not have had any effect. You may have a bunch of bad cable, or one cable with a pin 1 to 2 (or 3) short that's killing the strength of the DMX. In many cases, DMX will still run with such a short, by everything becomes really iffy. (Which is exactly what you are describing.)

Also, any of those cables have pin 1 and the case (shell) jumpered? Works fine as mic cable but is a big no-no on DMX.

 

[kendal69]

Mstaylor,

Man it was funny because the second their Video dumped my lights started working and there was only 30 minutes before doors opened and I stopped sweating and they all started grabbing ladders and sweating.

It was a better them than me situation.

Anyway holy molie is there some talent on this site or WHAT? That trick of reeling out the cable 200ft and trying the light is VERY creative.

You guys are doing you best to make me feel more and more like a Moron, but that's good I needed that because I was at my wits end.

I just pulled a light and checked out the two ft, in and out cable and it's some real lame cable I mean real lame cable. If that's it it would be cheaper to buy a lot of 2' pieces than hundreds of feet of new DMX.

 

[MNicolai]

Mstaylor,

Man it was funny because the second their Video dumped my lights started working and there was only 30 minutes before doors opened and I stopped sweating and they all started grabbing ladders and sweating.

It was a better them than me situation.

Anyway holy molie is there some talent on this site or WHAT? That trick of reeling out the cable 200ft and trying the light is VERY creative.

You guys are doing you best to make me feel more and more like a Moron, but that's good I needed that because I was at my wits end.

I just pulled a light and checked out the two ft, in and out cable and it's some real lame cable I mean real lame cable. If that's it it would be cheaper to buy a lot of 2' pieces than hundreds of feet of new DMX.

Actually, unless you make your own DMX cables, you get more for your money by purchasing 25' and 50' cables than 10' and 5' cables. The cable is cheap, while the connectors and soldering aren't, so if it comes to purchasing brand spanking new DMX cables, you'd be surprised how relatively minimal the price difference is in purchasing a dozen 10' cables versus a dozen 25' cables. While I'm not a fan of having lots of extra cable in the air, it's hard not to go for the 25's when the 5's and 10's are still as expensive as they are.

 

[jfleenor]

Hello there!

Just have a few questions for you:

1. Do each of these LED fixtures have multiple LEDs in each fixture? It may be possible that you have a separate receiver for each LED, which would surpass the 32 load limit per driver rather quickly. Typically, each fixture presents one load, but poorly designed fixtures can present more than one load to the driver.

2. The 200' cable run that you speak of. Is that from your driver to the first fixture, or is that between each fixture?

3. Do the symptoms change at all if you remove the terminator? If so, what change is there? We have seen some terminators that have 12 Ohm resistors, instead of the required 120 Ohm ones.

4. Do you have any photos of the fixtures, or a sketch of the setup? That may help us in troubleshooting as well.

Hope we can help you out!

 

[MNicolai]

FYI, that's Janell Fleenor. Fleenor as in Doug Fleenor Design, aka "Dr. DMX."

If you thought the other advice you were getting here was helpful, we're mere peons in comparison to the DMX knowledge-base that is the Fleenor empire.

 

[kendal69]

Link to pics showing the light and some long runs in ballrooms. Big room was about 200'x200', the other was 100'x100'

Site Title

Lights have 52 - 3 watt LED Green Red Blue and White. The typical 200' + run would be cumulatibe total run with 24 LED's seprated by about 10', but using 20' DMX cables.

Terminator actually helped, lights were flashing and going wacky then I terminated and they calmed down. When I say calmed down they stopped acting wacky but I had the lights working to light 7 or 8 and then no control from 7 or 8 on.

Set up was from board at one end of the run and terminator at the far end.

You can see in the photos that these lights are AMAZING and powerful being 3 watts for each LED. Now if i can just control them all at the same time.

millamber
On the Mega bar it was a whole different day and useage that's what made be go nuts and wonder is everyone else have these weird DMX issues with LED's.

The Mega Bars were connected end to end with a DMX cable and also power IN and power OUT from each bar, and connected to one outlet. All lights fine - last one was going bonkers. No control, controlled from the AMSTER light which drove all the others.

1:30 am I'm hitting the sack, man I'm beat today, if some of this doesn't make sense I'll rewrite in in the morning after some coffee.

OH yes and thanks all you guys a great, if only the entire world worked like this site.

 

[WooferHound]

Eliminate the extra cables altogether and just plug the lights into each other using the attached DMX Tails that are perminatly attached to the fixture.

What are you using for a controller ?

 

[mstaylor]

The troubleshooting skills on this board are pretty dang good but they are because we have all stood there going,"Well, that didn't work, what else can we try?" Don't feel like a dork bacause it is easier to troubleshoot when you aren't looking at the clock and trying figuring how close to doors you are.

 

[kendal69]

Wooferhound.

I've used several boards from a cheepie all the way up to a Road Hog Full Boar.

When I attached the Full boar I though this baby will blow out any "kinks" in the line, but nope. That was the ovard that was hooked up when the video went out and then lights started working. I'm pretty sure the two aren't related but threw as much info to you guys as I could.

I'm sitting her laughing because you guys all remind me of a HOUSE episode but for lighting guys and DMX.

You know that's funny.

Ms taylor.
Dude I know my hair turned gray that night. It was a huge client and I was this close to going to them with my tail between my legs and throwing in the towel.

I plugged un plugged switched lights switched cables laid new runs for 4 hours on a simple 16 up-light job. You know you drop 16 lights on the ground plug them in and walk away. One hour MAX with a lunch break.

I'm not going to leave everyone hanging, the secong I get a change to duplicate my problem I have several things you guys suggested.

Hey I was thinking of getting that DMX cable tester form Doug Fleenor RAD ZIPI- thoughts? A lot of money but.......if it's what will find the gremlin.
Will it do any good for testing this problem?