Thoughts on this light... World's smallest RGB DMX light

Let's be blunt here.
DMX was designed in the days when it went from a console to a dimmer rack. Neither of those are terribly lean on real estate to accomodate XLR5 connectors...
There is no valid argument for using XLR3s but there is a very valid argument to use an alternate connector where the use of XLR5s would mean DOUBLING the size of the instrument.
Sure, Darklight could go and use a LEMO or the like for the input and the output, but at that point the connectros alone will bump the price up to like 4x it's current.

Strictly speaking, if Darklight supplied an adapter of XLR5 to 1/8" TRS, the unit would be compliant...

Maybe the standards committee needs to address the issue of small form factor dmx connectors more comprehensively in the next edition of the standard... 5 pin TiniQ does exist...
 
Let's be blunt here.
DMX was designed in the days when it went from a console to a dimmer rack. Neither of those are terribly lean on real estate to accomodate XLR5 connectors...
There is no valid argument for using XLR3s but there is a very valid argument to use an alternate connector where the use of XLR5s would mean DOUBLING the size of the instrument.
Sure, Darklight could go and use a LEMO or the like for the input and the output, but at that point the connectros alone will bump the price up to like 4x it's current.

Strictly speaking, if Darklight supplied an adapter of XLR5 to 1/8" TRS, the unit would be compliant...

Maybe the standards committee needs to address the issue of small form factor dmx connectors more comprehensively in the next edition of the standard... 5 pin TiniQ does exist...

Chris, your points are exactly what we considered during development. It is impractical for us to fit a 5 pin XLR into our system (let alone two: one in and one out) as the cable headers will outweigh the light fixture.
 
- I hope I won't be strung up for this one, playing devils' advocate here -

Though on the one hand their wiring scheme does look really bad (and lets not even talk about the cabling recommendation), and gave me the same close to alergic reaction as most people above.
The Y-splits here are actually more like the 'normal' EIA-485 bus since every Y serves 1 fixture (which is in essence what also happens inside of any fixture), this is of course on the condition that you use no more then 1 fixture per Y and have 1 'bus' and that the tail from the fixture is not too long (no longer then 3 feet/1m IIRC).

Obviously obeying standards is important and situations with lots of different plugs and polarities for the same signal are just annoying for the users but DMX-512 does allow other plugs when the fixture is unable to accommodate full-size XLR plugs due to size restrictions.

Keeper,

The reason for using the TRS plug isn't necessarily for customers to go and use audio cable extensions and splitters to hook up the system (although practically speaking, they can up to certain distances --- as demonstrated in the lab video above). The real reason behind using TRS is simply to provide a plug and play adapter to the bare wire pigtails. This way if you want to swap out a fixture, the cabling (both power and signal) can remain intact and you can simply unplug the light. This feature is coherent with our entire product line.

Our official recommendation is using a single DMX bus and tapping the fixture signal wires into the bus by hard wiring:
proxy.php


There are many RGB fixtures out there without a pass-thru connection, and the many fixtures that DO have a DMX pass-thru are directly connecting each pin of the DMX out signal to the DMX in signal. You're right --- it's essentially what we're doing in the above diagram except the tapping in point is within the fixture.
 
Keeper,

The reason for using the TRS plug isn't necessarily for customers to go and use audio cable extensions and splitters to hook up the system (although practically speaking, they can up to certain distances --- as demonstrated in the lab video above). The real reason behind using TRS is simply to provide a plug and play adapter to the bare wire pigtails. This way if you want to swap out a fixture, the cabling (both power and signal) can remain intact and you can simply unplug the light. This feature is coherent with our entire product line.

Our official recommendation is using a single DMX bus and tapping the fixture signal wires into the bus by hard wiring:
proxy.php


There are many RGB fixtures out there without a pass-thru connection, and the many fixtures that DO have a DMX pass-thru are directly connecting each pin of the DMX out signal to the DMX in signal. You're right --- it's essentially what we're doing in the above diagram except the tapping in point is within the fixture.

My suggestion to avoid problems is to make up a cable that is part of your light that has both the power connection AND the In and out TRS connection, that way you control that part of the system, and them make SURE your customers use trs properly wired connectors. this would eliminate the customer getting some sort of Y adaptor Where difficulty is going to arise is IF you customer decides to use a dual 1/8 to single trs adaptor that is designed for two TS (MONO JACKS) then you are likely to short out the dmx connections. I agree if you look inside a lot of dmx fixtures in fact many of them (on the low end) simply loop the xlr ins and outs. Obviously you also want to make sure that there is NO way for the customer to plug the power 12v into the dmx connection

Sharyn
 
If there's space, I'd suggest a pair of RJ45 connectors on the fixture (or one and a 2-way splitter), for both data and power. CAT5 is at least 120R twisted pair, and this is commonly done in the animated (Christmas) lighting world.
/mike
 
Thanks Sharyn and Mike both for your input. RJ45 is an option we are considering for a future revision.

Mike --- when you suggested the dual RJ45 ports, you said to use them for both signal AND power? I assume you mean putting the power on pins 4 and 5 of the RJ-45? In terms of space, there's absolutely NO SPACE for RJ45 headers on the light fixture itself --- the only possibility is a dongle with those ports connected to the fixture cable. Cosmetically, it may look a bit awkward compared to our current design but I agree that it will at least follow DMX512 standard.

Does anyone see a potential problem assigning the undefined pins 4 and 5 of the RJ-45 cable to 12V power and ground? I guess there's the potential problem of a user accidentally taking our DMX RJ-45 cable (with power on pins 4/5) and accidentally plugging it into a computer or something...


Sharyn --- you're right about taking control of these potential variables. We will look into the option of a different cabling interface for the user.
 
It could happen but with a properly labeled unit that provides the power, it should be okay. you could have the power-supply/DMX receiver have the appropriate labels it should be perfectly fine since it would require them to take an Ethernet cable between the two. the most it could do is flip the internal breaker on the computers power supply. Since most computers run their networking devices off of 12 volt anyway.
 
It could happen but with a properly labeled unit that provides the power, it should be okay. you could have the power-supply/DMX receiver have the appropriate labels it should be perfectly fine since it would require them to take an Ethernet cable between the two. the most it could do is flip the internal breaker on the computers power supply. Since most computers run their networking devices off of 12 volt anyway.

The advantage of putting power inside the RJ45 cable is we now only have 1 jack to deal with instead of two. You could also potentially daisy chain the lights with a single cable that has both power and signal. However... the wires in the RJ45 cable isn't very low gauge --- over long runs and multiple fixtures, there will be significant voltage drop. So this connection method will be limited in distance and number of daisy chained fixtures.

Does anyone see a potential problem of the power on pins 4 and 5 causing interference with the DMX signal on the same cable?
 
The advantage of putting power inside the RJ45 cable is we now only have 1 jack to deal with instead of two. You could also potentially daisy chain the lights with a single cable that has both power and signal. However... the wires in the RJ45 cable isn't very low gauge --- over long runs and multiple fixtures, there will be significant voltage drop. So this connection method will be limited in distance and number of daisy chained fixtures.

Does anyone see a potential problem of the power on pins 4 and 5 causing interference with the DMX signal on the same cable?

I think a lot of networking devices already do it, and they send a lot of data on there... Also I believe that if its twisted pair the interference becomes moot? Been a while since I took E&M...
 
I think a lot of networking devices already do it, and they send a lot of data on there... Also I believe that if its twisted pair the interference becomes moot? Been a while since I took E&M...

Can you offer an example of such devices? I guess they are using pins 4 and 5 for power and ground... but which is power and which is ground? If this is something to consider, we want to pick the pinout that is compatible with the greatest number of 3rd party devices... would hate to have it in reverse and have some 3rd party device blow a fuse.
 
Can you offer an example of such devices? I guess they are using pins 4 and 5 for power and ground... but which is power and which is ground? If this is something to consider, we want to pick the pinout that is compatible with the greatest number of 3rd party devices... would hate to have it in reverse and have some 3rd party device blow a fuse.

Quan if you go the RJ45 route, we can then easily use RJ45 terminators as well which makes it a little easier to terminate than your current line up.
Best,
Bill Rod.
 
Quan if you go the RJ45 route, we can then easily use RJ45 terminators as well which makes it a little easier to terminate than your current line up.
Best,
Bill Rod.

It would be nice to have power inside the RJ45 cable... i just found a spec called "Power over Ethernet" or PoE --- it requires 48V and only up to 15W, so we'll have to figure out how something like that can be used, this is likely a completely different system and setup than what we currently have.

For this current project, the devices will have to be 3.5mm TRS but I do have a backup solution using RJ45 cable (just for signal) if the signal doesn't seem reliable. I'll have that available to you as a backup option during installation.

I don't foresee any major issues with your install because:
we've verified over 60 fixtures to work on a single DMX universe over 900ft (the video only shows 32 fixtures at 500+ft). Your setup will be multiple isolated busses with much shorter runs and less fixtures per run. Our setup condition uses over 900ft of unsheilded 3.5mm extension cable wrapped in loops... these are pretty bad conditions, and if we can verify in this condition, the haunt shouldn't be very difficult.
 
It would be nice to have power inside the RJ45 cable... i just found a spec called "Power over Ethernet" or PoE --- it requires 48V and only up to 15W, so we'll have to figure out how something like that can be used, this is likely a completely different system and setup than what we currently have.

For this current project, the devices will have to be 3.5mm TRS but I do have a backup solution using RJ45 cable (just for signal) if the signal doesn't seem reliable. I'll have that available to you as a backup option during installation.

I don't foresee any major issues with your install because:
we've verified over 60 fixtures to work on a single DMX universe over 900ft (the video only shows 32 fixtures at 500+ft). Your setup will be multiple isolated busses with much shorter runs and less fixtures per run. Our setup condition uses over 900ft of unsheilded 3.5mm extension cable wrapped in loops... these are pretty bad conditions, and if we can verify in this condition, the haunt shouldn't be very difficult.

Oh I completely agree for this project we should be set with the TRS, I was just suggesting for future revisions on the light... :)
Best,
Bill Rod.
 
Our setup condition uses over 900ft of unsheilded 3.5mm extension cable wrapped in loops... these are pretty bad conditions, and if we can verify in this condition, the haunt shouldn't be very difficult.

If (which it sounds like it is) a short term project, yes, it will probably work out great. However, I would think seriously about the suggestions that have been thrown out in this thread. Regardless of connector type, the recommendation of using audio cable instead of data grade cable is, in my opinion, a bad idea for business because it will fail. Maybe not now. Maybe not in your shop test. Maybe not on this particular install. But it will when you get any significant number of hours of use in variety of venues. These units strike me as being used much more in a architectural/permanent install type of situation, where reliability is key, than in a theater where you generally have more access to the fixture and knowledgeable electricians around to fix it.

My $.02
-Tim
 
Well, I'm just going to say that I think the idea of a tiny, DMX controllable, hide-a-way RGB LED lighting unit is a great idea. I for one support companies who try new and creative ideas instead of just copying other makers models.

That said, I would like to see a more "professional" product in the future. I think the idea of combined 12v power and DMX signal for the next model is a good idea. (ColorKinetics uses a power supply and a 4 pin cable don't they? Why not go that route if one is already using a non standard, that is, not five-pin XLR DMX connector?)

thrilltainment, while I can see what you are trying to get at, the short and simple answer is that using a wye cable with DMX signal, instead of a signal splitter, is neither approved nor recommended, and is, no matter what the reasonings (be they cost, 'so n so does it' or 'well it works for me' ) a gamble.
 
Please excuse the crudeness of the drawing, but would everyone feel better about this type of cabling system? The only difference between this and what we currently have is we split the signal off for you instead of having you use audio splitters and that we're using RJ45 ports instead of audio ports. So this should be DMX512 compliant...

proxy.php


advantage: we're using DMX compliant cabling...
disadvantage: much bulkier tail
 
If (which it sounds like it is) a short term project, yes, it will probably work out great. However, I would think seriously about the suggestions that have been thrown out in this thread. Regardless of connector type, the recommendation of using audio cable instead of data grade cable is, in my opinion, a bad idea for business because it will fail. Maybe not now. Maybe not in your shop test. Maybe not on this particular install. But it will when you get any significant number of hours of use in variety of venues. These units strike me as being used much more in a architectural/permanent install type of situation, where reliability is key, than in a theater where you generally have more access to the fixture and knowledgeable electricians around to fix it.

My $.02
-Tim

We agree that audio cable is non-compliant and we would never recommend it for permanent installations either. Our system comes with a pigtail connector which splits off into 3 wires --- it is the user's decision on what type of cabling should be used for the install. We recommend data cable with hardwired connections to our 3 signal wires. The 3.5mm TRS form factor was intended for convenience of removing the fixture for temporary installations, it wasn't intended to guide the user into using audio cable for extensions (although there isn't anything stopping them). Some of our customers make custom shielded cables with 3.5mm adapters at the end.
 
Well, I'm just going to say that I think the idea of a tiny, DMX controllable, hide-a-way RGB LED lighting unit is a great idea. I for one support companies who try new and creative ideas instead of just copying other makers models.

That said, I would like to see a more "professional" product in the future. I think the idea of combined 12v power and DMX signal for the next model is a good idea. (ColorKinetics uses a power supply and a 4 pin cable don't they? Why not go that route if one is already using a non standard, that is, not five-pin XLR DMX connector?)

thrilltainment, while I can see what you are trying to get at, the short and simple answer is that using a wye cable with DMX signal, instead of a signal splitter, is neither approved nor recommended, and is, no matter what the reasonings (be they cost, 'so n so does it' or 'well it works for me' ) a gamble.

we will be working on higher end and higher output models that have some of the features mentioned in this thread --- I am grateful that Bill has posted this question and the forum is able to offer some insight into the needs of the lighting industry. We are a new company (established spring 2010) and have lots to learn, research, and develop. thanks for your valuable input! we'll put it to good use =)
 
Please excuse the crudeness of the drawing, but would everyone feel better about this type of cabling system? The only difference between this and what we currently have is we split the signal off for you instead of having you use audio splitters and that we're using RJ45 ports instead of audio ports. So this should be DMX512 compliant...

proxy.php


advantage: we're using DMX compliant cabling...
disadvantage: much bulkier tail


In reality the tail wouldn't be much larger that it already is. By the time you factor in the end user having to add in their own TRS adapters and splitters, and the power connections that are already there, providing one simple tail that has all of it integrated works out significantly easier for the end user. It will also be a much cleaner install.
 
I think it's great how thrilltainment is actively listening to the suggestions posted here, rather than endlessly defending his product. He could easily claim that it's "just fine the way it is". This is how great products are developed --- by listening to customers/potential users.
 

Users who are viewing this thread

Back