Works with 2 power sources, but not 1...

Hi Everyone,

I've got a fading LED thingy going. Yes, thingy is a technical term.

Anyways, I have an UNO that is driving 2 LUXDrive BuckPucks. There's also a momentary switch attached to an interrupt pin controlling program behavior and such. I have the UNO on a 9v battery and the 2 buckpucks on a seperate 12v L-ion battery. Ideally i would like to have only the L-ion battery in the system. Is this possible? When i tried to connect the drivers and the UNO to the battery, i lost output one the LED's. If i have to guess it would be a level reference thing, or the driver is sensing a different relationship to the control, or maybe a well placed diode... This theory is where i need to get some help - If you guys could take a look at the attached drawing and let me know what you think, it would be super helpful.

Cheers - Jon

Just taking a quick look at the schematic, I did not see a shared ground connection between your two power sources. All components need to share a common ground (AKA reference).

Hey thanks for the quick reply ---

the drivers have a separate power source completely. The 'ctrl' pin of the drivers are connected to the UNO, and the 'ref' connected to the UNO. I could not get the thing to work when i connected the two to the same power source in a sense making a common ground?? make sense?

(the switch is connected to the ground pin on the UNO, this isn't supper clear on the drawing)

cheers

make sense?

Nothing you just said makes sense.

All components need to share a common ground (AKA reference). In other words ALL ground pins should be tied together (unless you are using opto-couplers or transformers.)

okay...

for conversation sake. if i were to connect a 24v battery to the LED drivers, totally acceptable to the drivers spec's. what would be the best way to power the UNO, as 24v is outside the spec for the UNO?

if you say a second battery, perfect, that what i have, but there is no shared connection to the batteries, and when i share a connection between batteries, it stops working. could do something as simple as connect the (-) of the two batteries, with out fear of blowing up the UNO?

thanks

"Best" is an undefined term. There are lots of ways to convert 24V to something in the Uno's range. A simple LM7805 could be used (in which case you would connect directly to the 5V pin) or maybe even a LM7812 (would create less heat than the 7805). Or you could build a buck converter.

could do something as simple as connect the (-) of the two batteries, with out fear of blowing up the UNO?

That's your only option. You have to connect the negative of both batteries, otherwise you won't have a common ground.

Imagine this situation:
You have two Batteries: Battery1 and Battery2. Battery1 is 12V while Battery2 is 9V.

If you measure the + and - on Battery1, you will get 12V.

If you measure the + and - on Battery2, you will get 9V.

What would you measure if you connected to + on Battery1 and + on Battery2?
You would measure, 0. There is nothing common between the two. Connect the (-) terminal of both and repeat. (Then what would you measure?)

sounds good, i have some experimenting to do. in my projects current form it's installed and not very accessible, I'll build a breadboard version and see what i get.

if i imagine the battery question correctly, its not series, nor parallel - so some form of an equation would give me the anwser... so it is easier to build the actual and just measure it... :slight_smile:

thanks for the help, i'll get back to ya (work is finished, gotta catch the bus)

_J

hmm... so i guess plugging 11v into the 5v output on the board isn't the best idea, "hmm, i wonder what that smell is..." any guess if it's just the chip or other parts on the board? :confused:

there's a good chance that would destroy everything.

board 1... yes broken :frowning:

board 2... working right along just fine, just needed to hook the pins correctly. I did get lucky, the drivers output 5v on the reference pin, which was hooked to the 5v pin on the arduino. but this time around, i didn't have 11v also on that side of the bread board, so the board was fine, survived 5v in to the 5v supply.

a diode installed inline, to make sure the reference doesn't get to the UNO. This also solved the 2 battery issue... :slight_smile: it was strange that the board was working when there was no power attached, but the driver was being powered - "where is this power coming from??" it was coming from the driver...

luxdrive buckpuck 5 led's single battery pack

and an updated drawing...

Hi, about those tieing grounds issue.. What about possible potential difference between grounds? Is that an issue? How to solve that? wouldnt different ground potentials change voltage levels (ex: a regular +5V change to +7V if the reference ground got +2V due to potential diferences.) Since boards are not really earth grounded, isnt that a problem?

Thanks!

it was strange that the board was working when there was no power attached, but the driver was being powered - "where is this power coming from??"

If you do a Google search for "AVR back-powering" you'll see this is a common issue with the ATmega. Another case that gets people is when they try to use the Analog Inputs to measure a battery's voltage. If the battery is connected directly to the pin, it will "back-power" the chip.

wow, super intresting, and i would have gotten really stuck on this sometime down the road - thanks for the follow-up.

cheers
Jon