Control/Dimming Certain lights going out after crossing 75% intensity

[Aerial]

I am using:
Strand 300 console
Strand CD-80sv dimmer rack

I have managed this issue a couple times on small shows but it is a hassle. Most fixtures work fine between 0% and 75% intensity. Whenever the intensity passes somewhere between 74% and 82% (haven't pegged it to a specific value yet) some fixtures on the system will go off--a few each on our onstage electrics, catwalk and house lights. When the intensity is brought back down below 70%, the problematic fixtures come back on. This past week I've just set select channels to not to go past 70% and I didn't encounter problems.

My first diagnosis was a wiring connection issue. I did see one obviously loose wire in the processor rack and tightened it up. This did not make a noticeable fix. Because I already brought a good bit of equipment into the room I went ahead and did the yearly vacuum clean-out, filter change, etc. in our racks down there. When I booted up the console again it prompted me for a cleanup, so I did. I've wanted to clear out some old show files and such from that computer so might as well do it at the same time. Everything was set back to default--to this point I haven't found any hidden fxs, profiles, time settings or anything else in the system to throw things off. Only problem with the 300 is getting the iPAQ wireless hand held to integrate into the system again, but that’s for another day.

I've yet to determine if the issue is with the CD-80. The panic button brings everything to full as it should. I had an error code for the processor not being seated. It has since been securely placed back in. I referenced the dimmer.com diagram of how the CD80 rack is mapped out for the ramp (which shouldn't be on an sv) and PSU card addressing (CD80 racks). I not sure if the PSU info corresponds to CD80sv or not. What I find interesting is comparing which dimmers are having problems to the diagram...all the issues are coming from the third PSU; the third and sixth columns of the racks. All those dimmers have the go out at about 75% issue. Fuse doesn't make sense because I would think nothing would come on at all, and I can still get full intensity through the panic button. This brings me back to my initial guess on a wiring connection in the system after the dimmers.



Does this sound familiar to anyone? Anything else to check on or tests to run to possibly pin down the source of the issue? The rack appears to be okay, and the console software doesn't have anything odd sticking out to me. I'm still not 100% certain on all the mapping of the other parts of our system (luckily its very small) but this issue sure is helping that department. I'm going to study the manuals for the other parts of the system that I'm even less familiar with and hope to get somewhere (including S110 node, Strand Premiere processor, Doug Fleenor Design DMX Combine, and some baseline switches)


As an aside, I have found some interesting info on both ControlBooth and other sites regarding the processor of the CD80sv. Apparently people seem to have had issues with the battery. When I re-seated the processor I took some pictures of a yellow encasing that looked flexible (I'm not about to touch an unknown something on a circuit board) that had some brown goop on it. Not sure the gooey stuff is normal but that stuck out to me. The pictures are poor so I won't post 'em unless someone asks, and I don't want to remove the processor again unless need be.

 

[chausman]

Have you checked the fixtures at all?

 

[JD]

Post the picture of the goop so we can make sure it is not a leaking backup battery or electrolytic cap. When the dimmer is set to 75%, is the light on at 75% or 100%?

The reason I ask that is there is some importance to the 33% and 67% mark (liner) which may actually translate to 75% as the curve is not liner. In any case, those points are where the zero voltage cross resets are happening on the other phases. The said "goop" may be conductive and there is a chance this may be related if some traces are allowed to cross talk. Bit of a shot in the dark, but the turn-off sounds so distinct.

 

[Aerial]

I have tried switching fixtures and dimmer modules

It does appear to bring itself up to full intensity before going out. A 25% intensity difference should be noticeable, and I got another set of eyes to check it out as well. There is no noticeable change in intensity between the threshold level and the panic(full) level. Also when scaled relatively to other fixtures going to 100% and the problem ones to about 73% on a submaster they appear to track simultaneously to full intensity (tried with multiple problem dimmers and good ones all at once scattered in spots of equal distance/focus straight to the cyc).

I have checked the dimmer curves and voltage levels on the rack and everything is the same there. Everything that I have came across in the CD80sv menu is set to equal values throughout for the settings that can be changed for individual dimmers.

 

[JD]

Hard to diagnose blind from 2000 miles away, but two things come to mind-

1) Software error causing dimmers to be controlled based on firing information from the wrong phase. (Check single phase vs three phase configuration on the dimmer software settings.)

2) Hardware failure due to trace damage on control board from a leaking cap or battery. (Pic of "goop" would help.)

The concept that the dimmers are attaining 100% output when set to 70%+ and then go out has distinct implications that the firing signal is going out of timing range by eclipsing the correct zero voltage cross point.

Let us know what you finally find out.

 

[Aerial]

here are the pictures that sort of turned out

 

[chausman]

here are the pictures that sort of turned out

View attachment 5270View attachment 5269View attachment 5268

The middle picture is rather artistic!

 

[JD]

That's a transformer. The brown stuff on it is called MFP (Moisture/Fungus Proofing) and is normal. So, that's good news as damaged PCB traces are real trouble.

 

[TimMiller]

That looks like the coating on the transformer which looks normal.

 

[FMEng]

This sounds like a control processor problem, not dimmers, not connections. Time to talk with Strand or get a technician on-site.

 

[FatherMurphy]

Do you have more than one rack? If so, swapping the processors and seeing if the problem follows to the corresponding modules in the new rack would tell you if you definitely had a processor problem. Also, if you have access to a different console, you could eliminate the 300 as a possible cause.

I'm leaning towards 'time to call Strand tech support', too.

 

[Aerial]

Thank you everyone. I haven't responded due to being out of town. After talking to Strand, it does appear to to be electrical. The dimmer rack is receiving power phases of 120v, 124v, and the problem one only 98v. Next course of action is getting an electrician over here.

 

[FMEng]

I glad you've identified the problem. You have an open connection or open fuse on one phase.

This might be a good time to explain how to check the voltages on three-phase power. Let's call the three phases A, B, and C. Take three measurements:

A to B, B to C, A to C. All three should be ~208 V for a wye connected service (wye is typical to stage lighting). A bad phase will be indicated by TWO, very out of whack readings, so that makes it fairly obvious.

Why not measure phase to ground? Because a dead phase may show voltage via sneak paths through the load, from the other phases. Your 98 V reading is likely a completely dead phase, but the reading is just a slightly out of kilter. The only way measuring three-phases to ground can be definitive is if all loads are entirely disconnected, preventing sneak paths. If the phase were really a solid 98 V, your dimmers would work OK except lights up full on that phase would be dim.

Here's an example:
A to B = 208
B to C = 100
A to C = 65
This means phase C is dead, because that's the one common to both bad readings.

Voltages rarely sag for long because a voltage drop in a high current circuit means a bunch of expended heat. Heat leads to melting insulation, bad smells, and eventual fireworks. Most of the time, a three phase feeder problem without heat, odor, and fire simply means one or more open phases from bad fuses.

 

[Aerial]

Voltages were checked via multimeter on all phases of power throughout the system: into the building, at the company switch, in the rack on both sides of the fuses (also tested with different fuses) and at the dimmer module plug-ins. Everything appeared fine and was all at around 120v. However, the processor still read that it was receiving voltage in the low 90's.

Solved!

The folks at Strand were very helpful as they always are. After resetting the CD80 to default, the rack still read it was getting a voltage level in the low 90's on the troublesome phase. The solution was to go in and manually set the phase voltage; to let the processor know that it was receiving a proper 120 voltage. In the CD80 manual this was one of those settings with the description to contact Strand first.

The next step is to pull out the multimeter again and make sure to set the rack to the proper voltage value (I did not have a DMM on hand while performing this task). If I'm resetting it, then it might as well be exactly correct.

Everything seems fine right now and nothing is cutting out. I had set dimmer profiles and such to stop the problem from popping up in shows. That all can be cleared now. A little more playing around and hopefully I can soon be 100% certain this problem is fixed. Now if I only knew the original cause...

 

[derekleffew]

...A little more playing around and hopefully I can soon be 100% certain this problem is fixed. Now if I only knew the original cause...

Sounds vaguely similar to the issues raised in this post referencing a harmonic mitigating transformer. You may need something more sophisticated than a true-RMS responding DMM to diagnose the problem. Do your electricians have access to, and knowledge to use, a PQA?

 

[JD]

Sounds vaguely similar to the issues raised in this post referencing a harmonic mitigating transformer. You may need something more sophisticated than a true-RMS responding DMM to diagnose the problem. Do your electricians have access to, and knowledge to use, a PQA?

Unless the transformer was recently changed, harmonic filtering causing ZVC errors is unlikely. Also, it sounds as if the problem has been resolved for now by recalibrating the monitoring circuit. Most likely it is that the monitoring circuit itself shifted way out of calibration. This may indicate a ongoing failure in the circuit itself. If it starts to drift again, then you have a component in failure within the circuit. Another possibility is simply that stored values were corrupted.

If it walks, a follow up call to Strand would be indicated for assistance in obtaining a replacement.