Wiring Trick for Numerous LED Strips

DIYLED

Active Member
Levers make things easy. Without these, it would have been a real pain in the ass to do this a bunch of times:
LED_wiring_PCB.JPG


It's kind of hard to see in this 2nd photo, but the big white wire is connected to a big red wire in the upper left:
LED_wiring_PCB_PSU.JPG


The load is spread out on 6 fuses for this single power supply that has 2 pairs of 24VDC output wires. The small black wires are actually the positives coming from the LED strips. The little white wire powers the DMX decoder board, and the big black wires (negatives) and the big colored wires (PWM'd negatives) also connect to the DMX decoder.
 
Nice neat work on the PSU. Not portable I suspect for install, and with all home runs. Fuses nice, except during a show - loading over sized instead normally from the auto-reset breaker PSU the normal way to do it. Circuit board very nice and clean way to do this, low voltage amperage must be quite some calculations in figuring out it's circuit board construction. Love lever locks, mostly the individual ones when remote or inside a PSU. Take stranded or solid wire well in varying gauges. Only one mounting bolt, not worried about the spin factor? Well done and nice organized. Not clear what is doing what for follow up tech people to service in understanding. More labeling and wiring diagram perhaps if spending the money on such a circuit board. IKA... better idiot proofing. But looks nice.
 
Thanks. It is actually portable. On site we only have to connect Romex and network cables (for art, not music shows, otherwise I would use XLR's).

There's actually 5 machine screws (mini bolts). The next time I might move them farther away from the terminal blocks so they aren't in the way of the wires.

There's nothing that could catch fire in this setup, so the fuses basically protect the MOSFET's on the decoder and prevent the embarrassment of smoking wires. This PSU does not have short circuit protection, it is current limited to 25A max which could potentially cause issues without the fuses. Yes, amp calcs are very important even at 24 Volts, and there's a basic formula for circuit boards based on trace width, copper "weight", and allowable temp increase. On a 5-Volt system with addressable LED's the amps can easily be higher than people might expect.

I suppose many years later this would be a bit unconventional to service. Hopefully they know Kirchoff's current law and hopefully I make a diagram...
 
Great workmanship again, but the labeling.... what is Kirchoff's current law I don't know or know that I use in not knowing I use it. That related to how you calculated the circuit board to carry the current which is a black hole in knowledge to me in specifying? Something I don't get in seeing the gauge of wire feeding the device, how it's circuit board distributes the load - how you calculated that circuit board in transferring the load science I don't yet know.

Is this in a set piece where your panel travels in not having to re-connect the wires each time other than for data? Noting or you would use XLR otherwise. Other than in a set piece, having to connect 60x wires +/- a few each install is not fun and would be problematic in having to re-strip the wire often. Would take time.

I don't note a decoder so it's dimmer controlled for RGB somehow or just white? Not seeing control of channels or RGB. Again, not attempting to give you a hard time, more looking to understanding it in thanks for what you are showing. Looking to learn more.

I count 20x home runs for the blue line. But theoreticlly fused to 7.5A for three circuits. A simple 600w24VDC transformer should power up what you are breakering for if my math is correct. Most of those transformers are self breakerd' I would think. This in auto- turning back on if overload as opposed to having to replace fuses.

Good workmenship and clean wiring, but don't understand what you are doing. So yes better labeling as per a first step.
 
Here's a really good web site for learning electrical theory in a practical way:
https://www.electronics-tutorials.ws/
https://www.electronics-tutorials.ws/dccircuits/kirchhoffs-current-law.html
The terminal blocks for red, green, and blue wires are just a wide double trace connecting all wires of each color together. So each big green wire connects to all the little green wires, same for red and blue groups. The fuses connect the big red or white positive wire to the little black positive wires.

I basically wired it similarly to this but added a fuse for every 8 to 10 RGB LED strips (not full reels but cut to length):
https://shop.dmx4all.de/images/product_images/original_images/dmx_leddimmer_X9HR_d_581.jpg
Unfortunately I'm not able to take more photographs because everything is assembled and bubble-wrapped right now (pink anti-static bubble wrap) and ready for transport.

Every power supply is connected to 2 of my custom PCB's and 1 DMX decoder.

Online calculator for circuit board trace width requirement:
https://www.4pcb.com/trace-width-calculator.html
There's some oddness in the shape of traces near the fuses. That's because the KiCAD software is not normally used for what I was doing, so I had to get creative.

Here's the datasheet for the power supply:
http://www.meanwellusa.com/webapp/product/search.aspx?prod=HLG-600H
In the 2nd diagram on the 3rd page, it shows how the constant current function works. It looks like the fault shutoff condition is when the voltage drops below 50% of nominal voltage (dashed line), so it looks like it would shut off with a complete short circuit. A partial short circuit might be different, but maybe thermal expansion would either close or open the partial short. Curious what somebody else has to say about that. I trust fuses.
 
Last edited:
Actually I realized that tutorial site might be missing some subtleties of circuit design. For example, adding capacitors to high-gain opamp stages ("pre amps") can prevent amplification of very slight DC offsets. But it is a very good web site.
 

Users who are viewing this thread

Back