Frequency Error on ETC Sensor CEM

StradivariusBone

Custom Title
Premium Member
Fight Leukemia
We've recently begun noticing a flash from our front wash lighting when in any of our standard wash presets/shortcuts. Typically these hold the fixtures at around 30-50% and the accompanying flashes of light are about the length of half a second. We noticed tonight that it's just the front wash, the house and work lights aren't affected, though they are off when this occurs.

While watching the CEM display it did flash "Frequency Error" when one of the flashes occurred, leading me to believe that it might be related. We've also had a lot of A19 failures recently, which may only be coincidental to the other issue.

All the manual says about this error is that the mains power has a frequency issue. I'm planning on putting in a work order to have our electricians come check our building transformer, but also probably going to call ETC helpline in the morning. I wanted to also get the board's thoughts if you'd be so kind.

It's odd to me that it's only the front wash (and all of the lights in the wash seem to be similarly affected), but maybe it only occurs when it's under significant loading? I'll try turning everything on tomorrow and see if the house/work lights behave the same. I watched the dimmer signal on a few of the affected lights at the rack and it showed no change from the preset that should be on, so I'm reasonably convinced it's not a DMX issue, even though it only seems to be messing with a particular wash.

Another issue that I just remembered is that we had one dimmer channel sticking on full recently even though the rack showed no signal. I chalked it up to a dying dimmer and turned off the breaker for now, but I wanted to include all the symptoms.
 
Are there other lights on this rack that are not affected?

Does the flash only happen during specific cues?

Can you try another CEM in the rack?

A frequency error on building power is extremely unusual and can be really only be one of two things. Most likely it is a CEM going bad, but it can be repaired. It probably isn't a power issue but it could be something on the order or a loose or missing feed neutral. Usually the transformer feeding this rack is feeding something else in the building and you would know if there were issues with that transformer. Although it is possible that the transformer is failing, that is so very rare.

Let us know what you find and/or send this info on to phone support this morning.

David
 
It seems central to our front wash lighting look, both for meetings and concerts which use some of the same dimmers. I run a PC with LightFactory on it and it happened when I had a shortcut running from that controller and when the arch control had an identical look running.

I turned everything on to 50% this morning (the front wash looks are programmed to that) and no flashes. Like any good transient issue, it's not cooperating with my schedule. We have two racks, I could swap the CEM's as it appears that the flashing is only occuring from dimmers in rack 1. The transformer does have an additional switch, which I think feeds audio and additional non-dimmed lighting, but I'm not 100% sure. I'm only noticing problems in the aforementioned areas.

I did manage to film it and put it up on YouTube-
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.


The first flicker is visible in the first 5 seconds of the video. The second half of the video I slowed down the part where it flashes as my tech was changing presets. Note this is being operated by the arch control and not the LF computer, but both were at one point or another exhibiting flashes.

It seems like it's a handful of dimmers only. I'd have to sit down with our plot and watch the video to figure out which ones are doing it exactly. We have had problems in the past previous to the computer based controller (which stays on 100% of the time) when we'd shut off the Express board often times random dimmers would just turn on at various levels without any input. When the rack gets DMX from the controller though, this problem goes away. I'm thinking now perhaps something between the arch control and the CEM might be causing this? Still haven't had time to call this in, but I thought I'd update with more info.
 
DMX Death Gasp. Normal, and not related to the current anomaly.

My first thought too, but it was well after the console shut down. We've got some problems with a few of the single-button stations with arch dimmers. We did troubleshoot it to a point, but after we started leaving DMX running 100% of the time to the rack it stopped being an issue and got put behind other stuff.
 
The CEM does frequency and zero cross detection on all three phases individually, so you could get a scenario where only loads on one phase are affected if just one part of the detection circuit is failing. That brief flash you're seeing looks a lot like what happens when an SCR gets stuck on for a half-cycle when it should be dimmed, so it's a reasonable symptom for a CEM that is momentarily confused about the incoming power.

While you could get flicker another way, it's pretty unlikely that anything related to arch-control could cause the frequency error message to pop up. I would definitely go with David's suggestion of swapping CEMs between the racks to see if the problem follows, then send the bad one in to the nice folks at ETC for repair.
 
It looks to me like there is an issue only on one phase as not all the lights show the glitch, if I'm seeing it correctly. Having video is great - thanks for shooting that.

Yes, please swap the CEMs and see if the problem moves but I think the CEM has an issue on one phase. I think we will find it wants to come back to Middleton for a visit.

Let us know,

David
 
The fact that you are seeing an error on the CEM would narrow it down to the rack and power feed. Phase chop dimmers firing circuits use two different sets of math depending on if they are running at 50 hz or 60 hz. It could be there is noise on one of the phases that is causing it to misjudge the mains frequency. That would cause a big change in the dimmer curve at almost any setting. Impossible to know without having knowledge of the actual software programming in the CEM. Bottom line is, although the noise could be on the line itself, it also could be a failing component in the zero cross detector in the CEM. Best way to isolate that is to swap the CEM. If the problem follows the CEM you have your answer. If it stays with the rack, then most likely it is a power issue.
 
Just swapped the CEM's this morning and the "look" that had generated the issue before followed to the 2nd rack (albeit on different dimmers), but that troubleshooting narrows it down to the CEM if I'm understanding all correctly. It does appear to be phase A as well. Will be in touch with ETC shortly and figuring out a plan to send our little CEM on a big adventure. Should be interesting trying to run this place with only half the dimmers for a while. Thank you all for your insight!
 
Nope. The addressing was all different since I didn't switch it when I swapped, but if anything that told me that it was in the CEM rather than the power since it was lights flashing that had not done that before post-swap.
 
Just swapped the CEM's this morning and the "look" that had generated the issue before followed to the 2nd rack (albeit on different dimmers), but that troubleshooting narrows it down to the CEM if I'm understanding all correctly. It does appear to be phase A as well. Will be in touch with ETC shortly and figuring out a plan to send our little CEM on a big adventure. Should be interesting trying to run this place with only half the dimmers for a while. Thank you all for your insight!

Good. That nails it down. I reckon we can find another CEM for you to use on the meantime. Just talk to the folks here and tell them about your situation.

David
 
Good. That nails it down. I reckon we can find another CEM for you to use on the meantime. Just talk to the folks here and tell them about your situation.

David
Thanks, as always, David. Got a plan going, getting a spare and hopefully this CEM is repairable. Our shop here has had a few that were beyond hope and were implying that a new rack might be in order. Hoping to avoid that as I can't get the county here to pay for annual rigging inspections, much less $5-8k for a dimmer rack.
 
Thanks, as always, David. Got a plan going, getting a spare and hopefully this CEM is repairable. Our shop here has had a few that were beyond hope and were implying that a new rack might be in order. Hoping to avoid that as I can't get the county here to pay for annual rigging inspections, much less $5-8k for a dimmer rack.
At one point ETC had a program to upgrade older racks to the current CEM just replacing the CEM and leaving the rest of the rack intact. If your CEM isn't repairable that might be easier than replacing the entire rack.
 
Sent the CEM off today. The spare is working well, good confirmation of our troubleshooting. Many thanks again and I'll update as warranted for posterity.
 
The saga continues- to catch up after a great deal of troubleshooting and hard work on the part of the ETC techs in WI, the issues continued unabated. The solution for now was to retain the loaner unit and return my stubborn CEM.

Fast forward to today. I'm now seeing DMX input errors on Rack 1 (the troubled rack) from both A and B. Our controller is on and the arch control is active. Controller is not sending anything. I pulled the fuse on the arch control and the error was replaced with the "no signal from B" warning, but input A still showing an error. I'm seeing random bumps to 100 from multiple channels, seemingly centered around phase B, but I haven't looked that far into it. It seemed to get worse not better when the arch control was taken offline.

Got a class of techs coming in soon, but I thought I'd put it to the hive mind prior to my call to tech support to figure out this latest development. Arch controller has it's issues- prior to using a computer based DMX controller, lights would randomly turn on by themselves when the board was off (not related to death-gasp) among other weird occurrences. Most of the single button push stations are not functional as well. At least we don't have any big shows in the too near future.
 
You made the correct determination that the original problem was the CEM. This looks to be a new issue. Contact ETC. Loaners can have problems too.
 
So after some basic troubleshooting I found a single, long DMX cable patched into one of our optisplitters going to nothing, apparently forgotten by some of my techs after loading out our last rental. After unplugging it, I haven't seen any DMX errors, but the strange thing is the errors did follow the CEM when I swapped racks. Not 100% convinced that solved it, but it seems OK at the moment. It was behaving like a transient, unterminated DMX run issue though. I just never thought about it registering as an error on the rack itself? I guess that makes sense.

I know we have some issues with our arch controller, a few stations are non-functional and we've had strange issues when patching in controllers in different inputs around the building (temporary loss of control, all dimmers get stuck at full). So I'm thinking I will take this chance and try and troubleshoot that issue first. From the looks of it, the DAS is very much tied into the input side of both ports going into the Sensor, which makes it reasonable to me that a fault within that system might affect either input ports.
 
I would say in general that if you see errors persist at the CEM after you have powered down the DAS system then the problem is most likely not in your architectural system. The DMX line from your console to the rack also lands in the DAS system for recording purposes. Your wall stations most likely have worn buttons and can be repaired. DAS systems sometimes are glitchy which is why the master reset of pulling the fuse. I would not be shy at all about contacting ETC Technical Services.
 
Something simple to try: Grab a cheap board (like a chauvet) and patch it directly to the rack, eliminating all in house wiring and systems. Just the board, a 10 foot cord, and the rack. Set up a scene and let it cook for a few hours.
A) no errors and the system works: Your problem is not in the rack.
B) Still have the problem. CEM, Funky power problem, funky problem elsewhere in the rack.

Everything sounds like it points to the rack, except for two things- You have a loaner CEM in there, and the change of symptoms with the stray cable. Time to keep it simple and divide and conquer.
 

Users who are viewing this thread

Back