Control/Dimming Y cable to split DMX

In the theatre that I commonly work in we have one dmx line that goes to our dimmers and another that goes to the stage for some led's and movers. The board has two outputs that are both on the same universe. This board is becoming unreliable and the new board has only one output. Since everything is 3 pin dmx is it possible to use a y cable out of the board to split the two lines?
 
... is it possible to use a y cable out of the board to split the two lines?
Possible? Yes. Advisable? NO!

From Doug Fleenor Design - DMX Primer :
Splitting

Never split a DMX signal with a "wye" cable. Use a splitter device that buffers each line separately or daisy chain from device to device.

The problem with splitting the signal with a wye cable is that the signal going up one leg of the wye is reflected back down and corrupts the signal going down the other leg. The longer the legs of the wye, the worse the problem. A wye where each leg is 10 feet may work fine but may fail if the legs are increased to 100 feet. A lot has to do with the quality of the cable, the strength of the signal at the wye point, and your relationship with God.

If you must split the signal, and you don't have a buffered splitter, do it as close to the source as possible, i.e. at the console. If you have an isolator available, place the wye at the output of the isolator. If you have two isolators available, place the wye at the input to the two isolators (which is like making your own splitter).
Doug "Dr. DMX" Fleenor and others have been answering this question since (at least) 1993. The only prudent answer is to use an opto-splitter, of which there are many offerings at various price points and qualities.
.
 
I should probably just make a new thread, but I always see people say "opto-isolated splitter". Are there none opto-isolated splitters? If so what is the point? It seems like optical isolation (high voltage protection) is something you would want on a show so I can't fathom a reason for a non opto-isolating splitter.
 
I should probably just make a new thread, but I always see people say "opto-isolated splitter". Are there none opto-isolated splitters? If so what is the point? It seems like optical isolation (high voltage protection) is something you would want on a show so I can't fathom a reason for a non opto-isolating splitter.

I'm pretty sure a non-opto-isolated splitter would be a wye cable. And probably a bad idea.
 
I'm pretty sure a non-opto-isolated splitter would be a wye cable. And probably a bad idea.

Sorry, but WRONG.
A true opto isolated splitter has at least N+1 power supplies for N outputs, one for each output, one for the input and "split", but ideally these 2 are separated also.
These get expensive, so you can have an active splitter that doesn't opto couple everything, just uses transceiver chips fed from a common supply. I understand some of the cheaper splitters do exactly this.

Data wise it's fine and provides all the benefits of an opto splitter.
Fault protection wise, not so, one spike will take out all of the outputs rather than just one if you were using an opto isolated device...

As always, use the right tool for the job, opto isolation will be needed for some things and optional for others :)
 
In my theatre I have three Opto splitters, with feedthrough input and 8 isolated outputs. These are used for all of our permanent runs and a few convenience outlets. I also have a three way splitter that has isolation optically and electronically from the input to the output, but not between the 3 outputs. I have another that has no isolation at all. The input, and two outputs are just transceiver chips. These last two are used as last minute installs for small shoebox dimmers driving practicals. My last unit is a non-loop through input and 8 outputs on 3 pin XLR. It has no isolation at all and the transceiver chips are all surface mount. It is powered by an external 5 volt walwart supply. I purchased it for $35 dollars free shipping on ebay from China. I mostly purchased it just to see how good it could be for that money. It has come in handy a few times to drive some very old Chauvet Color splash juniors. They are 3 pin XLR and frankly we use mic cable with the setup. If any part of it fails, it is no tradjedy, as there are so many of them, and they are used as supplementry effects.
My point is that you use opto-isolated distribution where it is important. If you get some good deals on some lesser spitters, they too can have a place for use when it is not important. It is also important that the person installing understands the difference and uses the best available for the application.
 
As others have put forth, a DMX 'wye' is a bad idea. While there is one place it can be used, it's just a bad idea as it is so easily misused by uninformed technicians. While not always necessary, opto-isolation is always a good idea. Opto-isolators serve to break ground loops and to maintain data integrity. Good splitters provide isolation from input to output and isolation between all outputs. Again, this is to break ground loops and to separate runs to keep failures confined to a small portion of the system.

While we're on the subject, always terminate the end of each run with a 120 ohm 1/2W resistor (or larger) between data plus and data minus. Also, there is no such thing as "3 pin DMX512". OK, I'm off of my soap box for now.
 
I know a company that does a lot of LED and Moving Light shows. That company does carry electronic splitters to their shows but they aren't used unless there is a problem. But they use Y-cables all the time without conflict. I confronted one of the Technicians about it and he said that it's just a Y-split inside of the fixture so there was not a reason that he shouldn't split DMX with a Y-cable. It seemed to make since to me because I had been through a few "High End" classes and the only fixture they had that electronically split the signal inside the light was the Intellabeam.
 
Yes, in most cases, it is a passive "Y" split inside a fixture.
BUT, there is one critical difference in doing that inside a fixture and doing that on the line.
It's all about the length of the spur off the main trunk...

In a fixture it might be at worst say 2 or 300 mm, compare that to a DMX line of what 50, 100, 1000 metres and the spur length is a very small fraction of the total line length.
Basically, it can be ignored.

But the relative length is only a part of it, the other is back down to physics. The signal does not take much time to travel down 300mm of cable and so by the time it reflects back at the other end, the signal probably hasn't changed and so it doesn't interfere much. When you start to get longer spurs, you can get to a point where the reflected signal arrives back at a later point in the data stream, which is also why the faster the signal, the less tolerant it is of "non standard" cabling topologies.
There are LAN systems that run architectural lighting control and things like that that run at a baud rate of 1200, and have a permitted topology of "whatever", because the 1500 metre limit of an RS485 LAN means that the drivers have run out of oomph before you reach the distance of reflections distoring the signal beyond recovery.

So it comes back to the simple reality that just because it works, does NOT make it right...
 

Users who are viewing this thread

Back