Battery doesn't power microcontroller as expected

[bonbonbaron]

So I got a Launchpad from Texas Instruments. It came with an MSP430G2553.

I programmed this thing to play different notes through a speaker-- while the IC was still attached to the LaunchPad. Then I took it out of the launchpad, plugged it into the same breadboard as all the other components for the speaker , and attached 2 AA batteries (3V total) to Vcc and a 3.3 uF input capacitor. I connected the GND pin to the negative terminal of the batteries.

Now, however, I only hear the speaker make one click when I turn it on. Is powering a microcontroller not as simple as connecting a battery to its Vcc pin?

Thanks
B3

 

[(*steve*)]

Does it work when you use some other power supply?

 

[bonbonbaron]

Yes, it does when plugged into the launchpad device (which in turn is powered by my PC via USB connection). Sorry, I didn't make that clear.

I'm gonna look through the documentation and make sure I set it up to use its own internal clock when separated from the launchpad.

 

[bonbonbaron]

Unfortunately making sure that the clock was set correctly didn't fix my problem. Multimeter shows there's indeed a connection between the wires I was connecting to the batteries.

Something's wrong. When I attach the wires to the batteries, I hear the speaker click, and it doesn't "unclick" until I disconnect the batteries. It's acting like there's no microcontroller there at all... as if my batteries are only wired to a speaker alone.

Any other useful things I can do with my multimeter here? I also tested the overall resistance between the wires: 140 kΩ. So with 3V batteries, that's a 20 μA current draw.

The speaker's an 8 Ω, 5W speaker. Don't think that matters much as it still clicks.

Hmmmm... You pros have any pointers?

Thanks
B3

 

[HellasTechn]

Yes, it does when plugged into the launchpad device (which in turn is powered by my PC via USB connection). Sorry, I didn't make that clear.

re you sure that it can run on 3 volts and that it does not require 5 ?

 

[bonbonbaron]

re you sure that it can run on 3 volts and that it does not require 5 ?

I'm sure. The documentation recommends around 2V-3V Vcc. However I had that thought too-- I tried as much as 9V (very quickly before anything could fry), but it did no good. Your suggestion just now made me doubt, however, so I went back and tried 6V. Still nothing more than a click upon connecting to the batteries, and unclicking upon disconnection.

 

[BobK]

Is there a reset pin? Are you driving it with something?

Bob

 

[BobK]

I looked at the datasheet and answered my own question. Yes it does have a reset pin.

Try a 10K pullup on the ~RST pin.

But...

You probably killed the chip when you put 9V, or even 6V on it since the absolute max is 4.1V. Does it still work in the launch pad board?

Bob

 

[bonbonbaron]

Try a 10K pullup on the ~RST pin.

But...

You probably killed the chip when you put 9V, or even 6V on it since the absolute max is 4.1V. Does it still work in the launch pad board?

Luckily the launchpad shows the chip to be alive and well. I'm back to 3V.

I've also implemented your suggestion for the 10 kΩ between power and the reset pin. (Also tried that with a button in between, in case the voltage wasn't supposed to be constant. Did it both ways-- a button that would stop current to the reset pin when pressed, and one that would conduct current when pressed.) Nothing yet.

 

[BobK]

Just looked at the user guide.

It calls for a 47K pullup and 2.2nF to ground on the ~RST pin. Try that.

Bob

 

[BobK]

And, looking at the lanchpad schematic, it has a 47K pullup and 10n capacitor to ground.

Bob

 

[bonbonbaron]

Looks i need to order that sized capacitor as well as just 7k more in resistors. (My 40k and 0.1uF didn't work.) Thanks for your help Bob; I'll let you know how everything plays out when I get those things.

By the way would you mind sharing with me where you saw those things in the user guide?

Thanks,
B3

 

[Bluejets]

I tried as much as 9V (very quickly before anything could fry), but it did no good.

As Bob says, this is never a good idea and will only end badly.
In case you are not aware, electricity travels at the speed of light (186,000 miles per second).
I doubt you are quicker than that.

 

[bonbonbaron]

Thanks, I'll pay more attention to the specs next time. (But not faster than light, huh? Just watch ;-)... but there wouldn't be much to see, if you know what I mean.)

 

[BobK]

I would expect 40K and 100nF (.1uF) to work. So maybe something else is wrong. Datasheets for PIC microcontrollers generally have schematic for the minimum hardware configuration. I could not find same for your TI part.

Do you have any smaller capacitors? In the range of 1 to 50nF?

Also, did you try the decoupling cap I suggested?

You might want to try 4.7K and 100nF, which would give the same time constant as the 47K and 10nF that the launchpad board uses. Basically the resistor and capacitor are used to hold ~RST low for a short period of time, while other things stabilize, then go high and hold it there. The rate at which the voltage rises might have to be withing certain bounds.

Also, if TI has a forum for user of the MPS series, you might want to try there.

Bob