/diy/ - Do It Yourself

[Anonymous No.1343596]

Hi /diy/
I am wanting to build a 200 LED series-parallel light array and I am looking for advice on the resistors needed for it. The array will have two different LEDs with different forward currents and voltages. Using this calculator http://led.linear1.org/led.wiz I figured I need 180 ohm 1W resistors for the first set of LEDs and 68 ohm 1/4W resistors for the second set.

Is it ok to mix resistors like this? I originally was going to use one rated resistor for both LEDs on each string, but then looked at the forward current to forward voltage graphs and thought they might not light up if I don't fall within the graphs curve for each LED.

The LEDs are 630nm Red with 2.2 Vf and 30 If and 850nm IR with 1.5 Vf and 50 If.

Here are the data sheets for both.

https://www.mouser.com/ds/2/90/C503B%20RAS%20RAN%20AAS%20AAN%20RBS%20RBN%20ABS%20ABN%20RCS%20RCN%20ACS%20-269115.pdf

https://www.mouser.com/ds/2/445/15400585A3590-1113580.pdf

[Anonymous No.1343605]

>>1343596
Forgot to mention the that supply voltage will be 24 volts and each string will be 10 LEDs long. The entire array will be 10x20.

[Anonymous No.1343613]

Yes, it's fine. You're going about building the infrared array incorrectly if you need 1 watt dropping resistors for a 50 mA current. A buck converter set for constant current would be a much better bet

[Anonymous No.1343616]

>>1343613
Premade buck converter set to 15 volts with 1 ohm resistors for the 5 infrared strings (in parallel), and just run the 5 red strings off of the 24 volt with the recommended 68 ohm resistors (in parallel).

[Anonymous No.1343630]

>>1343613
>>1343616
Ok thanks. I am trying to keep it relatively simple and compact and I don't know much about buck converters so it will just complicate things for me.

Would it have been alright if I went with my oringal idea and used just the 1 rated resistor of 1/2w and 68 ohms for both the LEDs? Each string would get the same resistor. Both the LEDs fall within the same tolerances, but the curves on the graphs are quiet different.

[Anonymous No.1344107]

>>1343596
>Is it ok to mix resistors like this?
each string is electrically separate. this would be fine, anon
>>1343630
>the curves on the graphs are quiet different.
then you'll probably get different relative output from the LEDs. this may not be what you want
consider rearranging the IR array by adding a couple of LEDs to the IR string eg. 12x8, 14x7, 15x6, 15x7, to bring the total Vf of each string closer to the supply voltage
also, if you're concerned about price breaks, note you can connect two of the 1/2W 68ohm in series to get a 1W 136ohm resistor or in parallel to get a 1W 34ohm resistor

[Anonymous No.1344279]

>>1344107
>then you'll probably get different relative output from the LEDs
Oh ok. That's what I thought originally, but it isn't desired. I want to be able to run all the LEDs at their maximum intensity with respect to their limitations.

I guess I should have worded the question better. What I meant to ask was if I were to use the 68 ohm 1/2W resistor for both the LEDs at 2.1V per LED, will the IR ones not light up because the voltage falls outside of the graphs curve. 1.6V being around the max those particular LEDs can run at. Unless I am misinterpreting the graph.

Also, what's the problem with using a 1W resistor? I did a calculation for what the minimum would be and it's 0.45W for the IR string. I figured that the calculator choose the 1W resistors because it wont overheat.

>if you're concerned about price breaks
Not really, the resistors are a few cents each and it will only end up costing around 5 dollars for 40 or 50 of them. The IR LEDs are where the major expense is.

Thanks for the help.

[Anonymous No.1344372]

>>1344279
>not light up
correct. it won't light up... ever again, because you will have burned it out
according to the radiant intensity vs. forward current graph, you will get maximum intensity out of the array by running at 50mA (but note the derating curve right next door if you're going to operate this device in a warm environment or an enclosed fixture). you can only do that at 24V on a 10-LED string by placing a 180 ohm resistor, and the power dissipation of the resistor is determined by all that
>what's wrong with 1W
just a lot of power converted into heat rather than light (see the derating curve)
this is why I suggest rearranging the electrical configuration of the array. instead of converting all that power to undesired heat, you can add LEDs that will instead turn it into the desired product. a 15x7 array, for example, will have a 22.5V total Vf, requiring the dropper resistor to only dissipate 50mA * 1.5V = 150mW, which a 1/4W resistor will handle just fine. thus, a minimum 75 ohm 1/4W dropper resistor on each string
NB, all of this assumes that your 24V is well-regulated. if it's more than a few percent high you might damage your LEDs. if it's more than a few percent low, they might not light up, if input power is poorly regulated you might want to take the loss in the dropper resistors and drop a few units from the strings. again, mind the temperature and the derating curve while you do this

[Anonymous No.1344500]

>>1343605
Stack more in series man. Instead of dropping 9V across the resistor for the IRs you should put 14 or 15 in series and just drop the remaining 3-1.5V, much more efficient.

If you want to you can search ebay for parallel LED driver circuits, but 20 parallel is fairly large. Another option is just having 22 TO-220 transistors and a couple resistors set up as current mirrors, which would be more stable.

[Anonymous No.1344704]

>>1344500
note for the noob OP: in the diagram, R = (Vin - 0.7) / I, they don't make that relationship explicit

[Anonymous No.1344968]

>>1344372
>(but note the derating curve right next door if you're going to operate this device in a warm environment or an enclosed fixture)
>just a lot of power converted into heat

I plan to eventually put it in a box or something similar with a fan that blows over the entire circuit board and out a vent at the other end. I am not to concerned with heat, although it did cross my mind that it will have to be managed somehow. But the array will only be on for about 2 minutes max each time and then turned off for a bit over the course of 30 minutes to 1 hour. Which reminds me I should consider a push button switch.

>a 15x7 array, for example, will have a 22.5V total Vf, requiring the dropper resistor to only dissipate 50mA * 1.5V = 150mW, which a 1/4W resistor will handle just fine. thus, a minimum 75 ohm 1/4W dropper resistor on each string

The problem with this is that I want to have the red LEDs on a string and the IR LEDs on a string below the red ones and then alternate that pattern. Since both of them run at different voltages and amps I can't configure the array in a way that would be symmetrical. One string will have to be shorter or longer, or I'd have to sacrifice intensity on a few of the LEDs if I were to mix the RED and the IR in one string. Or get a buck converter or the transistors like others suggested, but that's too complicated for me. Maybe if it works for what the intended purpose is I'll build a more efficient version of it.

>all of this assumes that your 24V is well-regulated. if it's more than a few percent high you might damage your LEDs.
I planned on getting this power supply https://ca.mouser.com/ProductDetail/MEAN-WELL/RS-25-24?qs=sGAEpiMZZMsPs3th5F8koAOKs67AukcKKUWuofiQenY%3d or http://www.meanwell.com/webapp/product/search.aspx?prod=APC-25 I like the first one because you can adjust the voltage and it's not hardwired so I can potentially use it for something else in the future.

At any rate, thanks for the help, it was invaluable.

[Anonymous No.1345079]

>>1344968
>The problem with this
I literally do not see the problem with that. the business end of the LEDs can be arranged symmetrically as you see fit. if you go with 10 and 15 lengths, the 15s can be neatly interrupted in the middle on the non-resistor end. you can jam the other 5 in the ends of one string or w/e. just throwing that out there
>power supply
that's pretty well regulated, resistors ought to be fine then
anyway good luck with your project. whatcha cookin?

[Anonymous No.1345500]

>>1345079
>the 15s can be neatly interrupted in the middle on the non-resistor end. you can jam the other 5 in the ends of one string
I haven't had time to think about this, but with my limited understanding I am not sure how to go about doing this without a lot of complicated wire arrangement and soldering. Forgive me for I am a brainlet.

>whatcha cookin?
I am trying to make a Red Light Therapy array and test it on myself. There are a lot of studies out there on their effectiveness, I was curious about it and it seems fairly simple to make a small array. These things are ridiculously over priced on commercial markets and you don't know what type of LEDs they are using and for the price they have very few on them.

When I get the chance to order the parts and make it I'll post a thread about it if you're interested.

[Anonymous No.1345555]

>>1343596
What the fuck, Anon? How the hell am I supposed to get Tunak Tunak out of my head now?

[Anonymous No.1345572]

>>1345500
>Red Light Therapy
L-looks fun!

[Anonymous No.1345594]

>>1345555
One is never supposed to get Tunak Tunak out of their heads!
>>1345572
Be weary of these tanning salon red light therapy booths as it's not true RLT. They still use fluorescent which has too large of a spectral bandwidth and emit a range UV (which can be beneficial for killing bacteria in the 400-415nm range). With lasers and LEDs you are able to get a more precise nanometer wave length which is supposedly required for the reported benefits this type of therapy offers.

[Anonymous No.1345645]

>>1345555
have you tried red light therapy

>>1345500
oh, neat
it's actually not that bad tbqhfam. I make diagram for brainlet
red dots are red LEDs, dark red dots are IR LEDs, black lines are - supply, white lines are + supply and resistor leads, blue/green lines are LED-to-LED wiring for red resp. IR strings, small black dots are fleas marking - side of LED, tan blobs are hotdogs used as resistors
the notable features of the diagram are the - supply bus running down the middle to which the - side of the 15-strings are connected, and the symmetrical polarity reversals
the top figure shows one way to run the 15-strings (fold twice), the bottom figure shows another (fold once). whichever one suits your autism best
>b-but there are only 90 IR LEDs
yeah, just run the other 15 around the left and top edge however you like

[Anonymous No.1345839]

>>1345645
>filename
ayy

[Anonymous No.1346065]

>>1345645
Oh wow, thanks I really appreciate the time and effort you put towards this. I saved the picture with your description for further pondering. Also, great filename. lol