woo hoo!

[niall]

I was given a DMX install todo in work and I finally finished it today. A total of 12 wall boxes, 14 XLRs, and over 600m of DMX cable run.

and here's a diagram to illistrate the install!

red indicates Generics dmx line, and blue is for intelligents. M denotes a male box and F a female one, the grey boxes are our trusses.



took its time, but i'm glad its over, best bit is, it will be properly utilised at a gig in a couple of weeks! (i just need to redo the lighting plan i've been sent - they want 32 parcans along the back of the stage, we have 12...)

[alhaynes]

Looks impressive, but what exactly is a "male box"?

[niall]

wall box with a male XLR connection, the 6 of them daisy-chained back to the distros.

[alhaynes]

I thought so. Does the splitter created at one of the middle "male boxes" create any problems from interference due to the unconnected portion towards the stage?

An example of this (http://www.dmx512.com/web/light/dmx512/term.htm) where the "male box" is the twofer is:

[niall]

not that i've come across so far, tho it's not been tested in a gig situation yet.

edit: this problem did cross my mind, but when it worked fine in the intial testing i didnt see the need to make up any terminators.

[alhaynes]

Doubt terminators would help - it might even make things worse if used on a branch. The general theory of RS-485 (the serial protocol used to drive DMX) is that a bus (with short banches) should be used, terminated at each end (note: one end is inside the lighting desk). In the highly unlikely event that they was no termintor in the desk, then the stage end should have a termintor.

See http://www.maxim-ic.com/appnotes.cfm?ap ... 3&CMP=WP-1 for examples of wiring RS-485 (and DMX).

[Flaps]

Hi Niall,

It has been an awful long time. I have only just found this site again.

Just out of interest, was the demux box you used optically isolated on all outputs? what make did you go for?

Also sorry to bore you but y splits, which effectively your male boxes have the potential to be are not really the best idea as there is the chance that you will get reflection on the line. Also ensure that if you do terminate the line (i would) only have one of two ends terminated, i.e on of two ends of the Y otherwise so will run into all sorts of problems.

In and ideal world, each splitter would be opto isolated of course.

[niall]

hey, i cant remember the exact model, the venue have had it for ages, tis a pretty standard DMX to D54 converter. The two Distros are opto-isolated, that was something i made sure of!

I realised about the potential of bounce back, but *so far* everything is alright. I'll prolly make up a couple terminators just incase something goes wrong!

[alhaynes]

A simple (-ish) fix could be done as follows:

At each male box split the through connection leaving the socket attached only to the DeMux end of the line. Then connect an XLR plug (which would fit the socket) to the other part of the wire. Then there should be one socket to plug the lighting desk into, and one loose end of DMX cable (with plug), which can be plugged in when used in pass-through mode.

Alternatively connect a switch (triple-pole double-throw) to switch between the stage-ward end and the socket. The common part, is of course soldered to the DeMux end of the wire.

Either will avoid the use of a y-connector (but should provide the same functioality).

I know the first of these is fully acceptable, but I am not totally sure about the latter (which would look more professional). The principle of either is to isolate the stage-ward (unused) portion of the DMX-cable, and in both cases ensure that a lighting desk cannot be used (at the male box) without this portion being isolated.

[Flaps]

of course the cleanest option would have to be to get a console which distributes over a network, then you can have multiple input points without any of this hassle!!

Flaps