Passively splitting DMX.

[Lampy512]

Hello smart people.

A colleague and I have been discussing the topic of passively splitting a DMX signal. 

I am strongly against passive splits for DMX.

My colleague insists they are perfectly fine to use.

I know that it will work but it isn't the recommended or "proper" way to do it.

If possible I would like as in depth reason why or why not passive DMX splitters should be used as possible. 

We both are fairly smart people and have a good understanding of how DMX and rs-485 works.

Any additions would be greatly appreciated.

Thanks all. 

[Don Allen]

It is not recommended to passively split DMX512 as it can cause signal reflections which leads to data corruption.

In reality most LED and moving had equipment, scanners etc. may have passive splitters. The DMX in and out sockets may be on a pcb with mm of pcb track between them or the input cable may go in via a few cm of cable from the input XLR connector, when it is joined onto the output cable, which goes to the output XLR cpnnector via a few cm of cable.

[kgallen]

Passive splitting is not the designed electrical topology of DMX which is designed to be a daisy chain.

You will split the signal power two ways, so a 6dB drop in each leg. This means the signal amplitude will be halved in each so you will reduce significantly the noise guardband at each receiver possibly resulting in erroneous reception  

How would you terminate? You can’t terminate both passive branches because the parallel termination impedance would be wrong. This means you are guaranteed reflections from at least one leg. 

For short lengths and/or of a couple of  lights it might work. But otherwise it’s the same answer as always: it will work until it doesn’t. 

Edited February 16, 2022 by kgallen

[Richard CSL]

I once had two identical dance floors 30 ft apart and ran a passive split (actually 3 ways ) no problems, never used a terminator. I know it goes against all the rules, but it worked. As my old boss used to say, you could run DMX through spaghetti if you tried hard enough.

Rules are made to be broken.!!

 

[Brian]

One of the counterintuitive things about reflections and lack of terminations is that you might find it not working at a given cable length. Make the cable shorter OR longer and it'll start working. Keep making it longer, for example, and it'll stop working; longer still and it'll start again.

Likewise, with a passive split, make a change on one leg and the other will stop working.

If you do it properly, with an active splitter, properly wired cables, and terminations, then it will always work.

Once you try to do it on the cheap then it might work one day and not, after a slight change, the next.

[Robin D]

I bought one of These some years ago and it's been faultless used 2-4 times a year. I couldn't justify spending then £120 or more out of my pocket. 

[gareth]

Sometimes you can get away with striding out into a busy road without looking, and make it to the other side unscathed. It's an incredibly bad idea, though, because the chances of it ending badly are high. The same applies to passive DMX splitting.

[richard_cooper]

I once helped trouble shoot a music venue where the lighting rig had been in place untouched for many years, but everything failed when they tried to add a couple more fixtures. All the cables between fixture were made to fit using cat5 of some description, replacing any single cable with something a different length caused the whole rig to behave erratically. Once we mapped out the maze of DMX cable we realised there must be a split hidden somewhere (it was in some trunking). It seemed that the original setup was literally the exact one situation which worked and they got lucky, as people have said until they ran out of luck.

[sunray]

3 hours ago, Robin D said:

I bought one of These some years ago and it's been faultless used 2-4 times a year. I couldn't justify spending then £120 or more out of my pocket. 

Cor blimey, I have 2 of those, purchased 2016 at £9.99 and 2017 at £12.99...  Now 40 quid!!!

As you say they have worked faultlessly since although when one started rattling I found the circuit board was only held in with hot melt (Personally I've never found anything hot melt actually sticks to). Now mounted with nylon screws and nuts... However they are not at all tidy inside.

 

Edited February 16, 2022 by sunray

[gareth]

3 hours ago, Robin D said:

I bought one of These 

a) Three-pin 😡

b) Someone needs to buy him a square! I don't think any of those connectors are at the same angle as any of the others 😆

[Jivemaster]

The standard says that 485/DMX needs terminating but some people swear by terminators and some people swear at them. Data takes so much then it stops completely.

I knew a DJ who ran DMX to the top of a stand then sent it to each of four lanterns in parallel via bell wire! But it worked for him. 

[sunray]

In a nightclub, there was a DMX network controlling a grid of 13x13 RGB PAR64's.

However some 6 way dimmer packs for incandescent dimming on a new DMX network was creating a problem... 

We found the  DMX was conventionally wired as a row of 13 fittings, without a terminator except the very last row. But the start of each row was connected using choc bloc 'back bone' style to the main DMX run, ie 12 passive splits/13 rows being fed in parallel from the single feed from the controller.

As a test I made up some m/m & f/f cables to be able to feed the DMX backwards along alternate rows to see if the problem went away, fully expecting to return but we never did.

 

 

Its not uncommon to have a stand with a number of fittings, think mobile DJ, with the DMX to simply be split to each but the longest run is only a couple of feet or so... oh and depending on the 'quality' of the manufacturer of such systems it was just as common to be a bit of fig8 speaker flex.

 

[sunray]

25 minutes ago, gareth said:

a) Three-pin 😡

As I the only piece of DMX kit with 5 pin connectors I have is a Swisson 120, that's somewhat of a relief😄

 

29 minutes ago, gareth said:

b) Someone needs to buy him a square! I don't think any of those connectors are at the same angle as any of the others 😆

They are made in a sweatshop in the far east...

[dje]

Interestingly when you look inside a lot of fixtures, the male and female DMX connectors are just soldered onto one another in parallel.

Thus in essence if you are daisy chaining these fixtures, then you are passive splitting on the rear of the fixture. One leg is going into the electronics and one leg is going out to the next fixture.

Not saying it is/isn't OK... just saying, devil's advocate, maybe more people are passive splitting than realise it 😀

[kgallen]

2 hours ago, dje said:

Interestingly when you look inside a lot of fixtures, the male and female DMX connectors are just soldered onto one another in parallel.

Thus in essence if you are daisy chaining these fixtures, then you are passive splitting on the rear of the fixture. One leg is going into the electronics and one leg is going out to the next fixture.

The resulting stub length on the transmission line within the fixture is a couple of centimetres, not long lengths of cable. Electrically, at DMX frequencies, this is a very different situation to making a passive split then hanging tens-of-meters of cable off each leg. Electrically, a DMX network is a transmission line, which has very specific electrical characteristic and behaviour. The tolerance to stubs and impedance mismatches causing reflections is a function of frequency. At DMX frequencies, short unterminated stubs like those inside the fixture can be tolerated whereas such stubs would be completely unacceptable at RF frequencies.

Edited February 17, 2022 by kgallen