Go Down

Topic: Power issues powering multiple arduinos (Read 3792 times) previous topic - next topic

gregj64

Hi all,

I'm using a 9v input to power 3 arduinos at the minute, it seems to work ok but as of today one of my arduinos seems to have a fried regulator. It works fine over USB but when I plug it into the 9 volt supply the power pin barely lights up and it doesn't run.

Could anyone shed some light on how I am going wrong?

I have the 9 volt supply coming in and onto some stripboard where the two lines feed off to the Vin and GND of the 3 arduinos in parallel. Is there a safer way to do this?

Thanks!

RuggedCircuits

It actually doesn't sound like you're doing anything wrong. Is the regulator on the bad board getting hot at all? Hotter than the other two boards?

--
Check out our new shield: http://www.ruggedcircuits.com/html/gadget_shield.html

gregj64

Thanks for the quick reply!

I checked it there, it doesn't appear to be getting hot at all.

I plugged the 9v DC directly into the dodgy arduino and it didn't power at all but the other arduino plugged in parallel did.

Also when it is by itself the power light is VERY dim but when I plug the shield that I am using with it (its a datalogger shield) the light turns off completely.

About a week ago I was testing a different DC in, a 2amp one, and it was faulty and blew up one of the arduinos (Quite literally, it was very dramatic) Could the second one have been slightly damaged and limped along for a couple of days before it broke?

RuggedCircuits

Quote
I plugged the 9v DC directly into the dodgy arduino and it didn't power at all but the other arduino plugged in parallel did.


There's only a diode between the 9VDC direct input and the Vin pin so unfortunately that doesn't give too much more information. At least your diode is working :)

Quote
Also when it is by itself the power light is VERY dim but when I plug the shield that I am using with it (its a datalogger shield) the light turns off completely.


Makes sense...the more current you draw from the regulator the lower the voltage gets. I don't suppose you can whip out a DMM and actually measure what the 5V pin is?

Quote
About a week ago I was testing a different DC in, a 2amp one, and it was faulty and blew up one of the arduinos (Quite literally, it was very dramatic) Could the second one have been slightly damaged and limped along for a couple of days before it broke?


If you had them all connected in parallel and the faulty supply had much too high a voltage then yes, it's possible.

--
Check out our new shield: http://www.ruggedcircuits.com/html/gadget_shield.html

gregj64

when I placed the probes on the center and right pin it read 2.03 volts, not a good sign I'm presuming!

So if I blew both voltage regulators can I assume it was the faulty DC in I was using and that the way I'm going about powering the arduinos in parallel into their Vin ports is safe if I swap out the second faulty board?

RuggedCircuits

Quote
when I placed the probes on the center and right pin it read 2.03 volts, not a good sign I'm presuming!


Errr...what's the "right pin"? It depends on the way you're looking at it. The middle pin is the output voltage. Measure that relative to GND.

Quote
So if I blew both voltage regulators can I assume it was the faulty DC in I was using and that the way I'm going about powering the arduinos in parallel into their Vin ports is safe if I swap out the second faulty board?


It could be the power supply you were using. Can you measure its output voltage when it's not connected to anything?

Powering Arduinos in parallel as you are doing is safe.

--
Check out our new shield: http://www.ruggedcircuits.com/html/gadget_shield.html

gregj64

thanks again for bearing with me!

The output of the DC in is settled around 13.25V. Is that ok when its not under load?

I can't get any reading from the output of the regulator out now (I stupidly was reading the wrong setting before with that last comment, sorry!)

I can order a new power supply if mines dodgy, is there any max current I should look for or will the arduino just draw as much as it needs?

RuggedCircuits

Quote
The output of the DC in is settled around 13.25V. Is that ok when its not under load?


Hmm...yeah, that's just fine. I was kinda hoping it would be really high which would explain why things are blowing up. Now...not so much of an explanation.

Still, with 13V in and 5V out the power dissipation will be 8I where I is the current draw. Just the arduino itself is about 0.03A. Are you powering anything else off the Arduino (you mentioned you had a datalogger shield...any idea how much current that draws?)

Once 8I exceeds about 1 (Watt) you might start having thermal issues with the regulator. I'm not favoring this as a theory though as I would have to be about 125mA and I don't think you're anywhere close to there.

If you are going to order a new power supply get something closer to 5V but above 7V. 9V supplies are usually a good start and will minimize regulator heating.

--
Check out our new shield: http://www.ruggedcircuits.com/html/gadget_shield.html

gregj64

OK brilliant, I can't measure the dodgy dc in that blew my arduino in the first place (I had to throw that out)  

but the one I'm using now is 1amp @ 9 volts written on it. It seems to be ok, (its the one that runs at 13 volt under no load).

As far as current draw the 3 arduinos I use have different draws, 2 have datalogging shields on them and draw maybe 100mA (At a Max) and the 3 is powering 3 servo's on a shield I built, they're low power, the most I've measured that arduino drawing is 860mA.

So a new power supply that can give 9v at about 1.5amp would be good? I want to upgrade from the 1Amp I'm using now for greater potential draw on the servos, that's why i tried the dodgy dc in that blew everything in the first place.

RuggedCircuits

Quote
As far as current draw the 3 arduinos I use have different draws, 2 have datalogging shields on them and draw maybe 100mA (At a Max) and the 3 is powering 3 servo's on a shield I built, they're low power, the most I've measured that arduino drawing is 860mA.


WHOA! That would be the problem.

860mA is too much to draw from the Arduino's voltage regulator when the input-output difference is 4V (9V --> 5V) as it means the regulator is dissipating 4*0.86=3.4W. It is going to overheat and possibly fail (or in your case, already failed).

You're going to have to power your servos separately (5V regulated supply at up to 1A) Connect the grounds of all power supplies together.

Arduinos provide logic, not power  :)

--
Check out our new shield: http://www.ruggedcircuits.com/html/gadget_shield.html

gregj64

Sorted so, I'll power the servo's separately and hopefully there'll be no more expensive issues with voltage regulators exploding!

Thanks so much!

Go Up