Ray Wu 400x WS2811 & T1000B - flickering beyond 100 pixels

aususer

New elf
Joined
Jun 24, 2010
Messages
5
Hey guys,


just received my Ray Wu 4x 100 pixel 12VWS2811 strings and the T1000B I ordered with it.
(only one faulty led on the end of a string!)


If I connect the T1000B to any of the 4 100pixel long strings (each with seperate 12V PSU) - each string works as expected.


If I connect the Data IN on PIXEL1 for all 4 strings together and feed the T1000B into it - they all work expected (ie. all four strings working in parallel).


However:
If I run the strings in series (ie. LED 100 DO -> LED 1 DI on second string etc...)
The second string onwards always "flashes" each light in a random fashion (note: they do show colours - just not coordinated). Just appears to almost be a signalling issue.


The plan is to make a mega tree - 2.5Meters high, with 25 pixels per string and 16 strings "long" (essentially just a big 25x16 matrix).


My understanding is the WS2811 protocol can easily handle this, and plenty of others appear to have done similar - and the T1000B is advertised to do 2048 Pixels... so I can't see a problem.
BTW: I have programmed the T1000B with 100, 200, 300 and 400 long strings with LEDEDIT2013.. just in case.. made no difference


Short of pulling the CRO out and having a look at the signal on Pixel 100 DO... do you guys have any ideas?


Is there a limitation of the T1000B to only 100 PIXELS? I've shot an email to Ray to ask - but I ask the collective in the meantime.


Thanks.
Mike
 

fasteddy

I have C.L.A.P
Global moderator
Joined
Apr 26, 2010
Messages
6,648
Location
Albion Park NSW
Im sure you have already taken this into consideration but how are you injecting your power to your strips as this can be caused by ground loops.
 

aususer

New elf
Joined
Jun 24, 2010
Messages
5
Yup - no ground loops. But good point - might be time to crack out the CRO and have a look on the 100'th DO and see what it looks like.
 

aususer

New elf
Joined
Jun 24, 2010
Messages
5
you might be onto something here... it might be that these 100pixel lengths are being affected by some grounding issues on the data line.. a quick "dodgy" both-end grounding test is looking promising..
(as lecturer said recently: if you hear hoofsteps - think horses before zebras)
thanks for putting me on the right track!
back with more detail as it comes to hand.. (and when I can find where I put those damn CRO probes!)
 

aususer

New elf
Joined
Jun 24, 2010
Messages
5
Your a gem!
Problem solved... looks like there was noise on the data line on 100DO.. grounded both ends and Bingo - problem fixed..
Thanks for your awesome idea.

400light mega.JPG
 

fasteddy

I have C.L.A.P
Global moderator
Joined
Apr 26, 2010
Messages
6,648
Location
Albion Park NSW
Glad i helped so easily. Ive had similar issues in the past so i knew thats one of the first places to look
 

Beefer

Apprentice elf
Joined
Feb 19, 2012
Messages
55
handy to know! Will keep in mind for my own 100+ runs of pixel strings...
 

reminder

New elf
Joined
Sep 3, 2013
Messages
1
Hello,
can you explain how you grounded both ends (data/clk)
thanks
reminder
 

fasteddy

I have C.L.A.P
Global moderator
Joined
Apr 26, 2010
Messages
6,648
Location
Albion Park NSW
The data and clock is not what is grounded.

what the user did was just connect up +v and ground to the end of the strip as well

so the begining of the strip will have +V, Ground, Data and Clock connected
the end of the strip will only have the +V and ground connected

Make sure you use the same power supply for this
 

Charl Marais

For my twins was the excuse I started with.
Joined
Jun 27, 2013
Messages
187
Location
Secunda
I had a similar issue but found that changing power injection did not help.

I could replicate the problem by playing with the intensity settings on xLights/Nutcracker and isolated the problem to one or two pixels which I swapped out for other units and this resolved the problem for me.

So in my case it seems that one or two individual pixels were corrupting the data flow at reduced intensities i.e. "dirty" data.

I will be trying the grounding effect as well. Thanks.
 

dannyp

Full time elf
Joined
May 12, 2010
Messages
262
Location
Sydney
Re: Ray Wu 400x WS2811 & T1000B - flickering beyond 100 pixels

ɟɐsʇǝppʎ said:
The data and clock is not what is grounded.

what the user did was just connect up +v and ground to the end of the strip as well

so the begining of the strip will have +V, Ground, Data and Clock connected
the end of the strip will only have the +V and ground connected

Make sure you use the same power supply for this

So just that I have this ABSOLUTELY correct, at the end you simply join the +v and ground wires together? I thought that this would blow a fuse.
 

fasteddy

I have C.L.A.P
Global moderator
Joined
Apr 26, 2010
Messages
6,648
Location
Albion Park NSW
Your assumptions would be correct if you just joined the +V and ground together in that fashion

You connect the ground wire to the ground terminal or cable and the +V to the +v terminal pr cable at the end of the strip

So what i meant by connecting is not connecting them together but instead connecting them to their appropriate terminals/wires at the end of the strip/string
 

dannyp

Full time elf
Joined
May 12, 2010
Messages
262
Location
Sydney
Re: Ray Wu 400x WS2811 & T1000B - flickering beyond 100 pixels

thanks. that makes sense.
 
Top