Wihit not exactly white on rectangles.

[MrChristmas2000]

I have been building some coro stars using Ray's rectangels. I have begun to notice that using the SSC in test mode that not all of the rectangles are displaying white the same. Some of them are displaying a slightly pink hue. The rectangles are at different points in the line as well.

Is this to be expected because white is a combination of all 3 leds? Are some of the modules just not responding correctly to the white mode.

I also noticed that the majority of the rectangles displayed 'cool' white (slightly tinted blue).

I will try to capture a video of this if possible.

Tom

[rrowan]

The blue led over powers the red and green. Try lowering the blue until you get a "white" that you like.

All RGB lights have this issue. In fact the RGB nodes don't match up with any color chart that I know of, hence my writing the color finder program.

Cheers

Rick R.

[RJ]

If you run too many the voltage will get lower on the end ones and cause this also so make sure it is not the farterest ones from power that are doing it. you can inject power on the other end to over come it if you need that many of them. These modules use more power. And of you use some of the ones he added with additonal leds ect. they pull even more so as the current draw increases on the item you have to lower the number of them you use with out adding power.

RJ

[MrChristmas2000]

The test strings are no longer than 30 modules.

The position in the string doesn't seem to matter it happens up and down the line.

I am using an SSC with the test program at this point just for functionality testing. I would assume that the output from the PIC should be constant as far as RGB is concerned.

I can understand the adjustment for the imbalance but that would really drive one crazy during sequencing when you need to do white.

Here is a short vid of one star displaying this difference.
You are not allowed to view links. Register or Login

[tng5737]

How many strings are you using and how are they connected.  I just had two sets of rect. in series and a couple of rects on the second string exhibited the same pinkish color as you describe.  they were in the middle of the string and not next to each other.  As soon as I put the second strin on its own SSC they both look alright.   
When I built my flake I used 45 rectangles and originally had the same issue I solved it by tying power leads of each leg together at the center (and did the same for the gnds, obviously)  That solved the pink tint issue.  Here is a link to the thread:
You are not allowed to view links. Register or Login

[MrChristmas2000]

They are wired like this.

SSC --- 3.5ft wire ---- M1 - M2 - M3 -------- 7ft wire ---- M1 - M2 - M3 --------  7ft wire ---- M1 - M2 - M3 -------- 7ft wire ---- M1 - M2 - M3 -------- up to a maximum of 10 sets of 3 Modules on each star.

I am still building stars. I will have 30 stars when finished and plan to run 10 with each SSC. The theoritical max I believe is 80 of the Rectangle modules off 1 SSC so that should fit within the specs easily.

While I was writing this I had a thought, I do have a TH2010 tester so I went and hooked it up to one of the strings. The same result with the differences in the white, but in addition to that I noticed a difference in blends such as yellow as well. I had 2 different hues of yellow which followed the same modules as the white differences. The differences in Yellow was very noticible as well.

I wonder if there is differences in the controller chips batches? Agh.... this is going to be a pain I can see already.

[tng5737]

In my experience 7ft between modules is going to be problematic.  I suspect you will need some null nodes in between to regen the signal.
Did you change the 330 ohm resistor to 150 ohms in the SSC?   The TH2010 has a stronger driving signal than the original SSC.  RJ changed the to 150  ohm to povide greater drive.

[MrChristmas2000]

I changed the resistor. That did increase the drive capability.

It didn't matter as far as which device was driving the modules. The results were the same.

After I thought about it this effect was happening when I initially tested the modules when they were just inches apart.

Mabey RJ could come up with something like the lighting curves on the Lynx Express units. Mabey the PIC could be told to adjust the color balance for each individual module or group of modules. Just a thought.

I will be doing some more testing over the next few days.

[RJ]

You need to feed power additionally in from the other end since you have 70 ft of wire. 7 Ft is pushing it so this also is hurting you.

RJ

[MrChristmas2000]

How do you parallel 2 power supplies? If I connect another 12 supply to the far end of the chain won't that be the same as pralleling supplies?

I'll check the voltage today to see what it is at the end of the chain.

[RJ]

I meant the same supply

Sent from my Charge by Tapatalk

RJ

[MrChristmas2000]

Did some further testing today.

I marked the stars that had the red tinge modules. I then moved them around.

For example I had one that was the second star in a string and moved it to a second string of stars in position 7. The star had 2 cool white modules and the one warm (red) one. I replace that star with one that was all cool white and continued to be all cool white in the same position of the first star. There was no change in the white tinge of the stars modules.

I further tested the same star using my TH2010 tester which I have a 12 amp power supply connected to and there was no change in the white color tint level.

Given the results I don't believe that the color balance differences had nothing to do with power. I believe that it's a manufacturing problem, perhaps the difference in the chips from one batch to another. Unless you pay attention to that level of color detail a manufacturing facitily probably wouldn't care.

I appreciate your clarification on what you meant about power injection.

[MrChristmas2000]

RJ,

I have done some further testing.

I did add power to both ends of the string of modules (30 modules). The whole string of lights got brighter.

It did not do anything for the color hue though. The cool white ones stayed cool white just brighter while the warm (red tinted) ones just got brighter with more red tint.

Now that I have all the stars built I will do more testing.

[dmoore]

Given the results I don't believe that the color balance differences had nothing to do with power. I believe that it's a manufacturing problem, perhaps the difference in the chips from one batch to another. Unless you pay attention to that level of color detail a manufacturing facitily probably wouldn't care.

Having purchased thousands of LED modules from a wide variety of sources I can say that quality varies quite a bit.  You can purchase a basic RGB module for .75 cents and the "same" one for $1.25. 

[MrChristmas2000]

While I managed to squeeze out enough of the Cool White retangles for this project it didn't leave very many spares. I now have a whole pot load of the red tinted white ones.

I'm really not sure how you could get Ray to only send you the Cool white ones since with his tester you have to look really close and there has to be a cool white one along side of the others to initialy notice the difference. It is very noticible when you have 3 or more in a group.

The power injection on both ends does help with the intensity though. I measured about a 2 volt drop over 30 rectangles.