PHOSPHENE'S ON LINE BEGINNERS GUIDES ARE REGULARLY REVISED AND EXPANDED. BOOKMARK THIS SITE AND USE IT AS AN INFORMATION RESOURCE.
IF YOU HAVE SPECIFIC QUESTIONS JUST EMAIL ME AND I PROMISE TO GET BACK TO YOU PERSONALLY.
If you are a teacher or running a site that offers information you may distribute and publish the articles on our site free of charge. BUT you must publish the page in its entirety and unedited, and MUST include a note that the article originaly appeared on PHOSPHENE's site and have a live hyperlink back to this site. www.phosphene.freeserve.co.uk Please also drop us an email to say that you have done this and letting us have the url of the site the link is on so we can check that it is working.
A page of explanation about DMX, what it is, what it does, and how to avoid the worst of the problems that beset it.
DMX is a protocol that enables the control of dimmers and 'intelligent' fixtures (they are generally the things with waggly mirrors and the like on them) to be done down a simple piece of 2 core and screen cable.
In the earlier days of stage lighting control each dimmer had to have a separate bit of cable sending it information on whether to be on or off (or something in between).
Usually 0V down this control cable resulted in 'off' with +10V causing full on, and various levels in between.
There was an exception to this in the case of Strand Electric who worked on a -10V (neg 10V) system.
Anyway you needed a lot of connections between the desk and the dimmer racks.
This made for big hanks of multicore cable where more than a few dozen dimmers were involved.
DMX has simplified all this cabling.....but of course it has introduced a few problems as well.
Let's see if we can de-mistify it a bit!
Here's a simple analogy, and one of the best I have come across, to explain the basic principle behind DMX.
Think of the cable as a road. Each 'bit' of information is like a bus going along the road. Each piece of DMX controlled equipment, intelligent lantern, dimmer rack, etc. is like a man waiting at the bus stop.
When the bus comes along the man looks at the front of the bus to see if it is the number he wants (That is his 'address'). (In this analogous world the busses come regularly and in numerical order!) If it is not the right number/address he ignores it. If it is 'his' bus the waiting man (instead of getting on) counts the number of passengers in the bus.
The quantity of 'passengers' give the man a value. Based on that value he does certain actions. For example, if he were a dimmer a value of 0 would mean he did nothing (stayed off) a value of 10 might mean he faded up to full.
The bus passes on, with its contents un-affected, and is either counted (gives information to other waiting units with the same address) or ignored along the whole length of the road.
The address I refer to above is usually taken to mean the 'start' address of a piece of equipment. For example a dimmer rack might be set to address '1', but because it has 6 channels it will read information from the "busses" numbered 1 to 6.
(Usually this will mean that the next dimmer rack has a start address of 7 (not 2!!).
In the case of 'intelligent' lighting the number of addresses used by each piece of gear will vary, but it is not uncommon for one 'fixture' to use half a dozen or more addresses. (These might control: on/off, colour, gobo, iris setting, Y axis mirror position, X axis mirror position etc.)
The address of the piece of equipment will be set in various different ways depending on the manufacturer. Zero 88's Betapack 2, for example, has a very sensible set of 3 rotary switches, (hidden under a cover-plate in the top right hand corner of the front panel) on which you dial up any start address you want between 1 and 512.
However most manufacturers will have resorted to the 'dip' switch form of address setting. This does seem to cause confusion for some people. Don't panic! The address is set by selecting combinations of the little on/off switches. If you are lucky they may be numbered. If not here's how it goes:
From the left: Switch 1 = 1
Switch 2 = 2
Switch 3 = 4
Switch 4 = 8
Switch 5 = 16
(you see the mathematical progression?)
To set (say) 12 as a start address you switch on switches 3 & 4 (they add up to 12!)
Beware, some gear will then use the next number up as the start address (13 in the example we've just given)
Several pieces of equipment can share the same address. For example you might want to have all your 'intelligent' fixtures doing the same thing at the same time throughout the show.
Setting tham all to the same address will save you lots of desk channels, because you can drive them all simultaneously from the same faders.
Minor warning about waggly lights that have come from the disco world. Some manufacturers, (NJD is an example) realise that not everyone has DMX control desks. They therefore build the kit to accept control from DMX or Analogue or a built in memory/microphone.
This is very useful. The NJD Datamoon is a splendid example of this sort of flexibility. Unfortunately it involves you in the potential for address switching that is unique to the particular device.
(The Datamoon will run random patterns if you turn all the switches off. It will be sound active if you make its address 65, it responds to analogue signals if its address is 16 or 32 etc. etc)
Now what about the cables?
You will read that ordinary twin core screen mic cable will not work. In actual fact it probably will on very small scale shows.
The USITT DMX512 control convention, the industry standard, is for a 5 pin XLR connector, with the cable cores wired: Pin 1 = signal common/screen, Pin 2 = Data-, Pin 3 = Data+.
The remaining two pins (4 and 5) are reserved for an optional second data link that most of us will never use. However, be warned, some equipment uses these pins to return information to the desk from the dimmer racks etc. ("its' all right, I am a happy dimmer" sort of thing)
Unfortunately one or two manufacturers have decided to ignore the industry standard and use these two 'spare' pins for sending supply voltages. At the small scale end of the market this unpleasant idea is in use with Pulsar and its associated companies.
Be very wary of connecting any DMX desks to anything with the Pulsar/Horizon/Clay Parky name on unless you are sure. One good way round their attempt to blow up your desk is to make sure you are only using pins 1, 2 and 3. DIsconnect pins 4 & 5!
Generally the desk will have the female 5 pin XLR connector. (the female connector is always the one with the hole(s), the male connector has the pin(s) that go in the holes...work it out for yourself..if we explain it any more the page will get barred by family filters!)
Dimmer racks by reputable manufacturers will have the male version of this connector for input, and another female connector for output to enable you to daisy chain on to the next piece of equipment.
Never split or loop the DMX control path. When you get to the last 'fixture' put a termination plug in the empty DMX output socket.
Most equipment will claim to be 'self terminating' but use a termination plug anyway. Strange half-hearted attempts to waggle (on the part of mirrors) and dimmers that fail to go right out, or refuse to turn fully on, are symptomatic of an improperly terminated DMX line.
The termination plug is really simple, just a 5 pin XLR plug with a 150 ohm resistor soldered between pins 2 and 3.
Without this the DMX signal (which is running at really really high frequencies), bounces off the end of the cable and reflects back along the wire. This is a problem familiar to those of you dealing with radio transmitters.
Regrettably, while all the sensible manufacturers have stuck to the general convention for the DMX signal and how it works, there is less of a consensus on the type of connector.
Let me repeat: the connector for DMX is a 5 pin XLR.
That said: Ryger uses a 3.5mm stereo jack to control Tiny Scan (Oh Yuk!)
NJD uses 1/4 inch stereo jack to control Datamoon (Nearly as Yuk)
NJD uses both RJ45 (that's something like a telephone plug) or 3 pin XLR (standard mic lead type) for Spectre.
This 3 pin XLR crops up on Intellabeams (a product of a quality that ought to know better!) etc.etc.
Even more unfortunately there is no consensus on the pin configurations for these rogue connectors.
Stereo jack plugs, where used, generally seem to be wired Tip = Data+, Ring = Data-, Sleeve = Screen.
3 pin XLR however turns up with a variety of pin arrangements.
As a hire company Phosphene will always send DMX equipment out with adaptors to restore the DMX arrangements to industry standard from whatever abberation the manufacturers have had.
Bear in mind though, that the vaguaries of pin configuration could result in strange effects if the wrong adaptor ends up in the wrong gadget!
Each of the 'addresses' I mentioned above will relate to a fader on your control desk. Usually this will mean fader 1 controls address 1, fader 2 address 2 and so on. It is possible on most DMX desks to alter this arrangement by changing the DMX patch.
It is rare on small scale shows for there to be any real purpose in this. As when you have done it there is unlikely to be any external evidence that things have changed all you will usually achieve is berwilderment.
Operators pushing up fader channel 1 usually expect it to bring up address 1, not address 46!
Generally speaking, in very small scale work, you will end up running your dimmers off the first few channels/addresses of the desk and your 'intelligent' fixtures from later channels or vice versa.
A desk like Zero 88's XLS allows you to run 24 dimmer channels and a similar number of what they call 'aux colour control' channels via a simple keypad. Zero 88's Sirius 250 and 500 desks and the subsequent developments are arranged so that the first half of the desk is intended for 'intelligent' fixtures and the latter half for generics and simple dimmer channels.
Desks like these will allow you to use just one 'channel' to control all the various 'atributes' (usually colour, gobo, position etc.) of a particular fixture. We can't teach you, here, how to drive every desk on the market, as there are just too many idiosyncratic quirks that apply to specific manufacturers.
Here's a few notes on the terminology that you'll run across in the 'operator's handbooks' however.
I mentioned quirks. You'd not really expect too much oddball stuff in terms of the actual electronics of the DMX system by now. After all DMX has been the de-facto control convention for most stage lighting for over a decade. (DMX stands for Digital MultipleX). We've mentioned Pulsar's abberations, they call their convention PMX by the way, and we should mention Strand's similar (but different) weird system called AMX.
Within 'proper' DMX you will find some legitimate variations however. For example there's a distinction between 8 bit and 16 bits. The latter is achieved when 2 channels of information are stuck together to give a more precise control of moving fixtures. An 8 bit mode gives 256 steps of information which produces a rather jerky motion on mirrors and moving heads. (As a complete circle is 360 degrees, 256 steps result in over a degree of movement per step.) Due to another of those geometric type sums using 16 bits gives 256 times 256 steps (over 65000 steps) to a 360 degree rotation. (That's going to be accurate enough for all of us!) You won't need to actually do anything to benefit from this... the manufacturers have already made the decisions for you.
You should be aware of the different ways in which desks handle and dole out information though. Generally the generic channels will be what is known as HTP. HTP stands for 'highest takes precedence'. You'll intuitively understand this if you background is in 2 or more preset desks. If you set a channel to, say, point 5 on one preset and at point 7 on another preset... then shove up both master faders, you'll get the channel output at point 7. Desks treat generic channels like this although they're sending out the output of memories rather than presets. Movement 'atributes' are regularly treated on a LTP basis. LTP stands for 'latest takes precedence'. Since there's not really an 'off' or 'on' for a position this approach makes sense. What happens in practice is that a mirror, or moving head goes to a position... and stays there until some different position is sent to it. This saves a great deal of unneccesary (and probably noisy) waggling about in the rig. But do be aware of this different approach... it probably means your moving lights will stay where they last were when you fade to blackout, for example.
If you are looking into buying a desk to drive both dimmer channels and moving lights you need to ask a few questions. Chief among these will be do you like the style and operating methods of the particular manufacturer. (If you've read much of this site you'll realise that we like Zero 88's general operating style... although we challenge some of the more recent forays into the bigger worlds, with the blithe assumption that the only buyers will be Mac and High End users...(let's ignore the poor saps floundering around with low budget stuff)... and that we really hate certain other brands. You may have different views.) You must get this right for you when you buy, you're going to be spending a lot of time behind this desk. But do check also that it has the capacity to do all the chases and so on you may want to achieve.
This guide page brings in more supplementary questions than any other that we run.
Any specific questions email me and we'll get back to you with answers/advice.
| 
