Express & Expression issues (possibly midi?)

hobbsies

Active Member
Hi everyone, sorry for such a long post, I felt it was important to have the full background as I saw things.

I am the ME for a midsized theatre company in Chicago. I have been having some really funky issues with our lighting system that has boggled my mind. It's a long story, which I'll post below, but the TLDR version is that our Express 48/96 and now an expression 2.x is freezing/crashing. I suspect it's from midi, because sending Midi cues from the sound computer out to the Express will definitely cause it to freeze. We haven't had issues with an expression 3 or 2 until tonight, both of which we've used as backup due to a little accident during tech on the express.

The story:

The venue owns a 48/96 express console. About halfway through tech, some coffee was spilled on the tech table and seeped into the board. No one noticed the spill because it was mostly under the board and we didn't see it until the board starting acting weird. The track pad was engaging as soon as someone typed in a channel number, making it impossible to select a channel higher than 1-9. We took it to a local shop and had it fixed. In the mean time, we borrowed an expression 3 from another theatre, copied the show over and continued programming on that, and kept using it well into the run of the show.

When I got the Express back from the shop, I hooked it up to test it. I noticed some weird issues with the way it interacted with a leppy pack. The practicals (clip lights and bare lamp sockets) wouldn't fade on and off correctly. Instead of fading smoothly, they would jump to 30, 70 then full. There would also be a two second delay when trying to release those channels. This problem occured rarely, but I was able to reproduce it sometimes and wasn't fully comfortable swapping the expression 3 out for our fixed Express. I talked to the shop about the board, but they insisted nothing was wrong with it. So we continue using the Expression 3.

Fast forward 6 weeks or so, the theatre which we borrowed the expression 3 from wants it back to be their backup. Their Ion has failed in the past and they want to have it around incase **** hits the fan, fair enough, I know the feeling now. At the end of the week I tried swapping the express 48/96 back in. I didn't see any of the funkiness with the leppy pack, in fact it seemed to work perfectly. Midi commands were functioning, all of the cues were working, the everything seemed to be fine.

The next show is on thursday, and come thursday I get a call from the stage manager that none of the lights are responding at all. I had him reboot the board, reload the show and finally turn everything off and then turn it all back on. Nothing worked. I head over the to theatre to diagnose the issue, and discover that nothing in the first universe is responding. I try swapping the DMX to the 2nd universe, assign it to start with DMX 1 and we have limited control over the system. The conventional dimmers were working, but some were ghosting and flickering. Two of the moving lights were spinning widely with no control, we had no control over the LED Pars or the Leppy pack. Also, Midi cues sent from the sound computer (a Mac) start causing the board to freeze. I tried reloading the show and restarting the board. We still had no control from universe 1 and uni2 was still very limited. We started the show an hour and a half late with some subs I programmed with the conventional lights while the PM drove to the theatre w/ the expression 3 and got it for act 2.

Once we hooked up the expression 3, act II ran totally normally. Conventionals were working fine, movers were happy, had full control and midi wasn't crashing the light board. We had to return the expression 3 the following day, so I borrowed an expression 2.x from another theatre.

I took the Express 48/96 to a different shop this time to get a second opinion. I described the issues I was having to the tech there and they tested it out. They called me and said they couldn't find anything wrong with it, although they had no means to test the midi functions. I picked up the board, tested it myself, caused it to crash about 10 times by firing MIDI cues to it. I tried deleting the show and reloading it using 3 separate disks, restarting the entire system again, rebooting the board, still crashed.

The Express 2.x has worked great for two weeks until tonight, when it froze several times during preshow. I don't know the details of what happened before it crashed, if midi was tested on it or not. Initially I had the stage manager turn off the board, and turn it back on, but it froze again when we tried to go into the preshow cue. Turning everything off, dimmers, movers, board, leds, leppy, hazer, then turned it all back on seemed to fix the issue. The show ran fine the rest of the night.

Some details:
The SM is running the show from an RFU
The length of the DMX chain is roughly 500'
The device order in the DMX universe 1 chain looks like this: Board <- 2.4k 24 dim sensor rack times 3 (72 dims total) <- 2 circuit, 6 dim 3600W leppy pack <- Radiance hazer <- Design Spot 250 Moving Light <- several LED pars, then a Mover, several more LEDs, Mover and finally several more LEDs. 3 Movers, 10 LEDs, 1 Radiance, 1 Leppy, 3 24 dim 2.4k Sensor racks. not terminated
Second universe operates a projector dowser.

Here's how the midi settings look:
1 - ETC MIDI Channel: 1
2 - MIDI Show Control Device IDs: 1/Disabled
3- reserved
4 - reserved
5- time code input: disabled
6 - time code frames per second: 30

I'm at a complete loss. I don't know what the problem is and I don't know how to fix it. Two shops say the Express 48/96 board is working perfectly, and we haven't had any expression 2 issues until tonight, and zero expression 3 issues through 8 weeks of shows. Has anyone experienced MIDI issues with an express? I plan on calling ETC tomorrow, but am curious if anyone has any suggestions.

Thanks so much :)
 
Last edited:

JWilsonLX

Member
I can't speak to any of your midi/output issues, but I can say that I've seen that very steppy dimming issue on multiple Lep packs, mostly the ULD 360. Also, if you're crashing multiple, otherwise stable systems with the same midi triggers, is it possible that this is where the issue lies? Seems unlikely, but worth checking out...
 

hobbsies

Active Member
It is a uld 360. Haven't seen it have a step issue since that one day though, even on the express.

What could I change about the midi triggers to make them work on the express? I know very little about midi or q-lab.
 

Clifford

Active Member
Whenever you test the system, are you testing it with the whole rig? Unplug everything but one of the Sensor racks and see what happens. Add everything downstream of that, then add the MIDI back until you run into the problem. Remove what caused the problem to see if it goes away, if it does, you've found the issue. If you add everything and nothing gives you trouble until the system goes down again, it's seems it's more likely the board. If that's what happens, call ETC (1-800-688-4116) to see if they can help.
 

HansH

Active Member
Premium Member
Haven't seen it have a step issue since that one day though, even on the express.

Speaking generically, as I don't know the ULD-360 that well -- The step issue could've been a zero-cross sampling problem -- usually after a power reset of the dimmers, this goes away. Essentially, the dimmers are trying to dim the circuits at the wrong times in the sine wave, causing erratic dimming.

What could I change about the midi triggers to make them work on the express?

It may not be the MIDI commands specifically that you are sending, but rather the MIDI hardware that you are using. What are you using to transmit MIDI from QLab? If you have another MIDI interface, or the MIDI interface has another port, use that to see if you get the same result on the Express - it might then be an issue with the MIDI circuitry on the console. In any event, Clifford's suggestions make a lot of sense -- try and isolate the issue, narrow the field of possibility, and feel free to give us a call (800-688-4116).
 

FMEng

Well-Known Member
Fight Leukemia
If the MIDI hardware at either end of the circuit, computer or lighting console, do not adhere to specifications for isolation, you might have a big ground loop inducing noise into the console and DMX. I would examine the specs of the MIDI port on the sound computer to ensure it uses opto-isolation on its input. If it uses a game port instead of a DIN connector, there is your culprit.

One thing you can try is to make sure the computer is on the same power as the light console, and then break any ground loops to the sound console.
 

hobbsies

Active Member
The sound macintosh is using usb to some sort of mixer which takes a 5pin din midi out to 3xlr runs to the express with another 5pin din adapter to the 5pin din in midi port. If that makes sense.

Edit1: It doesn't make much sense to me if the expressions are working fine but the express isn't if there's noise being introduced into the data signal. Wouldn't the expressions have the same issues?

Edit2: I unplugged the dmx from both uni 1 and 2 and still caused the board to freeze after about 15 midi cues sent from the sound computer, so it's definitely something between the express and the sound pc.
 
Last edited:

FMEng

Well-Known Member
Fight Leukemia
If you are having to adapt the MIDI I/O to XLR cables, that screams ground loop. I doubt you'd be doing that unless there was some distance between the computer and the console. Without examining the system at the component level I cannot say why it is happening. But, again, I'd be willing to bet getting the computer on the same power source as the lighting console, AND breaking grounds in the audio connections with dropped shields and transformers will fix it.

Try just moving the computer next to the console, keep the MIDI cabling short, and disconnect all of the audio. If the system works, then we'll know why it is happening. You could accomplish the same thing by moving the lighting console next to the audio area.

As for why one model console works better than the other, it could just be a minor difference in how they are built. Just the size and routing of a ground trace on a printed circuit board could change the result. A loose screw inside could be the culprit. In the audio world this is known as a "pin 1 problem." It is why one piece of audio equipment has hum and buzz problems and another doesn't in the exact same circumstances. Neil Muncy, an audio and electrical engineer, has done a lot of writing about proper grounding for EMI and RFI problems for the Audio Engineering Society. It's clear that many engineers and manufacturers either don't know or don't care much about the subject because the problems only show up once in awhile. If they sell thousands of units, and one customer complains, it isn't enough to increase the cost of making it to avoid the problem.
 
Last edited:

hobbsies

Active Member
Ok I'll try keeping the cable short and putting the computer on the same power source and see what happens. Also, I'm not sure I understand or know how to do what you're suggesting when you say
breaking grounds in the audio connections with dropped shields and transformers will fix it.
Can you elaborate a little?

Thanks for your help, I really appreciate it.
 

FMEng

Well-Known Member
Fight Leukemia
What I'm trying to get across is to test the system without any audio connections. If that works, chances are that restoring the audio connections will screw it up again even with the computer on the right power source. Or, audio hums will develop that you didn't have before. But, there are ways to connect the audio that won't cause the ground loop and hum to come back.

Ground loops develop between connected equipment when there is a difference in ground potential between them. That causes current to flow, which induces electrical noise into the circuits between the two locations. The classic example is hum and buzz in audio systems, but it can happen with data circuits or any other kind of connection.

A balanced circuit, like used for analog audio, DMX, and AES-3 digital audio, have immunity to noise by what is called "common mode rejection," but it isn't to hard for the ground potential difference to be too strong for balancing alone to keep things clean.

For audio systems, we can break the ground path by doing a couple of things. First, there is no need to have the cable shield connected at both ends. When two pieces of equipment are separated by distance, and/or are on different power sources, it is a good practice to connect the shield only at one end of the line. (The exceptions are mic lines, which need shield continuity to work.) In the connector at one end of a cable, the shield is left unconnected and tape or shrink tubing is used to ensure that shield whiskers or foil does not make an un-intended connection. Another way to do the same thing is to make a ground-lift adapter with a Neutrik NA3FM and cut the wire between pin 1.

Neutrik USA Inc Neutrik XLR Male to Female Adapters XLR to XLR Adapters at Markertek.com

In your situation, you need to drop the shield on every audio cable between the computer and the audio console, and then ensure that the computer is powered from the same receptacle as the lighting console.

If dropping the shield doesn't solve the problem, or if audio hum develops, we can achieve total isolation by employing an audio transformer in each line, usually at the input end of the line. A transformer passes audio, but will not allow any common mode current to flow. They are miraculous devices for eliminating hum problems. The one catch is that a cheap transformer can sound poor and not provide the amount of isolation it should. Don't cut corners with audio transformers. I highly recommend Jensen transformers. For your situation, their PI-2XX is probably what you need. On a tight budget, the Sescom IL-19 would do the job.
JENSEN TRANSFORMERS, INC. - ISO-MAX® Audio Isolator Products

Sescom Sescom IL-19 Inline Pro Audio Hum Eliminator THE INDUSTRY STANDARD Impedance Matching Transformer at Markertek.com

I keep of pair of Jensen transformers in my audio tool kit for solving unexpected problems. One example of what transformers can do is running analog audio thousands of feet between buildings, on un-shielded, twisted pair (telephone) cable with totally clean results. It would not be possible without them.
 

hobbsies

Active Member
FMEing, thank you for your suggestions. I made sure the board was on the same power source as the sound computer, and replaced the midi cable altogether which seems to have fixed the issue. We also installed an opto-isolator that will hopefully fix the moving light shenanigans that the Stage Manager has been experiencing. I cannot thank you enough for ending this nightmare for us.
 

Users who are viewing this thread