Re: Custom LED Circuit - Wire Gauge?
 

I question the 3.1 volt spec. Normally I see a drop of less than that, but these may have a built in resistor or current limit built in. Working with LEDs it is good practice to add the series resistor to prevent over current. Can you provide a link to where you bought them.

Re: Custom LED Circuit - Wire Gauge?
 

i tried to find the listing to no avail. lots of copycats though.

they were discrete led's. no built in resistor. they were rated for up to 3.6 v, but dissapate 3.1v.

the blues were a similar story.

i think they are bright LEDs. hence the high voltage drop.

Re: Custom LED Circuit - Wire Gauge?
 

OK,
You are buying "Super Bright" LEDs. They do have a higher voltage drop but the same rule applies for current limits. You do want to have a resistor in the string to keep the current at specified level, usually in the 20-40 ma range. If you are making each shape with parallel strings, then the wiring within each shape can be #22 with the resistors wired right at the string. If you wire each shape back to a Spike, make that wiring #18. Each shape is then the custom circuit and all rules are satisfied.
The only caveat I can supply at this point is the relative brightness will be a subjective thing. If the inspector, ref or FTA think they are too bright or if a team complains they are too bright, it will be up to you to correct the problem. That will mean, make them dimmer, or remove the breakers that feed them.

Re: Custom LED Circuit - Wire Gauge?
 

Do not try to avoid using a resistor. LEDs are not linear devices, meaning that when they get a little warm, their current consumption increases. More current leads to more heat leading to more current - this is known as thermal runaway. It ends in smoke.

Go ahead and use your four LEDs in series, but then add a small-value resistor - say, 20 Ohms or so (do the math) - in series. This will prevent thermal runaway.

The voltage drop across an LED is determined by its chemistry (and therefore the color it produces). Blue & White run at about 3.5 volts, red runs at about 1.8 volts. This is a physical quantity, not generally subject to change. Keep this in mind.

My advice is to use a milliammeter to measure the circuit current, and adjust your circuit to reach the necessary current, ignoring voltage.

Oh, and if you want, you can use ONE spike to power 2 colors. Hint: what's the "D" in LED stand for??

::safety::

Re: Custom LED Circuit - Wire Gauge?
 

i know that. I also don't like backfeeding LED's.

if the LEDs were not ordered yet, i will see about getting the kind with the resistors pre-installed (ebay has ones like that for 12v). if they were ordered, i will toss in a resistor.

I forgot about diodes not being linear... the flu must have rotted my brain.

Re: Custom LED Circuit - Wire Gauge?
 

This is a fine point I reckon (and a very understandable mis-interpretation) but my advice comes from a detailed reading of the NI 9403 spec. The DSC is NOT strictly spec-ed for 2mA per DIO output. The DIO channels on the DSC are directly connected to the 9403. The 9403 is spec-ed for 64mA total on 32 channels and TEST data shows 4.4V out when driving a 2mA load (on a single channel). It does NOT say that sinking more than 2mA will drive a single channel into over-current protection mode. And this type of spec is pretty typical. Page 8 of the NI 9403 Operating Instructions and Specifications shows an example driving an LED. So one can drive a few very high-efficiency LEDs (2-5mA) and be careful not to exceed the total 64mA. I have a few chip LEDs on a board laying around, tried this real quick and it works fine.

DIO is not the best way to go though. I would not recommend a team w/o an EE mentor use the DIO channels for LEDs. The solenoid outputs or relays or custom circuits are better choices. For example, it would be easy to drive scores of LEDS with some I2C high-current latches.

HTH