Please help me, my mind is blown :(

[daveymuso]

Hi

 

 

 

Here's a totally complicated issue I need to resolve.

 

 

My setup

cakewalk sonar om laptop ->midisport interface -> NJD MDC 128 midi to DMX convertor.

 

 

have used this setup very successfully for years with NJD spectres & scanners & effect from other companies --- doesn't lose a beat!!!

 

 

I just bought 3x LEDJ slimline 229 par cans. Great piece of kit --- pick us out nicely from the spectre wash!

 

My Issue

1. I isolated the system bytaking the controller out of the racks so now have only the laptop->midisport midi interface -> NJD MDC 128 -> 229s *3 fed by 1x 3m lead& 10 inch link leads. All neutrik/van damme mic cables tested with a propersoundlab meter. Can't get a simpler setup than that…

 

2. when I plug in, to 229 #3 dmxout, the neutrik DMX terminator they all become totally non responsive?? So all other tests are without this…

 

3. when I plug in any 2 unitsthey work fine – I've tried all combinations and orders with 3 units. Thesymptoms were moving around but still not working properly

 

4. I did some focused testing.The ranges usually go from 0-127 on the mdc controllers (this translates to the0-255 DMX). With three 229s in the circuit '0' becomes non responsive on theRGBA channels only. If I change them to 1 they begin to work better. If I thensend a 0 to RGBA two units start to strobe with colour scroll. I thought I canwork with this now I know the anomolies, but then after a few minutes ofplaying, #3 went into strobe & colour scroll (with macro control at 0)

 

 

5. Finally I took my ball homewhen I changed blue on 229 #3 @ channel 103 and it changed 229 #2 strobechannel 74. This happen the same every time. I now believe this to be somethingmuch deeper than interference.

 

 

 

I have never had any issues withthe NJD controller – driving 10 units all different makes etc. It never drops asingle beat across 20m of cable.

 

 

My thoughts are – how come thelights take 0 then when the third unit is added, it accepts 0 on brightness,strobe & macro, but not on red, green, blue or amber?? This leads me to believe it is not the controller as it is sending out the correct signals

 

 

 

Do you think there maybe a dodgy 229 –the trouble is I don't have another 229 to swap into the mix…

 

[ianknight]

It's a dodgy 229 - ask your supplier to send it back to ProLight for them to look at, it won't take more than a couple of days. We had issues with one killing the DMX stone dead (and it's PSU) when a DMX cable was plugged in - scary for the customer! - returned it to ProLight, back inside a week.

 

Don't go poking around inside unless you've seen inside one before - the cables are kept short and tidy internally to improve reliability.

[TomHoward]

Have you tested all three individually to make sure that they are all working?

 

What's the addressing like - are they all addressed individually? What happens if you address them all the same, do they then all respond correctly? What about if all three are at the address of the one behaving oddly - do they all behave oddly the same?

 

Most problems I've had with intelligent lighting setups is either the unit being one address out from the controller (although this happens more often with DIP switches) which causes perceived 'erratic' behaviour, or one of the units along the line generating DMX - gets knocked into a stand-alone mode and then generates DMX for others in 'slave'.

[daveymuso]

Have you tested all three individually to make sure that they are all working? Yes, unfortunately I spent 4 days programming my whole set but only ever had 2 inline -- this was because one of the original units was dead on arrival so had to send it back. All work absolutely fine alone

What's the addressing like - are they all addressed individually? Yes

What happens if you address them all the same, do they then all respond correctly? No, when I add #3, #2 & #3 go loopy. (even if I swap them around???)

What about if all three are at the address of the one behaving oddly - do they all behave oddly the same? No , just #2 & #3 every time

Most problems I've had with intelligent lighting setups is either the unit being one address out from the controller (although this happens more often with DIP switches) which causes perceived 'erratic' behaviour, This is all OK & Proven

or one of the units along the line generating DMX - gets knocked into a stand-alone mode and then generates DMX for others in 'slave'. It is almost like #3 starts generating DMX and it goes back down the cable to #2

 

It's a dodgy 229 - ask your supplier to send it back to ProLight for them to look at, it won't take more than a couple of days. We had issues with one killing the DMX stone dead (and it's PSU) when a DMX cable was plugged in - scary for the customer! - returned it to ProLight, back inside a week.

 

Don't go poking around inside unless you've seen inside one before - the cables are kept short and tidy internally to improve reliability.

 

 

the trouble is I bought them from CPC online. last month. It may be easier to return for refund & then buy another three from somewhere else...

When I bought them, one unit would respond at all in DMX mode --- I returned this for a replacement. This tells me that they may have a dodgy batch???

[TomHoward]

The cables are in parallel so it's not unlikely that it'd go the other way down the cable.

 

To prove it's the unit and not any addressing, try addressing all three units to the same (ie A100, or whatever the address of #1 is) and add them individually until you have problems. If it still throws up problems with all the setups the same it's probably a dodgy unit.

 

If it is a dodgy unit, you shouldn't have problems returning on to CPC for an exchange - they're probably the easiest of most online sellers for replacement of units. However, personally I'd pick up the phone as it's always quicker.

[daveymuso]

Thanks tom

 

 

 

Already tried on same address -- no different.

 

 

I have some interesting movement.

 

 

 

 

 

I managed to borrow a Transension SDC-06 6 channel simple DMX controller...

 

 

All works perfectly! Although it cannot supply the 7th channel which is for colour macro. It even works fine with the DMX terminator in. This proves that the issue is with the NJD MDC-128. This seems to work for a while as long as I don't send any 0 to the RGBA channels. However, I have to send 0 to strobe & macro and this is when I get my issue. Also the terminator stops the DMX working on this unit!!!

 

 

 

I think I need to start thinking about an alternative way of driving my fully programmed lite show (oh the man hours is gonna be crazy :() Unless anyone has any further suggestions?

[dirkenstein]

This proves that the issue is with the NJD MDC-128. This seems to work for a while as long as I don't send any 0 to the RGBA channels. However, I have to send 0 to strobe & macro and this is when I get my issue. Also the terminator stops the DMX working on this unit!!!

Definitely sounds suspicious. Does the terminator stop DMX working when you plug in any DMX fixtures other then the 229s, or just the 229s?

 

From the sound of it, you confirmed that your cables are wired corectly and don't have a 'break' on one side of the DMX.

 

Things to consider:

 

Is the RS485 line driver chip on the NJD faulty?

The symptoms sound an awful lot like the behaviour I once saw on a DMX line being driven 'single ended' with the - lead disconnected inside the lighting desk. Similar behaviour could be caused by a faulty rs485/DMX chip or by a broken/loose connection on one of the DMX pairs on the NJD. Or a defective cable (which I believe you've eliminated).

 

The easiest way to tell this might be to make one of the LED DMX tester terminators various people suggest on their websites, like this one from bigclive:

 

DMX Tester

 

If you're not getting DMX transmission on one of the two DMX balanced lines that LED won't light up.

 

What does the NJD actually output in the way of DMX when the input values are set to 0 for a DMX channel? Does it ouput a fixed size DMX frame or drop values beyond the 'end' of the currently active controller or note? You can't really tell without a DMX analyzer but this is beginning to sound like a candidate for one.

 

What addresses are all DMX devices in your light show on?

 

One possible thing to do is to set a DMX channel to 255 _beyond_ the address range for your 229s. This will _force_ the NJD to generate packets with 0s on all channels the 229s use but that aren't currently being set to any values by note-ons/control changes.

 

Are you using the NJD in midi note mode or control change mode?

 

I'm also wondering of you are having issues due to scaling from a 0-127 midi range to a 0-255 DMX range. I had a Lanta DMX washlight which set colour channels to audio controlled when red/green/blue brightness was set to 255. 0-254 was brighness. Resulted in the totally counterintuitive behaviour of the fixtures turning off when brightness was set to 100%.

 

Do the 229s have specific DMX values / value ranges the need to be set on certain channels to set diffeent modes?

[TomHoward]

I'm still not clear - when they are all on the same address (ie A001), and are the only thing on the chain, do any given two work without problems, but as soon as you add the third you get erratic behaviour?

[daveymuso]

I'm still not clear - when they are all on the same address (ie A001), and are the only thing on the chain, do any given two work without problems, but as soon as you add the third you get erratic behaviour?

 

Bang on... wierd isn't it. If the DMX numbers are correct for 1 or two, why should they change when adding a third??

 

This proves that the issue is with the NJD MDC-128. This seems to work for a while as long as I don't send any 0 to the RGBA channels. However, I have to send 0 to strobe & macro and this is when I get my issue. Also the terminator stops the DMX working on this unit!!!

Definitely sounds suspicious. Does the terminator stop DMX working when you plug in any DMX fixtures other then the 229s, or just the 229s?

 

From the sound of it, you confirmed that your cables are wired corectly and don't have a 'break' on one side of the DMX.

 

Things to consider:

 

Is the RS485 line driver chip on the NJD faulty?

The symptoms sound an awful lot like the behaviour I once saw on a DMX line being driven 'single ended' with the - lead disconnected inside the lighting desk. Similar behaviour could be caused by a faulty rs485/DMX chip or by a broken/loose connection on one of the DMX pairs on the NJD. Or a defective cable (which I believe you've eliminated).

 

The easiest way to tell this might be to make one of the LED DMX tester terminators various people suggest on their websites, like this one from bigclive:

 

DMX Tester

 

If you're not getting DMX transmission on one of the two DMX balanced lines that LED won't light up.

 

What does the NJD actually output in the way of DMX when the input values are set to 0 for a DMX channel? Does it ouput a fixed size DMX frame or drop values beyond the 'end' of the currently active controller or note? You can't really tell without a DMX analyzer but this is beginning to sound like a candidate for one.

 

What addresses are all DMX devices in your light show on?

 

One possible thing to do is to set a DMX channel to 255 _beyond_ the address range for your 229s. This will _force_ the NJD to generate packets with 0s on all channels the 229s use but that aren't currently being set to any values by note-ons/control changes.

 

Are you using the NJD in midi note mode or control change mode?

 

I'm also wondering of you are having issues due to scaling from a 0-127 midi range to a 0-255 DMX range. I had a Lanta DMX washlight which set colour channels to audio controlled when red/green/blue brightness was set to 255. 0-254 was brighness. Resulted in the totally counterintuitive behaviour of the fixtures turning off when brightness was set to 100%.

 

Do the 229s have specific DMX values / value ranges the need to be set on certain channels to set diffeent modes?

 

 

 

Food for thought, thanks Dirkenstein :)

 

 

 

I have the NJD in control change mode... 0-127 addresses are 74, 81, 88. I did try moving the 81 up to 100 to seperate them at one point.

 

 

haven't got a tester but what I do have is NJD spectre which have an LED that changes from red to amber when DMX is good. I could put this after the 229s. Also I haven't tried the terminator in the spectres. Will add it to my todo list...

 

I like the sound of your Lanta washlight issue. The reason is that when I was using the simple DMX controller I noticed that when the strobe was on full, it wasn't anywhere near as fast as when I use it on the NJD. I would estimate that those sliders are outputing approx 100 on the midi range. When I was testing and trying to not use 0 on the RGBA channels, it worked lovely for a few control changes, but then went wrong -- this could have been due to sending 127 values...

 

 

 

I am covering all bases here --- I have returned the 3 units to CPC as I am at the end of my 28 days. When they refund the money, I will order another 3 and continue (they are worth the effort :))... if the original 3 were faulty, great. Otherwise I will try more tests and carry on the fight...

 

Watch this space :) Thanks for all your help so far guys (& gals if exists)