Intelligent Pixel strips and power injection question

[BeeJaay]

I am planning to move to the E681 this year with some power supplies. My question is if I have a long strip with 3 5m intelligent RGB strips daisy chained together, would I be able to plug power on both sides of the strip, or would I have to cut the strips in half, connecting the data and clock, but then adding power to the second leg of the strip?

I guess the simple question is, can I have multiple power supplies connected onto one continuous strip, or is that not feasible. (not much of an electrician)

Thanks!

[fasteddy]

If you download the running pixels with the ECG-P12R manual there are diagrams showing in detail how to inject power into pixels. the manual may be about the ECG-P12R but the power injection part is common for most pixel controllers.

But to answer your question directly. Yes you can use multiple power supplies as long as you isolate the power supplies from eachother and only connect the DC volts ground between power supplies. Power supplies should never have the + connected together unless they are designed for this (which most are not)

[neilric99]

It is not a good idea to have multiple SMPS connected to the same set of strings (or controllers), if one smps should fail then there could be an inbalance of power and that could cause damage to your strip or controllers. and as eddy said below

[timon]

If however your only running one PS you can power inject without isolating the +V and thus feed a string from both ends. The positive side effect is the entire strip or string will be more balanced.

[mschell]

Also, how often you need to inject power is dependent on what voltage strings you are using. 5v strips need it more often than 12v ones.

[cessnapilot]

If however your only running one PS you can power inject without isolating the +V and thus feed a string from both ends. The positive side effect is the entire strip or string will be more balanced.

I was testing this the other day because when I had a string of 50 pixels on full white they were pink at the end. When I only connected v+ to the other end it fixed it (no ground). Will this work long term or should I also connection the ground to the other side?

[timon]

I was testing this the other day because when I had a string of 50 pixels on full white they were pink at the end. When I only connected v+ to the other end it fixed it (no ground). Will this work long term or should I also connection the ground to the other side?

You actually only fixed it half way and at the same time possible caused your self a future problem. You still are loosing power because the ground was not returned as well. Raising the ground reference can cause signaling issues since the ground is the reference. BTW, if you had just put a ground at the other end instead of +V you still would have seen the string get better.

So always inject both ground and +V and if the +V is coming from the same supply you don't need to break the line. If from a different supply then the +V must be broken at the injection point. IMHO it's best to inject from the same supply.

[mschell]

Raising the ground reference can cause signaling issues since the ground is the reference. BTW, if you had just put a ground at the other end instead of +V you still would have seen the string get better.

I definitely agree with the first sentence, but disagree with the second.

Having a common reference for signal integrity is important, and can effect if the right data gets to a pixel way down the string(s). If the color change was due to signal integrity, a new ground connection might help. You definitely need a common ground throughout the same entire set of strips on one data input.

However, the color change in the pixels is most likely due to voltage drop of the supply voltage, due to the resistance of the wire/traces of the +V. Connecting ground to the far end of the strip would not have done anything to improve the voltage available to the LEDs near that end. Only connecting a +V to that far end would keep the supply voltage at a sufficient level to allow the LEDs to come to full brightness.

My longest 5V pixel/strip run is 32. Each new run of 32 has a new connection back to the controller.

On the other hand, my runs of 12V TM1809 strips are going to be 72 long, and I don't expect to inject power other than with the data at one end.

[cessnapilot]

Thanks.

I figured it was due to voltage drop but was supprised to see it be better as soon as I connected v+ and didn't change when also connecting ground.

[CaptKirk]

See without the +v at the other end, the electrons were getting too tired by the time they got all the way to the end of the string to be able to push through the LED. Injecting +V at the other end let FRESH electrons in and then slide home on the -V rail on a free ride.



Ok, I KNOW it is actually electrons being sourced from the -V rail because electrons are negative, but common current convention says the current flows from the positve to the negative so do not nit pick my little story..