building a lighting console

[Junior8]

This thread has set me thinking. Has there ever been a time since, say, tracker operated Liquid Dimming, when building your own control system (including dimmers) would have ended up £ for £ equivalent cheaper than a purchased control system offering the same features? I only ask because I was amazed to find looking at some old catalogues today that my namesake cost £58 in 1970 (equivalent to £672 today) for 4 500/1000w dimmers and eight 5 amp outlets!

[Matthew Robinson]

Like most things in electronics, this can be split into the following stages:

 

Input

To input from a series of keypresses/faders, you'll probably require a system such as the one used in a keyboard: ie, a processor in the keyboard scans the keys and sends a scan code to the computer, which then inteprets in and converts to ASCII, etc. Have you considered butchering a keyboard by removing all the keys and wiring all the new keys to the place the old keys were connected to? Then, you can simply just use a key-mapping process on the computer to make each key do a certain operation. Don't use the number pad keys though- you'll probably want those for an actual number pad! Key mapping should be quite simple for a reasonable programmer. In terms of faders, you'll want pots attached to ADCs. Each ADC writes to a memory location, which is then read in as the memory location data changes or at specific time intervals.

 

Process

As you build your I/O devices, you'll write DLLs for windows or equivalent for other operating systems. Programming this should simply be a case of patching a series of values created either by your input buttons & faders; number keys; stacks; etc to the output. Don't forget, you'll probably want to be able to store loads of data about patch, stack, etc so this will probably be the hardest area for you. This has already been covered, so not a lot more for me to say, other than to consider your input method: do you want a command line; or push faders, click keys, programmed method of operation?

 

Output

Personally, I would be inclined to write all of the values you create for the DMX values into RAM. Then, a chip deals with all those values and spits them out as DMX.

 

Alternatively, you could build yourself a preset desk, whereby you just have n faders connected through n ADCs to n registers, then go for the chip discussed above that spits out DMX.

 

I've put quite a bit of thought into this, because I'd like to do this myself later in my degree, so please keep us informed of your progress!

[swampman]

Couple of notes:

Matthew.: "input issues" careful with keyboard controller chips: I have neve managed to rebuild a keybord (that worked before I fiddled with it!) that works again: the chips used seem to be very sensitive to static or whatever OR the carbon etc crosspoints (keypress points) get damaged. I stand to be corrected but that is my experience. A dedicated multiplexor would be better and seems the way most desks go.

 

Junior8: the classic problem of Return On Investment (the religion of accountants, managers and specialist in take-overs...) these are student cases (?) where they will derive a lot from the work as personal profit.

Dimmers NO: dangerous mains voltages, chokes etc were never really worth making not even when you might have considered winding your own resistors!!.

 

Controls yes: special applications especially if within the analog realm, but they are now becoming quickly redundant as colours are determined by levels (LEDs) and you need consistency from scene to scene. A custom built 20*10+4 channel with a pin matrix (memory) board with +&- 10v output (! ! ! ) I used to use for corporate events & some rock events is now a wall fixture talking/museum piece. Sniff ! Sniff ! It would never have been profitable (R.O.I. in the short term!!) for us to buy a desk at the time (late 1980s-early 90s) where Stonex and Avolites still ruled the pin matrix scene (or a memory desk at 10K+UKP) and Pulsar etc with dip switches was not sufficient for what we needed in one unit.

 

Others: as mentioned cutting aluminium for sliders etc: try laser/water jet CAD/CAM prototyping companies - we used a signwriters skills for the desk!

PC modules: use inbedded Linux, FreeDOS or better a student/academic licence for a real critical time Op. system like Wind River VXWorks would give you experience that is industrially useful. Microsoft seems too closed shop to bother about!.

Another idea: SIemens etc, industrial and domotic microcontrollers: plenty of different designs and programming is done from PC platforms with easy debugging tools.

[dbuckley]

Analogue boards were easy! Regulated 10v rail and lots of sliders for levels, DMX desks need much more thought and work and can be bought for a lot less that the parts cost.

And thats the thing.

 

Twenty years ago this would be a fun project, and thirty years ago home-made boards were quite popular. But back then a reasonable electronics hobbyist could build something that wouldn't disgrace itself.

 

But now we have DMX512 and computers, and everythging has changed...

[beware]

If you want to go the PC/Embedded programming route with little to no experience, I'd suggest something like TCL/TK (www.tcl.tk). It's easy to learn, extendable, embeddable, and also Tk is an easy to use user interface builder should you require on screen display.

 

I would think you could get a thin-client (very small, basic PC) with linux or windows XP embedded, install TCL/TK on it, add a USB/MIDI fader panel (http://www.dv247.com/computer-hardware/korg-nanokontrol--52990 or http://www.thomann.de/gb/behringer_bcf_2000.htm) or similar product, and some sort of DMX out, plug it all together and away you go. To make it cooler, take all the gubbins out of the separate boxes and put it all in one.

 

That's how I'd do it...

 

A concurrent post has been automatically merged from this point on.

 

total cost £250-300 ish.

[Brian]

If we're not careful this topic is just going to become a list of everyone's favourite programming language. The bottom line is that the actual implementation is irrelevant and trivial. Until the OP can fully describe what he wants to achieve and how it will work, in effect by writing the operations manual, the rest is moot. And once that manual is written the actual software part is trivial.

[GridGirl]

Twenty years ago this would be a fun project, and thirty years ago home-made boards were quite popular. But back then a reasonable electronics hobbyist could build something that wouldn't disgrace itself.

 

But now we have DMX512 and computers, and everythging has changed...

 

There's a 9-channel, 2-preset serial desk and dimmer pack sitting in the garage at my parents' place. Dad built it over 30 years ago (it's older than I am!) based on an old Strand dimmer module. The dimmer is seriously heavy - but fully shielded so it doesn't cause hum in sound systems - but it still works perfectly (it was well built in the first place and has been meticulously maintained) and still gets hauled out occasionally - I've used it on small school productions and the odd church production - plus the desk has a built in lamp and there's a telephone circuit as well so you have two-station comms! But the cable that runs between desk and dimmer is about the same size as a standard 32amp 3-phase cable with a multi-pin connector (looks a bit like a Harting) on each end...how things change!

[Matthew Robinson]

TBH, having re-read the OP, I think that we're taking this a bit over the top: what I think the OP wants is 10 pots with 10 buttons connected in such a way as to short circuit the pot. Then the output of each pot can be taken to an ADC and stored in a register, which is then read by a circuit which encodes values in the registers to DMX, which should be reasonably simple. Hopefully.

[paulears]

... but isn't this just re-inventing the wheel. Apart from the knowledge you gain from constructing the device, in terms of soldering and perhaps pcb construction - why would you wish to do it? The end product would be rather limited in what you could do with it? A 40 year old analogue design would appear to offer more?

 

As said further up the topic, what is important would seem to be features and benefits - as in what can it do for you? If the features are limited, then so will the benefits be?

[beninjam]

How about looking up Arduino (www.arduino.cc)? They're lovely little, easy to program microchips which are quite cheap, expandable and there is a DMX library for the software on the internet. I was also thinking about a little project myself/.

[JimWebber]

Surely if this is to be a "Real" project, we are all putting the cart before the horse.

 

As has been mentioned earlier, the first stage of any design process is deciding what you want said device to do.

 

Once you have your wishlist of exactly you want this "thing" to do, what features it will have, and its specifications, then and only then can you start thinking about HOW you are going to go about achieving these specifications.

 

And if you realise you do not have the skills/knowledge/experience/parts/money to fulfil all of your design spec, then you can look at revising your design and making any compromises...

 

As for programming PICs, do not come to me, the last thing I programmed was a SZ80, and that was twenty years ago!!!!!

 

Jim

[Matthew Robinson]

Quite :) but I've been thinking about the future of lighting desks.

 

Do you think there will be a market for a desk that can automatically map an area to be lit, and where each fixture either is focussed to (in the case of a generic fixture) or the limits of travel (for intelligent lights)?

[timmeh2]

Quite, but I've been thinking about the future of lighting desks. Do you think there will be a market for a desk that can automatically map an area to be lit, and where each fixture either is focussed to (in the case of a generic fixture) or the limits of travel (for intelligent lights)?

 

Hi

 

MA lighting desks already do this. If you define the size of your stage and position of every light and pieces of set and or/people, it can tell you what angle to put your generics at and will also work out pan, tilt and focus for your movers and point them at the object you want to light. Its a very powerful feature and is perfect for blind or on-the-fly programming.

 

All the best

Timmeh