Understanding 2811 wiring

[Kaden]

Hi All,


I just want to validate my understanding of how to wire 2811 pixel strings (12v 3x5050s per IC version)


I am thinking I will need to power sources (for convenience).


Assumption 1: All power supplies, controllers and pixel strinps have their gnd/- connected.
Assumption 2: The controller can use the same power supply as one of the strings.
Assumption 3: The controller's + output does not need to be connected to the strings (only gnd and data)
Assumption 4: Only the data wire needs to be connected from one end of the string to the start of the next
Assumption 5: The + and gnd wires can be connected anywhere on a string as long as each pixel is attached to it at some point.


I have included my awesome paint artwork to try and convey my understanding of what it will look like.

Notes: I understand the current requirements and voltage drop issues, I understand distance limitation between pixels, I understand there are dmx/e131 controllers - i have shown dmx for simplicity.


Am I correct, close or just plain wrong?

 

[AussiePhil]

In theory, that looks fine.

The unknowns
1. Should Pos and Gnd also go from the end of the strip to the start of the next strip ?

2. Should only Ground also be extended from End to Start?

I have wired my pixel Icicles as per 1. but i see some weird things at times but only at certain data rates indicating some weird cross talk issue, but the prototypes still have lots of excess cable that is looped around leading to increased crosstalk.

I'd built it and test if i was you, it should work fine

Phil

 

[fasteddy]

Your assumptions appear to be correct and your diagram appears to be correct, but there are many ways to do this

Refer to the ACL running pixels with the ECG-P12R manual on page 11 and 12 for diagrams on power injection

 

[Kaden]

Refer to the ACL running pixels with the ECG-P12R manual on page 11 and 12 for diagrams on power injection

That's where I got the information from I try to read all the ACL docs I can get my hand on.


My questions come from the 3 wire setup of 2811 (instead of the 4 wire 2801 style you have in the manual).

 

[fasteddy]

Refer to the ACL running pixels with the ECG-P12R manual on page 11 and 12 for diagrams on power injection

That's where I got the information from I try to read all the ACL docs I can get my hand on.


My questions come from the 3 wire setup of 2811 (instead of the 4 wire 2801 style you have in the manual).

The differences between a 3 wire and a 4 wire setup make no difference at all, the power requirements are exactly the same. So all the principles of connecting up a 4 wire are the same for a 3 wire, the only difference is that you dont have the clock wire in a 3 wire system.

 

[Kaden]

The differences between a 3 wire and a 4 wire setup make no difference at all

That is exactly what I had hoped I guess I am at the point where I just need to test to see if the theory is reality.

 

[fasteddy]

The differences between a 3 wire and a 4 wire setup make no difference at all

That is exactly what I had hoped I guess I am at the point where I just need to test to see if the theory is reality.

I have noticed a bit of flicker with the 2811 when using the 1804 code with the ECG-P12R so i would expect that we will have to wait for the 2811 code to be released by Ed to run it perfectly, but the 1804 code is good enough to test your lights.

 

[Kaden]

I have noticed a bit of flicker with the 2811 when using the 1804 code with the ECG-P12R so i would expect that we will have to wait for the 2811 code to be released by Ed to run it perfectly, but the 1804 code is good enough to test your lights.

I ordered one of these LT-DMX-1809(WS2811) to use for testing until the code and ECG-D2 ECG-P2 is released, as I prefer e131.

 

[Kaden]

Is there a reason to limit the number of pixels on a string or controller?

What I mean is the ECG-P2 has two outputs, is there an advantage at the pixel/IC level to split a single full universe of pixels in to two half full universes?


Same question about the controllers, do they run better if they aren't trying to output their max number of pixels?

 

[AussiePhil]

Basically


No, No and No


However some controllers DO limit the number of pixels on any single output, Stellascapes certainly do by design and i believe the E68x series does as well



Cheers
Phil

 

[j1sys]

i have to disagree a little: yes, yes, and yes ... but with the following qualifiers

it is all about time.

even at 750kbps it takes a finite amount of time to send the raw data to a given number of pixels. so the longer the string the longer the time. we are talking in the 5-10ms time period. so if you REALLY want rapid updates and sub 25ms lighting changes (strobe effects) then the shorter the better.

with that said, we are going to support multiple universe per string capability. for many applications 25ms to 50ms updates with a longer string will be more cost/wiring effective. but longer strings also introduce more risk that a bad pixel will take out more of your show. so it is always up to the display designer.

so the real answers are: YMMV, YMMV, YMMV remember my 'signature' block: There are no rules, and those are the rules.

-Ed

 

[Kaden]

So will having an artificial limit of 1 universe (170 pixels) per string/output limit provide me with good performance for everything except a strobe effect (which dmx isn't capable of anyway)?