Strange DMX issue -

[JChenault]

I had an "interesting" day today, I thought I should share. First in case if helps someone else - secondly in case someone could explain what might be going on.

We are doing a show in a high school venue. The lighting system consists of distributred dimmers . The DMX control for each dimmer pack runs directly to an rack of Pathway - DMX splitters ( DMX Repeater Pro). There is no daisy chaining of DMX. In addition to using the splitting functionality of the system, they are also apparently using to to merge DMX streams, and to act as a hub where the system will sense which DMX outlet is putting out signal and use that as the source.

The house system has an element for control.

I attempted to connect my control console ( Plexus) using a "DMX King" USB dongle. I have used this product many times going direct to DMX devices and have had no issues. I found that the response to the system was noticibly sluggish. There would be a noticible delay when I moved the slider before the intensity of a fixture would change. If I was fading, the fade would stutter for a time before completing.

I connected a second DMX King and had the same result.

The next day I attached one of Baxter controlls pocket DMX to the system ( for debugging) and things were smooth.
I then attached the Enttec USB Pro Mk2 and the problem was solved. Fades are smooth and all is happy ( I have not tried the older Enttec dongle yet )

So - if you are using the DMX-King, and have some hesitation, give the Enttec USB-Pro a try.

Secondly - does anyone have any idea what could have been going on here?

 

[Footer]

Could be a timing issue. The DMX King could be dropping frames so the merger could be dropping the connection then restoring it. The Enttec unit does not do this from my understanding Give DMX King a call, see what they have to say. Someone who knows more about how dmx works on the wire could probably give a better answer, but there is mine!

 

[techieman33]

I know on the ethernet version of the DMX king part you can change the timing with their configuration app, that may help you if it is a timing issue.

 

[jglodeklights]

Also sounds, to me, like a timing issue.

In the last changeover I did a couple weeks ago, we found our rented strobe caps from Diversitronics were firing randomly, in addition to when issued data over DMX. Adjusting the transmit speed from slow to fast on our ION improved the situation. Unlikely a bad cable as one was direct to the Diversified Electrics optosplitter, and the other was through a Wybron PSU connected to the splitter. Termination neither improved or worsened the problem. We rented a second splitter, but made by Elation, and used it to output to the strobe caps. Problem solved. Unless the ION has a physical problem with the first DMX output and DMX timing, this would seem to point to a timing issue created by the splitter, somehow.

I have had a similar problem with a cheap dimmer pack being the third in a daisy chain of four (the other three were of slightly better quality). The fourth would flicker. Swapping their positions on the DMX chain solved the problem.

 

[JChenault]

I know on the ethernet version of the DMX king part you can change the timing with their configuration app, that may help you if it is a timing issue.

I'm using The USB version.

I was not aware of a config application from DmxKing. Where did you find it. When I used the Enttec config tool the DmxKing would not let me change the speed.

 

[techieman33]

I'm using The USB version.

I was not aware of a config application from DmxKing. Where did you find it. When I used the Enttec config tool the DmxKing would not let me change the speed.

Looks like you want the ultra dmx configuration tool, it's the 2nd one up from the bottom on the download page. You could also try updating the firmware.

DMXking.com Downloads

 

[MiltonDavis]

There are a couple of possible problems that could cause the original issue. The first thing I would recommend is to connect the DMX King directly to one of the distributed dimmers (not through a splitter). If the problem does not appear in that scenario, it is likely that the timing of the DMX King is allowing the splitter input to time out as one of the first responders suggested. If the problem continues to display itself when you connect directly to the dimmers, it will be the dimmer that is actually at fault. In either case, a DMX re-timing device will take care of the problem. Most DMX testers are capable of doing the re-timing. (/start of commercial statement) Of course, I work at Doug Fleenor Design, so our DMX Decelerator comes immediately to mind for me. It's a $200 box that re-times the signal and opto-isolates it. (/end commercial statement)

 

[WooferHound]

A lighting company regularly brings a console into our theater that has the same delayed response that is being described here. I can't remember the name of the console but it has a single screen in the top center of it that is operated with a stylus. Our dimmers are ETC Sensor racks about 2 years old. But as Milton suggests in the above post, we started putting a DMX Decelerator on the output of the console and it solved the problem.

 

[DavidNorth]

A lighting company regularly brings a console into our theater that has the same delayed response that is being described here. I can't remember the name of the console but it has a single screen in the top center of it that is operated with a stylus. Our dimmers are ETC Sensor racks about 2 years old. But as Milton suggests in the above post, we started putting a DMX Decelerator on the output of the console and it solved the problem.

It's likely a Jands Vista and I bet you have CEM3 control modules in your Sensor racks.

A while back, after contacted by two Jands Vista users, we found that the UART in the CEM3s have an errata in the silicon design that keeps DMX signals of some very specific timing parameters from controlling the dimmers. Turns out that Doug Fleenor Designs Preset 10s also share the same timings. This is seen as very delayed response to changes as the dimmer rack is only picking up occasional frames of data.

The nice thing about this problem is that both Jands and DFD were great about coming up with changes to their output timings to reduce the issue. Also, ETC built a work around in code until we could come up with a hardware design change to work around the errata. Gotta love chip errors. Software engineers also like to gloat that most any hardware problem can be fixed by software.

There are several ways to resolve this timing issue and I suggest doing as many of these that apply. Update the software in the Jands console as the newer releases change timing. Update code in the Preset 10s as DFD has the ability to respond to this quite well. Update the software in the CEM3 to v1.4 and enable a feature called Force DMX Timing, which is a setting in one of the back door menus. We can certainly help you set this in each rack with a quick phone call or email. The work around from the software release notes is listed below. Channel shifts and reduced control response are the typical issues that arise.

RND 0021177 Long DMX break lengths require “force DMX timing.” Using the forced timing on incoming DMX that does not have
a long DMX break length (between ~200uS and 500uS) will cause the DMX patch to be off by 1 address.
Workaround: Do not use “force DMX timing” feature unless DMX break length is between 200uS and
500uS.

So far, your site I believe is the fifth or so to note this issue, but we have for a long time prided ourselves in DMX design. It's cases like this that give us some indigestion.

Sorry for creating a mess down there. Please let us know what we can do to help.

David

 

[jpk]

Funny thing is we've just released a firmware update that would fix this one for you. Don't know which ultraDMX variant you've got but here's a link to one of them ultraDMX RDM Pro
click on the Downloads tab. There is the odd piece of DMX gear out there with receive timing issues and what we did was wedge a DMX Decelerator right into the ultraDMX firmware across all 4 variants (original ultraDMX, Micro, RDM Pro and Pro). Certainly some USB DMX hardware might have different timing parameters which "fix" the problem but please don't label good hardware potentially bad unless you're really sure! In almost all scenarios it's the receiving end that is problematic or cabling/termination issues.

Interestingly over time we've also encountered another USB DMX issues that crops up occasionally when receiving DMX so we also added DMX RX Throttling control to keep the frame rate forwarded over USB limited. Some software packages have issues with particular lighting desks (Pangolin Quick Show is a good example) and we wanted to fix this independently even if you feed 800 DMX frames per second into it.

For reference the ultraDMX Pro can also act as a standalone DMX Decelerator when in splitter mode. It's become the Swiss army knife of USB DMX.

 

[JChenault]

Funny thing is we've just released a firmware update that would fix this one for you. Don't know which ultraDMX variant you've got but here's a link to one of them ultraDMX RDM Pro
click on the Downloads tab. There is the odd piece of DMX gear out there with receive timing issues and what we did was wedge a DMX Decelerator right into the ultraDMX firmware across all 4 variants (original ultraDMX, Micro, RDM Pro and Pro). Certainly some USB DMX hardware might have different timing parameters which "fix" the problem but please don't label good hardware potentially bad unless you're really sure! In almost all scenarios it's the receiving end that is problematic or cabling/termination issues.

Interestingly over time we've also encountered another USB DMX issues that crops up occasionally when receiving DMX so we also added DMX RX Throttling control to keep the frame rate forwarded over USB limited. Some software packages have issues with particular lighting desks (Pangolin Quick Show is a good example) and we wanted to fix this independently even if you feed 800 DMX frames per second into it.

For reference the ultraDMX Pro can also act as a standalone DMX Decelerator when in splitter mode. It's become the Swiss army knife of USB DMX.

Jason
Sorry to take so long to reply to this post - I've been in tech week.

I hope my post did not come off as labeling your hardware 'bad'. I have used it extensively in the past, and will again.

There have been times when the Enttec hardware failed, and DmxKing worked flawlessly. ( in that case it was low power from the PC driving the on board processor. Seems like you guys do not need as much power as the Enttec device).

This was a case where someone else's equipment did not co-exist nicely with yours.

My intent was not to say any disparaging things about DmxKing - you guys have a fine line of products, and I will be fascinated to learn more about the new features.

The response issue was occurring on both the ultraDMX and the ultraDmx RDM Pro. Since I was able to work around the issue I have not yet tried the new frirmware. I am not at the theatre now, but I will plan to install the firmware this evening and let you know how it works.

 

[jpk]

Not a problem John, perhaps we may catch up again this year at LDI.

 

[WooferHound]

This DMX delay issue has popped up again yesterday when a national act came in with a new "GrandMA 2 Light" console and we started getting sluggish operation with some dimmers not responding for up to a minute. Our house Element console seems to have a half second delay all the time.
I remembered that our facility had a Pathway DMX Ultimate Converter that wasn't being used that night, so I got it and set it up to decelerate the DMX signal. Worked wonderfully, but any new console may not be compatable with this system.

 

[JChenault]

This DMX delay issue has popped up again yesterday when a national act came in with a new "GrandMA 2 Light" console and we started getting sluggish operation with some dimmers not responding for up to a minute. Our house Element console seems to have a half second delay all the time.
I remembered that our facility had a Pathway DMX Ultimate Converter that wasn't being used that night, so I got it and set it up to decelerate the DMX signal. Worked wonderfully, but any new console may not be compatable with this system.

What is the network topology of the existing system? The element half second delay seems wrong

 

[WooferHound]

We discussed it a little bit above, but it's standard DMX cable throughout.
On our last show we had a tablelamp with a 100w bulb in it. Should have been super fast response but there was about a half second delay when pushing the bump buttons. David North ETC suggested that I upgrade the software in the CEM modules and reset some DMX timing. I believe that is going to be the next step.