Circuit for 12v LED PWM controlling?
Crosswinds
Posts: 182
Hi Guys!
Iam in the need for a LED controller for my fishtank, and will simulate sunrise and sunset with a prop.
I think i am home on the software side, but am a bit worried about the circuit.
Since the LED´s im using need 12v (G4).
I think the perfect thing would have been to just use a circuit like the ULN2803A for this. But since that one only can control outputs with the same voltage as the controller-side (propeller), that fails.
I hope im makeing my self clear here!
Does someone have a circut-suggestion for doing this?
Thanks!
/Daniel
Iam in the need for a LED controller for my fishtank, and will simulate sunrise and sunset with a prop.
I think i am home on the software side, but am a bit worried about the circuit.
Since the LED´s im using need 12v (G4).
I think the perfect thing would have been to just use a circuit like the ULN2803A for this. But since that one only can control outputs with the same voltage as the controller-side (propeller), that fails.
I hope im makeing my self clear here!
Does someone have a circut-suggestion for doing this?
Thanks!
/Daniel
Comments
The ULN2803A will accept drive voltages up to 30 Volts. The inputs will react to Propeller voltages.
http://oomlout.com/8TRA/8TRA-Guide.pdf
I use these chips to drive 12 Volt Solenoids all the time. Propellers and ULN2803A work well together.
Just have one more question. It says that it can handle 500mA per output, but is this simultanious?
Maybe I'm wrong but I think you have to look at power dissipation over the entire package. You will see in the data sheet that it's limited to about 2.25 watts for the whole package. And wattage = Volts x Current. So it's not just a question of current - it's also how many volts at which you are running.
Power Dissipation
(one Darlington pair) 1.0 watts
(total package) 2.25 watts
http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00000179.pdf
Hello! Thank you for your answer!
I have now measured the current of thise devices.
they use 130mA.
12 x 0,13 = 1,56w
Would that mean that i will draw 1,56 + 1w = 2,56? And it can only handle 2.25 total? :S
Let's say it was 0.5 volts. So he power consumed in the chip when on is I times V = 130ma times 0.5 volts = 65 milliwats.
Seems like it should be quite safe.
Beware! The 2803 is a darlington configuration which means that the output will not saturate below 0.8V. The power calculation is based of course as heater infered on the voltage drop across the device and nothing to do with the supply voltage. Frankly I don't see why good old discrete transistors aren't just used. Even at the smd level the tiny BC817s easily handle 500ma each.
EDIT: Just read the DS and it quotes around 1.1V max saturation voltage for a 100ma load.
Oh got a bit confused before this, i thought i should calculate with the wattage of the load!
Should i hook this up, and measure the voltage on the load directly to see the voltage-drop? Then calculate with that?
I have several 2803 laying around, and i tought it would be a neat package! But ofcourse a regular transistor should work. Do you have any you can recommend for this? No use for SMD´s here.
I confess, I was confused by what that really meant, too. Thanks to Heater and Peter for clearing that up.
Input from the transformer: 11,97v output from the ULN to the LED´s 11,17v and one LED draw like ~110mA when it gets 11,17.
I cannot see very much light-power change due to this, so maybe im in the clear with that.
But could i feed 4 of these with the ULN and be in the clear? because i guess it only can handle 500mA all together and not for each channel?
What do you guys think about this?
Ofcourse it would be great to in some way be able to feed them 12v. But that is not really a problem.
The led intensity is a function of the current through it, not the voltage. There is probably an internal current limiting circuit of some type built in to the led. As long as the voltage is above a certain point it has very little effect on the intensity.
The ULN2803 can not drive all 8 outputs at 500mA, it can however drive all 8 outputs simultaneously at 110mA per output. There was a similar thread a while back using the ULN to drive solenoids and the data sheet with the derating info was posted there.
The 12V is a nominal rating. If you look at the data sheet you will find that it can be 12V +/- x% (typically +/-5%) or has lower and upper voltage ratings listed somewhere.
0,8v
Thank you for clearing this up! Then it should be possible to use all the gates, since they pull about 100mA each then!
In short what are those things and would using PWM supply on them work as expected?
I from what i understand they are in series and use a current limit resistor. There is no integrated PSU or such stuff. but there is a rectifier. T
hese are the most simple ones out there. They have just made them work with 12 volt.
Ive changed my mind about using the ULN´s for this. I would rather have a way that i can feed the all 12pcs of LED´s through. And not having separate cables for each one!
It will be something like 1.5-1.6 amp 12v (for all twelve together) from my measurements. They (according to the manufacturer) draw 1w per pcs.
What would you recomend for this? Somekind of transistor? I do have TIP102´s laying around.
Check out Jon's Nut's and Volts article:
http://www.parallax.com/Portals/0/Downloads/docs/cols/nv/prop/col/nvp9.pdf
and another that incorporates DMX:
http://www.parallax.com/Portals/0/Downloads/docs/cols/nv/prop/col/nvp3.pdf
If you can get a schematic of the LED board, that would be nice
Firstly, thank you for the links! Will check them out.
Secondly im sorry if my two threads got blended together. They were meant as two. One for the hardware, and one more for the software bit!
But i guess we could stick with this then!