DMX pyro

[mrboo]

If your going to run a cable for the safey system you need in place your better just using bell wire as it cost little and use the system most people use. you need line of sight so need to be not far from your pryos anyway. Sometimes it gets messy trying to use DMX like with Lasers. if people wanted it on DMX it would of been done by now.

[Superpants]

If you choose to go ahead with the project you will obviously have a much better understanding of the issues that you may encounter in trying to make a system safe from this thread. Early on in your project I would suggest that it would be a good idea to produce an FMEA. If this is carried out thoroughly the scores gained from this should give you a clear view of where the design work needs to focus.

 

If design engineering is an area you wish to get into after University, familiarity with the FMEA process will put you in a good position with future employers, as will the skills you pick up in giving weight to the various areas of the project and the concerns that are raised. In my experience this is not something that is well covered in university courses and is an area we spend quite a lot of time and effort with new graduates working through the processes as it helps guide their designs. In many cases in our line of business (automotive)it is required as a project deliverable by our customers, as well as to satisfy ourselves that the design is thorough and robust and can drive a lot of the product testing requirements.

 

 

 

[itiba]

There is currently a west end show which uses their lighting desk to trigger pyro cues among other things, not too sure if I'm allowed to name names but the system in place is as follows:

DSM calls "standby pyro inhibit"

LX Op raises an inhibit fader on the desk

DSM calls "standby pyro dead mans"

An ASM on either side of the pros with direct line of sight of the pyros holds in a safety switch. if either of these switches are not held in then the pyro will not fire

DSM calls "LX101 and pyro cue GO"

LX op hits 'go' and this fires the pyro as well.

The inhibit fader automatically descends

[Brian]

As long as DMX is only commanding when to fire, not if it is safe to fire, then there is no safety issue with using DMX.

Exactly.

 

There are commercial systems out there that take in SMPTE timecode and if you thought DMX was unreliable you ain't seen nothing *but* the SMPTE is only used to time the operation, the pyro operator has already made the decision to arm the unit.

 

You need line of sight so need to be not far from your pryos anyway.

For small shows maybe. We regularly have devices on the end of a 300m cable run simply because of where we want devices.

 

 

 

As someone who designs pyro control systems I can categorically state that you do not want to use DMX from your controller out to your firing boxes. It's the wrong system for the job. By all means use some sort of serial control link but you need to put some thought into what that is and what features it needs.

[kitlane]

This makes interesting reading. (same link I posted on the fireworks thread). A description of a pyro problem at Treasure Island in Las Vegas

[Brian]

This makes interesting reading.

That piece is about the only written report about a pyro incident with any details. The key thing is that it required two faults to occur and relied upon the way some companies, particularly US ones, design their systems. What it does illustrate nicely is how things can conspire against you, which in pyroland can be A Bad Thing ©.

 

It is quite possible to control pyro over serial links...

...you just have to take a number of precautions to make sure things go off at the right time. This means taking a layered approach to system integrity/security.

 

 

 

The photo shows the prototype sitting on my bench of a system capable of firing over 3,000 devices.

[OllieDuff]

I guess as an exercise in designing a safe and effective device then it sounds like an interesting project and as other people have posted, a good chance to demonstrate a good understanding of what you're doing.

 

Re-reading some of the posts above, I think there might have been a misunderstanding about what you meant by three channels - am I correct in thinking you're proposing to use 3 DMX addresses per effect, and the effect will only fire if the correct DMX value is set on all three channels?

 

If I could make one small criticism, it's that "use a separate DMX universe for pyro" isn't really a safety feature that you can dictate as a designer/manufacturer; that would be up to the end user to decide if they want to do that. My reasons for this are:

 

1. You can't guarantee that users will actually use a separate universe. They might be running on el cheapo lighting desk which only outputs one universe.

 

2. In a modern desk with multiple universes and soft patching, there's no fundamental reason why it's harder to accidentally plot in addresses from universe 2 as compared to universe 1; you could patch 1/001 into desk channel 1 and 2/001 into desk channel 2. I suppose there is some advantage in that by having a dedicated cable run for your pyro, you reduce the chance of nastiness from the rest of the rig, but in practice, has anyone ever heard of a dimmer or intelligent luminaire messing with the DMX signal thru? I'm just not sure what specific mode of negligent firing you're trying to mitigate by using a separate universe.

 

The dead man's handle is also an issue from a design point of view; if you have to have a DMH for each channel, you may as well have an off-the-shelf Lemaitre six way so that you can select, arm and fire channels locally. At the risk of going all engineer-ey, it all depends on the problem statement for your project; if the purpose of the project is "design a safe DMX-controlled pyro firing system" then yes, this is potentially a good project to do with the constraints that you describe. If the purpose is "find a safe way of controlling pyro for live performance", then the solution is buy something off the shelf or pay an experienced professional to build something custom, and you really don't want to round off your write up with "I've designed this thing, but in the real world you'd buy something different". I suspect that you can probably write the project brief in such a way that the thing you want to build fulfils the original specification, but keep in mind that in the real world, you start with a market need and design a product to meet it, and that the person marking it might well ask why you would ever use your product over currently commercially-available products. Of course, you might well have reasons why your product is superior in certain ways but as others have said, if it doesn't exist already there's probably a reason (violation of the DMX standard being chief among them).

 

Agree with others that you want as short a data run as possible to your receiver/DMH and then longer runs of power to the effects themselves (with due consideration of course to voltage drop but that's easier to mitigate than possibility of random EMI messing with your control).

[Brian]

Agree with others that you want as short a data run as possible to your receiver/DMH and then longer runs of power to the effects themselves (with due consideration of course to voltage drop but that's easier to mitigate than possibility of random EMI messing with your control).

I have to disagree with that statement. Let your data-link do the long runs and keep your two-core short.

[Sam Jelfs]

Sounds arrogant of me, but there is a dissertation that should be in the KR library on the design / build of a pyro controller using a CANBus interface for the data transmission... not quite what you are looking at but there is some info in the literature section on the use of DMX / other protocols... not the best thing I've ever written, but the thing made stuff go bang :D

[dbuckley]

I have to disagree with that statement. Let your data-link do the long runs and keep your two-core short.

 

I absolutely second that.

 

Assuming the power side is not moronically designed, it also significantly reduces the chances of the sorts of problems that the Birket paper describes.

 

I'd have thought that it would be cost and space effective to have the necessary electrics and electronics in the base of a pyro pot holder.

 

I've deleted my thoughts on how to do this conveniently and safely as some numpty will go and do it and get it wrong and it'll all be my fault!

[kitlane]

Sounds arrogant of me, but there is a dissertation that should be in the KR library on the design / build of a pyro controller using a CANBus interface for the data transmission... not quite what you are looking at but there is some info in the literature section on the use of DMX / other protocols... not the best thing I've ever written, but the thing made stuff go bang :D

 

They don't keep dissertations in the library any more. However, I have a copy of yours on the shelf in my office.

[Sam Jelfs]

They don't keep dissertations in the library any more. However, I have a copy of yours on the shelf in my office.

 

Do they not, I think that's a real shame, I found it very useful to go and have a flick through previous students work whilst writing mine to give an idea of what is a good and what isn't a good dissertation. I'm sure I still have a pdf of mine on disk somewhere, along with the actual device in a box in the attic.

[preselipete]

Not sure I agree with the statement "the pyro fires at the wrong time.... is not a safety issue"!!!

[kitlane]

Not sure I agree with the statement "the pyro fires at the wrong time.... is not a safety issue"!!!

 

But you have quoted this out of context. The principle that was being discussed was having a human operator enabling a window of opportunity for the automated firing of the pyro. As long as it is safe to fire the pyro (because the area is clear or whatever other criteria are being applied) then if it is fired at the wrong time it is an artistic but not a safety problem.

[LeeT]

I think it's the safey issue more than anything - You need someone who has a complete visable view of the pyro, the actors, audience, and anything that could cause harm to anything/anyone.

If your using a Deadmans switch, presuming that it would be operated by a third party, what would be the need for a DMX Controlled fire, when you may as well, just use the main link pyro fire?

 

I wouldn't attempt to use a DMX, however easy it would be. Safety, should be the first prioritty...