LED strip lighting

[Scottl]

I am planning to suspend LED strip lights over my layout and all the materials have now arrived from the slow boat from Asia.

I am using a 5M strip of double density LEDs (600), type 5050.   Based on the specs, the maximum current would scale to 144W @ 12V for the string, so I purchased a dedicated 120W 12V power supply.  This factors in approximately 20% de-rating to be conservative.

So, some tests.  No pictures because those LEDS are bright.  I can't believe that I thought 1 amp of current illuminated them!

The power supply is operating to spec, 12.2V.  I wired the string to the leads, taking care to match polarity.  They light up like no tomorrow, almost painful to look at directly.  However, my power meter indicates the supply is only drawing 35W from the main, so dismissing inefficiencies for the moment, this is only taking about 1/3 of my potential current and power. 

After thinking about it a bit, I considered that amperage flow to the strip is limited by conductor size- c. 20 ga leads to the strip and then the flexible PCB on the strip itself to convey current.

A second test:  wiring the two ends of the strip (as a loop) to the supply, careful (check twice) with polarity, and turn it on.  Radically brighter to the eye, and now consuming 86W at the main.  These things are bright- wow.  People who light them up with the 1A output from a power pack would not recognize them.

Am I correct in assuming that my conductors were limiting current?  If I give sufficient conductor to deliver current, that this system will approach the draw of the power supply (120W)?  I'm no stranger to electronics, but I'm not used to working with high power circuits.

My installation plan is to cut the strips into three (unequal) lengths, and then feed current to the middle of each strip.  I was going to use 16 ga feeders, just because that is what I have on hand, but I could change that if needed.

Any thoughts?  Advice?  Thanks in advance.

[C855B]

20AWG is rated for 10 amps, so it's not the wire per se, although you didn't say how much wire - you're getting into possible voltage drop territory there. I'd more suspect the PCB path contributing to voltage drop. But... truthfully... I'm mystified. There's something else going on in your initial test. If it was voltage drop due the PCB and you fed it from the end, you'd notice a reduction in brightness further away from the feedpoint.

At any rate, you're OK with your game plan. Since you're cutting the tape, #16 is more than adequate, center-feed will work great, and you have plenty of "oomph" with that power supply. You'll be fine.

[Scottl]

I'm glad I'm not the only one a bit mystified. I think you are right, it is the PCB board, since the 20 ga feeders are only a few inches long. 

On the open ended test (power fed on one end of the strip), could I measure voltage drop at the end of the strip?  The LED's must tap into a 12V buss long the length, so in theory I should have 12V at the end, or it gives me a measure of drop over the length?  They are so bright, even with only 35W that I can't really look at them and perceive a difference in illumination.

 

[Scottl]

I measured the open ended strip voltage at the end of 5m and it is 8.75V  I think we have the culprit.

Off to the hardware store for parts for the fixtures:  copper pipe, pvc pipe and caps...  Let there be light! (after I dig out of the snow... )

[peteski]

We discussed just this type of situation here recently in https://www.therailwire.net/forum/index.php?topic=35228.0 . Most useful (wire gauge) info is towards the end of the thread.