Studio Spot and DMX conflict

[mstaylor]

I am usually pretty good at troubleshooting but this one has me stumped. I loaded a show in yesterday and they are using four Studio Spot 575s, that is the only thing on that line of DMX. I was called in today to move a projector, I am the lift operator/rigger. While I was there the LD asked to go up and check two lights that weren't responding, they were both flashing a shutdown message. He asked me to check the connections on both lights, the last one working and the first not.
Now to the troubleshooting, I didn't like the way the jumper from two to three was so I reseated it and had the LD restrike three and four. They showed they had signal and the addresses were correct but the still wouldn't take commands. Also, once they homed the display would go blank. I jumped around three to see if four would work solo, nothing. We decided to pull four down and use the spare. In the process we discovered if four was unplugged from AC then three would work. We hung the spare and got the same result so we changed address to see there was a conflict, same thing. We changed power to another curcuit in the breakout, nothing so I ran a homerun of power, same thing. Then three started a lite error so we replaced it with the original four. So now the old four is in the three spot, the spare in four spot with the same symptoms. I also tried putting an isosplitter between two and three, same result. To get the three light working I had to unplug four from power but I had to have the DMX line attached. If I took the cable out between three and four, I lost control of three.
The only two things I didn't do was replace the cable between two and three or try a terminator. Neither the lighting company nor the rental house believe in them. I am constantly having issues with these two outfits. Three pin cable comes from the rental house and I have never been able to determine if the cable is made for DMX or just audio cable.
Any ideas?

[SteveB]

" In the process we discovered if four was unplugged from AC then three would work. We hung the spare and got the same result "

Quick read makes me say "bad data cable" between 3 & 4. And FWIW, the SS's are old enough to want terminators and I use one on mine.

I would start by getting Fix 1 to work by itself, then add Fix 2 and see what happens, then 3, then 4, trying to narrow down where the problem starts. If 4 has an issue I would first suspect a bad data cable. If you have 3 fixtures working and no spare data cable, then take 3 off-line, and run a direct line from 2 to 4. That may narrow down to cable or fixture - having already seen fix 3 work with a (in theory) good data cable.

[mstaylor]

One and two were never a problem. Three and four was the issue. As far as data goes, I took the cables out of three and patched them together to see if four would work by itsself. I also took the cable from two and took it straight into four. I could never get four to work, although it did when moved to the three slot and had the data plugged and power unplugged on four. Darndest thing thing I've ever seen. I have probably convinced the lighting company to get terminators.

[SteveB]

Lot's of uncertainties that are driving you nuts:

1) Is it mic cable or real data cable ?.

2) Is the cable good ?.

3) Are 2 of the fixtures good ?.

And a terminator is needed in any case.

[mstaylor]

The thing that I was struggling with was what does what is plugged or not plugged into a downstram light that isn't on have to do with a preceding light. I can understand a light not working or doing weird things because of bad cables or lack of terminators, but being dependant on an odd combination of plugging and unplugging of a light passed it I can't begin to understand.

[FMEng]

The thing that I was struggling with was what does what is plugged or not plugged into a downstram light that isn't on have to do with a preceding light. I can understand a light not working or doing weird things because of bad cables or lack of terminators, but being dependant on an odd combination of plugging and unplugging of a light passed it I can't begin to understand.

Think about waves in a bath tub. They travel along until they run into the side of the tub and then reverse and travel the other way. The signal inside a properly matched and terminated travels from the source to the last device in the chain and stops. The signal in an improperly matched system reflects and bounces back and forth just like the bath water. At certain points along the cable, the amplitude of the reflections hit maximums and minimums. Therefore, at certain points along the cable, the reflections will interfere with the desired data signal. If a receiver happens to be at that point, it won't work. Where that happens along the cable is mathematically predictable, but not intuitive.

In your case, unplugging the one light was changing the impedance of the receiver slightly and moving the reflections slightly. Changing the length of cable might have had the same effect.

This is why the proper cable should be used and the terminator installed on the end of the line.

[mstaylor]

But what does plugging power or unplugging power to the next light have to do with anything? I understand the reflections or slap of the DMX and without a terminator I could fix that problem, the AC had me bamboozled.

[derekleffew]

But what does plugging power or unplugging power to the next light have to do with anything? ...

Possible grounding issue? See the LightNetwork threadf by Bob Goddard I referenced in this post.

[mstaylor]

Interesting, I was trying to figure a groound problem but couldn't sort it out in my mind. Let me give you an idea what they did for power. They used a mushroom, square molded box with recepticals around the outside, cams off the side and breakers under a removable top. They used a soco breakin to plug all the circuits, a soco up to the truss and a breakout to power the movers. So could an audio three pin with it's ground tied to the case cause a problem on the power side? If so then that explains all the problems.
That would give me a reflection or bounce back of signal caused by lack of terminator. By patching the DMX line between the last working light and the next changes the length so the moment that they meet is not at the light. By unplugging the power it affects a suspect ground loop in some manner.
I don't have to fully explain to them exactly the mechanics of what happened, as long as I can explain enough so it can be avoided in the future. The other problem I am fighting is the LD is good with what and how to use the lights and movers but he is not adept at troubleshooting, plus he is afraid to make me made. I can't understand the last part but he is always saying he is sorry I am up on the truss fixing his stuff. I keep telling him that the equipment isn't his so all we can do is make it work.

[porkchop]

I agree with Derek on this one, that sounds like a dirty ground. I had an issue a ways back where I had a chain of LED pars, maybe 7 on one data chain, and the second or third in the chain was freaking out, all the others where fine. I spent hours and hours trying to figure out the problem, replaced lights, replaced cable, etc... turns out the cable had become badly damaged right by the last unit it was just the one light several down the line that was being effected by it. Learning from that experience, I'd take a good look over all of the data cable in use, not just the cables plugged into the offending units.

If you are still completely boggled after that you might compare potential difference between the ground pin on the DMX line and the ground on the power supply. That's a REALLY big long shot, but most moving lights are chassis grounded and most boards are grounded to the chassis as well, if there is a significant potential difference that could be a problem, although I have no idea how you would go about solving it.

[mstaylor]

When I went to loadout today, I was told a couple of the lights have bad boards in them causing the problem. First I'm skeptical that every combination had a bad board. Second, if that was true, what the hell are they doing sending them out on a rental?

[FMEng]

But what does plugging power or unplugging power to the next light have to do with anything? I understand the reflections or slap of the DMX and without a terminator I could fix that problem, the AC had me bamboozled.

Simple. The receiver circuit inside the light changes input impedance between powered and unpowered. The impedance changes affects the behavior of the whole system.