Because of a space requirements (aesthetics) I would like to place the capacitor on the end of every 1m ledstrip (connect trough 1 to 2 meters of 1.5mm2 cable).
As far as I could think of there will be no electrical reason it shouldn’t work, as the LEDs on the ledstrip are connected in parallel (power lines) they could draw power from anywhere (just like with power injection)…
Is this correct? Or did I miss something?
Thank you!
You likely do not need the capacitor(s) at all if you have good wiring and power supply. I would add a level shifter though.
Would be even better if I didn’t need them aesthetically… I’ll test them without before I glue everything together 
I do have a level shifter in the [logic] part of the drawing, didn’t want to draw the whole diagram…
Thank you!
Not sure why say “You likely do not need the capacitor” because in some cases you do else I can also say you do not need a shifter because i never used one . In reality you should follow the KB best practice
If you have good power wiring and a capable power supply, there should be no need for the cap’s as the PSU and wiring are up to the job for intended load. Adding the caps is like a duct tape fix to a larger problem. I would MUCH rather a person skip the caps than skip the level shifter. The OP is also using 1M LED strips and if they are pulling that much juice to cause an issue or brown out then their whole setup needs an overhaul.
100% agree with this statement. A capacitor is there to counteract cable inductance and (maybe) help with EMI but you would need pretty bad wiring or very long power cables that it actually turns a ‘non working’ setup into a ‘working’ one.
So thank you guys!
I made 2 shorter prototypes
In the end the capacitor wasn’t the problem. The cable strain relief from the solder joints I made took me hours to fuguire out as the presumably good solder joints keep on pulling the pads. The problem was when rolling the cable just a little the length of the 3 cables changed inside the led profile… Its a 1.5mm2 cable that is so rigid it pulled the pads…
That cost so much effort but also space on the led profile that I also just placed a capacitor on the end of the led strip eventually. Not because testing showed I needed it, because it didn’t look so bad, and it could only prevent yet unknown problems.
Anyway I have a working prototype powered by a power bank that I keep moving around to check if the cable stays put, so far so good till now…
In a week (if the prototype is still working) I start doing the other 12 I meter ones, those will be powered trough the fusebox and a 40a PSU…
One more potential problem I discovered is the SSR-DD (solid state relay) creates a voltage drop of 0.7v… When on full power (I will allow 4amp or 5aamp per led strip I think) the end of the strip it has a 1.4v drop in total at 5amp (with the most longest cable I need) strip specs say 3.7v min operating voltage. Making the SSR-DD the limiting factor to 4amp, a mechanical relay would probably allow me 6 or maybe 7amps, but hey are hard to find in the 30 or 40 amps I need at 5v…
On a 1 meter stip the black isn’t that annoying after all (this is 0.5m)
The cable strain relief is just a small circle in all the separate conductors, and a tywrap at the very opening… Capacitor is just a little to big to fid inside the profile so I cut the defuser middle where it stuck out. And because I needed a ‘end of the light part’ otherwise it looked weird when the lights were on I just used thick heatshrink
And I made a very small breadboarded controller for it , The ESP is soldered on the back with just a few pins getting trough, in between are the resistors for the output…
I took this way too serious for a prototype… 
The PSU got the Capacitor inside that is needed to run the full Amps savly
if you pplan for more then 200LEDS per strip ypi need to incect after the 200leds
i prefer to do so after 100 LEDS WS2811
if you run ZIGZAG its best to incect on both sides only the Positiv
stripes are made the GND Flat out so they handle the Amps return quite good
only the forward power is to be concerned
looking at a WS2812 stripe both +5V and GND have about the same amount of copper on them if that its what you meant by ‘flat out GND’ then it’s not so.
I’m using SK6812 strips, I wanted the white leds in it…
What do you mean flat out GND? Both traces on my strips look the same width and thickness…
If I didn’t want the whites I would have gone for a 24v/12v strip to keep the amperage down a little…
The easy way to deal with the switching of high current DC power is to; Power the Esp with a small 5v supply -either through the vcc pin or from the usb port and then power the LEDs from your big supply. You can then use a much smaller relay to switch the AC side of the big power supply.
If my prototypes survive the next week of moving around I’ll consider that…
Any advice on a ~220v relay? Fan I maybe control the relay itself trough WiFi within WLED?
I thought maybe I’ll add another esp with WLED, add only the relay and the strips connected to other WLED ESPs on over DDP making that one the “master controller” and hide the other inside a VLAN or something so they don’t show up on everyone’s phone…
But I don’t know if the master controller will switch the relay as there are no actual strips connected…
This should work fine: https://www.amazon.com/gp/product/B093LC9DNJ connect d1 on the relay shield to w/e gpio you choose on the controller/in WLED, 5v to controllers 5v and gnd to controllers ground.
The master would still switch the relay. It does not care if there are LEDs connected or not and the LEDs gpio pin(s) have nothing to do with the relay’s gpio pin
Thanks for the info 
I’ll take a look!
