Appropriate Power Supply for LED Display

Go To Last Post
9 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have been working on a low resolution LED display controlled by a atmega164. it controls a 3x3 matrix of cells/pixels each with 3 LEDS. That means 27 LED's total per module. On average the system draws 0.25A. The completed system will be comprised of 54 of these. thats means about 14A at 5V. Any suggestions on a power supply/ distribution system? I was thinking about something like a high voltage power supply stepped down to source more current.

Here's a video of the module if you are interested.http://video.google.com/videoplay?docid=7051505557301001150&ei=WsJeSciVIoSM-QG0zbjQDg&q=pskled&emb=1

Thanks,
Andy

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Are you running this on 120 VAC?

Separate power supply(s) for the micro(s) and the LEDs?

Do you desire one big supply, or several smaller supplies?

The average current is good to know, but you will also need to determine the Peak Current.

Jameco, (The first catalog in hand's reach), has this 5V, 30 Amp, 150 W enclosed power supply: Jameco Number: 201912, listed as $75 quantity one in the catalog, but $96.95 on line..., seen Here .

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes It will run on 120 VAC, the micros are separated from the leds through mosfets so it is an option to have them seperate, I think thats probably a good idea. What do you think? This will eventually hang on a wall, so a single wire coming out and to a socket would be ideal, but I'm not sure if its practical in a power distribution sense. I suppose If i did the math and if every single one of the 1440 leds was completely on at the same time at 30 mA then the peak would be 43A.

Thanks,
Andy

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I would use a separate power supply for the micros.

If you elect to use the same power supply then make sure you filter it well before feeding the micro(s).

Unless your software will specifically prevent everything from being on at once, even as an unexpected error condition, or while developing your code, you need to allow for that possibility.

If you have to keep replacing a fuse, or pushing in a pop-out circuit breaker, it will get old fast.

You can search around for similar power supplies from Jameco, Digikey, Mouser, etc. You may find a better price, higher efficiency, smaller package, etc. I just picked that one as a starting point.

You could put two of the above in your project and move on to the next phase of its development.

43A * 5V = 215 Watts. I don't know what the efficiency of your power supply will be, but for the sake of argument and example, say 50%, so your power supplies will generate 215 W and the LEDs could generate < 215 watts of heat, which your enclosure must accomodate. (Think: FAN).

500 Watts on 120 VAC is not that much, one line cord will be fine...

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If you're not too worried about the shape and size of it, you can get a really cheap one, which'll supply plenty of juice by modding an ATX PC PSU...

There are plenty of tutorials out there, but you could start by having a look at http://www.wikihow.com/Convert-a-Computer-ATX-Power-Supply-to-a-Lab-Power-Supply to get some ideas - then it'll typically give you regulated output up to about 30A at 5V, or 12A at 12V...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

firestorm099 wrote:
If you're not too worried about the shape and size of it, you can get a really cheap one, which'll supply plenty of juice by modding an ATX PC PSU...
Here we go again :-(

Those simple "converted" PC PSU's are a fire hazard when used as a lab power supply. The "conversion" you pointed to shows exactly how not to do it. There is no sensible current limit.

These days 600W or 750W models aren't uncommon, and the first 1GW models have been spotted in the wild. They can put out a lot of current in normal operation. Even a low spec 300W PSU can do that. Enough to melt your cables, your breadboard and all your electronics.

In short, using such a simple converted PC PSU as a bench power supply is stupid.

Stealing Proteus doesn't make you an engineer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Meanwell make some nice SMPS DC power supply's- worth checking out.

oddbudman

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Here we go again :-(

Apologies, I hadn't scoured the forum to see that these discussions had already been had...

Not knowing a lot about these things, I suggested that site as a reference point. Thankyou for pointing out that it's the wrong way to go about it - I hate giving dangerous advice :(

I wasn't suggesting it be used as a bench PSU (as the website suggests), but in a dedicated application where current draw is likely to be 'relatively' low.

After looking at a few projects on the web (of reasonable repute) where 5V at 3A are required, this method has been used (although not necessarily the link I posted previously). Is there a SAFE way to do this, and avoid the extortionate cost of a high-current switchmode supply?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I just realized I messed up my math with how many leds I would have. The max current would actually be 29.8 A. Any recommendation on a good cheap 30A 5V PSU?