E682 and RGB issues

[ed26105]

I have a E682 board I am testing with 12VDC 2811 pixel nodes and pixel strips from Ray.  The pixel nodes work just fine under test.  However, when I enter 1, 2 and 3 in the test mode, some strips act as BGR and some act like BRG.  The universes overlap correctly.  I can't figure it out.  I must be missing something simple.

[wmilkie]

Just a shot in the dark, because I just did this myself;

when you power up the board, and look at the configuration page at 192.168.1.206, did you select  the correct outputs for the lights?

ie; at the bottom are the 4 boxes/menu choices for different light string types, are they all set to 2811?

Edited August 18, 2015 by wmilkie

[ed26105]

Yes, all correctly set to 2811.  First 4 are 70 lights each string, the rest are 50.  Each string is not necessarily 50 lights long, some are 23, some are 17.  But that shouldn't matter, should it?

[k6ccc]

The color order for strips varies from batch to batch.  You have no control of it at all.  It sounds like you have two different batches of pixels.  About the only thing you can do is to group the strips that are the same together in different clusters.  Then set the clusters for the different color order.

 

Are the strips different types?  Generally if you order a quantity of the same strips from Ray, they will all be the same, but a different type of pixels may be different.

[ed26105]

Yes, all are from the same order.  Nothing that can't be overcome with channel set up.

So what you are saying is "RGB" strip only means it has one LED of each color.  The order in which it is accessed via the software may vary per strip.

Also found this out with some LED modules - three in each rectangle.

Bottom line apparently is just deal with it.  And of course that is why we test.  Just running test pattern isn't enough.

I appreciate the insight. 

[k6ccc]

Let me give a little more detail on this answer.  As far as the LOR software is concerned, it will always be RGB.  You do NOT need to change anything in your LOR software to compensate for the color order in the strips.  The data in the ethernet data stream will always be RGB.  Where it may be different is between the E1.31 controller (your E682 in your case) and the RGB strips themselves.  When the strips are manufactured, the order that the data is sent to them is set in the manufacturing process.  There are six possible color orders, and your E682 allows you to set the color order for for each cluster to match the color order for the strips that you have.  In some cases, you will be told what the color order is from the source of the strips, and in some cases you will have to experiment to find it.  If you have to experiment, you use test patterns 1, 2, and 3 which should bring up Red, Green, and Blue respectively, and see what you get.  Then change the settings until you get the desired colors.  For example, if you have the setting RGB, and the colors actually come up GRB when you do test patterns 1, 2, & 3, then change the setting in the E682 to GRB.  If that does not work right (or you don't what to try to figure it out), just try all the options until you get one that works.

 

Based on what you have said, I am guessing that you have several types of lights.  Even though they were bought in the same order, because they are different types of lights is the reason that they are different color order.  As a general rule of thumb, if you for example buy 16 strings of RGB lights for a pixel tree, they will most likely be the same color order.  But a different type of lights may be different.  For example the pixels on my pixel tree are in one color order, but the strips that are used in the pixel star on top of the tree are a different color order.  Since the E682 sets color order in clusters of four outputs, I have the 12 strings for the tree in three of the clusters, and the four strips for the star in the last cluster.

[dougd]

Yeah, the RGB order can turn into a real pain in the butt. My pixel nodes so far have all been RGB, 3 different vendors. My 12 volts strips have come to me in 2 different RGB orders. My 5 volt strips have a different RGB order and my modules even a different order. I think I have recieved pixels in 5 of the possible 6 combinations.

[ed26105]

Although the E682 allows you to set the color order, this must be done in groups of four.  So I have to be careful with my assignments on the E682 board.  It is relatively easy if one has all the same type of lights on one board. But the way I am applying lights to the porch area of my house - spindles, rails and posts - I have to be careful to keep all the same color order lights in a group of 4.  And first I have to confirm the color order for each group of lights.  Keeping track of channel universe assignments is tedious with non-identical props.

Not being an electrical or chip wizard, I wonder what drives the color order and why it can't be consistent.  Apparently LOR can get the same color order on everything they sell.