Running xLights Sequence on Raspberry Pi (And maybe an ESP device?)

From memory the "80% rule" came about mainly due to people using "no name" power supplies. The output rating on them was always a bit suspect so better to run them at less than full rated power. It has no bearing on the efficiency of the power supply which is about conversion losses; more power (Watts) being drawn from the mains input than is delivered to the low voltage output.

Proper power supplies will also have a "derating table" which shows how the maximum available output power reduces with varying conditions. The main one being increased temperature. If you're in a cool climate there should be no reason you can't run a good branded power supply at it's full rated power. High temperature operation caused by high ambient temperature and / or poor air flow may reduce the maximum safe power output to as low as 50% of the rated maximum.
 
The 0.6watt per light figure used in the calcs has probably factored in enough overhead there anyway.
 
sorry yes what @David_AVD said, way more info and accuracy TY - my efficiency comment was related to efficiency rating which do vary and is usually 80% under certain loads but that is on the input conversion to output, a proper PSU will give you the full rating of course - afterall this isnt #HDDsizegate haha. 🤣
 
someoneAUS said:
The 0.6watt per light figure used in the calcs has probably factored in enough overhead there anyway.
Click to expand...
That 0.6W number is only correct for 12V WS2811 regulated pixels where they can get the full 48mA that a WS2811 chip provides at 12V (0.048 * 12 = 0.576W)

12V Resistor pixels generally have resistors that are too high, and as such limit the current to about half that. So yes, if you use that number, you have basically doubled your requirement over what you really need.
At 5V the wattage is different again (0.048 * 5 = 0.24W); about 2 & 1/2 times what you need.
12V GS8208 pixels are 0.015mA * 12 = 0.18W each, so the number above is over triple what you need.

This next bit will be contentious I'm sure, but I find that it is much easier to calculate using current, rather than power.
It's the limiting factor on a number of things, and you don't have to worry about the supply voltage in the calcs.

All pixel driver data sheets show the amount of current they supply for each LED (being constant current drivers).
* A WS2811 will supply 16mA per colour, so up to 48mA for all three.
- When used in say 60/20 strip with 3x LEDs in series per chip, the current doesnt change across the 3 LEDs, it's just not wasting power in a resistor/regulator
* A GS8208 will supply up to 15mA, but due to the series arrangement, it's irrelevant on the colour. i.e. Full red, full white and full yellow all use the same current, despite it being 1, 3 and 2 colours on at once.

On the power supply front, you need to look at the data sheet to find out the max power / current anyway.
eg. a LRS350 power supply puts out 29A at 12V (for 350W) or 60A at 5V for 300W max (despite being called a 350W PSU)
Add up the current draws of your pixels, presto; you can match it to the PSU.

It also means you can match the current draw of a prop to the fuse rating, to the required cable, etc. Voltage does not come into play for those.
(Except when doing V-Drop calculations and you want to see what's remaining, the dropped voltage is the same)
 
I completely agree that current draw is really what you need to look at. Measured current draw is is the ultimate way to go.

Only yesterday I was pondering about adding the actual current draw at 100% white on the voltage tags I already put on each prop male tail.

I'll be making up a simple test box with voltage / current meter that can be put inline with the prop under test. I think the F-Test may also show current?
 
Derating is a safety and longevity guideline. If a wire is rated for 10A then you should design for 8A. If a PSU is 100W then you should design to use only 80W. It is just a "rule of thumb" and many people push systems harder than 80%. I tend to want to protect my investment so I stick to the design guideline. :)

Martin
 
You guys "design" your shows? I just start hooking things up. Funny cause I'm a EE and the last thing I wanna do is pull out a calculator. If a fuse blows then I lower the brightness...lol. I remember in the early days I was running about 80% brightness cause I couldn't notice much of a difference between 80 and 100 so it seemed like wasted power. Everyone was ranting about designing for 100% brightness and I decided to calculate my power draw for one of my boxes. Based on number of pixels even at 80% it came to 41 amps. I was like hmm I've only got a 29amp power supply how does that work. I wasn't curious enough to take measurements but I figure something had to give so the power supply voltage probably drops when the load it too high and keeps the current under the max. I just lowered everything to 60%. Couple years later I was building a new housing for the controller and noticed the 30amp green connector was a little melted...lol. I ordered a couple new ones and soldered them on...good as new.
 
You make my case for me. You did not take care and you have melted connectors. That could just as well have been a melted house. I have had well designed props that have crisped wires due to unplanned intrusions (darn squirrels) that caused shorts and overloaded feed wires.

IMO The safer the design, the lower the chance that something really bad will go wrong. Something will still go wrong, but that is just Murphy having fun with us.
 
Our yards here are covered with rocks, the house is covered in stucco, and the roof is covered in spanish tiles. I don't worry at all about the controllers that are sitting out in the rocks....lol.
 
I get nervous about anything inside the house. One year I ran a string of pixels around the Christmas tree and then everyone started having these pixels issues and I quit doing it.
 
David_AVD said:
From memory the "80% rule" came about mainly due to people using "no name" power supplies.
Click to expand...
I got an Alitov 12v 30A supply, the company seems to have a presence, but of course that does not make it good. Is this a company I can keep using, or should I stay away in the future. I heard Meanwell is a good power supply provider, is that correct?

David_AVD said:
Proper power supplies will also have a "derating table" which shows how the maximum available output power reduces with varying conditions. The main one being increased temperature.
Click to expand...
I'll mostly be running them in the winter with temps from -20 to 50 (deg F). I may also keep some up year-round, but I'll only be using them at night. I've attached a whether chart of averages if that is helpful at all.
1705179644047.png

Is there a specific low temperature that I should never run the lights if it goes below that?

--

Also, someone said before (I think it was Martin) that you cannot get more then 4A though pixel wire. But right now for my 500 pixel test, I only have one power supply around the 250 pixel mark, and no injection runs. Since that would only carry 4A, why would all the lights light up if my calculations say that the power supply can only do 600 lights?

If 500 pixels can be run of 4A, why can't I run the rest like that? IT was also said that PI runs are only needed every 200 pixels (PI - 200 pixels - PI - 200 pixels - PI), so if 4A is the max in the wire, wouldn't 400 pixels only take a max of 12A? Or am I just taxing the system and an explosion is bound to happen?

Sorry, but I thought I understood it until I had that thought 😂

Thanks for your help!
 
Power supply brands other than Mean Well may be ok as long as they have data sheets with the relevant information available. I can speak for cold weather (we don't get it where I live), so will leave it to others to comment on that.
 
GR9 said:
Is there a specific low temperature that I should never run the lights if it goes below that?
Click to expand...
Not sure about the lows about that but it was -12 when I woke up this morning and the lights were running for the morning show! This is my third year and I never worried about how cold it was before I ran them. There are always temperature range ratings on most components which I think are more heat impacting than cold. My yard controller is literally buried under snow as we speak and it is running fine.
 
You must log in or register to reply here.
Back
Top