Did I fry my Leonardo Tiny?

Hi all. I'm an absolute novice to electronics but have a tiny bit of programming knowledge. Have been learning Arduino to try and upskill and having fun along the way.

I built a door alarm that used a Duinotech Leonardo Tiny Atmega32u4, an RGB led with resistors, an active buzzer module, and a thin film pressure sensor. I'd made a prototype on an Uno and breadboard before moving to the Tiny and a prototyping board.

I finished soldering everything today and as I began programming the functions I realised the green led pin (D10) wasn't functioning (it had been in the past). I used a multimeter and found that pin D9 had no power, despite being set to digital write High... D10 had 0.5v coming out, and D11 had 5v.

Everything else seemed to be outputting power at 5V when set high, but I noticed the Atmega32u4 chip was getting very hot to the touch.

I checked continuity of all the different connections I'd made and couldn't detect any shorts. I unsoldered the wires from D9 and D10 just in case, but now the board isn't detected by my laptop. I'm wondering if I've damaged it in my novice heavy handed approach to electronics?

I'm assuming that the Atmega32u4 typically shouldn't get very hot?

Your assumption is correct, it should not really get warm!

Without your wiring schematic, and a picture of your wiring, any advise given would be guess work! But the chip getting hot usually means you have a short circuit to ground somewhere! Probably on a GPIO pin.

1 Like

Did you make a prototype with a breadboard first? Always helps me.

Thank you.

I guess I will try again with a new board.

I had made a prototype on a breadboard but used an arduino Uno rather than the tiny. I also added the thin film sensor as a last minute change.

I guess another question is is it fine to connect all the I/O pins to the same ground? I'm assuming so otherwise only having 2 grounds and 10 I/O pins is a bit silly.

If this happens again I'll try to do a schematic but those are probably the most alien to me!

Could you also damage the chip from excessive heat by soldering?

Yes but a wiring error is more likely. If you get the polarity of the supply reversed it also causes the processor to get hot, as will excess voltage.

You don’t connect an I/O pin to ground at all, unless it is through a push button and that pin has been set to be an input. Connecting output pins to ground will blow your pins if you set that output to high.

1 Like

If I have a short to ground, presumably the best way to check would be a multimeter with the - on the ground pin and to go around the other pins using the + probe to see if I get a circuit using the continuity mode? I tried that and didn't find anything so am going to wondering if I did something wrong?

Excuse the poor photos and very poor soldering, this was pretty much my first foray into soldering. I think the iron wasn't hot enough for some bits so I likely cooked parts.

I'd removed the board at this point for diagnostics.

I don't belive any i/o pins go to ground unless either the Led, buzzer or pressure module had a fault so that the short was in one of them.

3x I/O pins went to the led with a single connection to ground on the ground pin of the led.

1x I/O to the buzzer, ground buzzer pin to ground on board.

The pressure sensor had 3 pins, 1x I/O, and 1x ground and 1x 5v. They were all connected appropriately.

I think for my next attempt I'll not try to keep the wires as tidy and just do single longer lengths to where they need to go to reduce the risk of a bad solder.

Thanks Mike. I used the wrong words, didn't mean I had connected the I/O pins to ground... More that there are enough pins to connect to other sensors and modules that may in turn need more ground connections than there are pins.

Eg, I had an Led, a Buzzer and the pressure sensor all requiring a ground, but there were only 2 ground connection points.

I do wonder if the pressure sensor is at fault as it was the only thing with a 5v connection.

The middle two horizontal strips, the lower of the two, second joint from the left. It looks like that is not making contact. It looks like the pin is sticking out of the surrounding solder. That happens sometimes when you don’t make good contact with the pin and iron.

I normally bend the wires slightly when pushing them through strip board, one to make sure they don’t fall out when you turn it upside down to solder and two that the wire is touching the end of the hole so that the solder filet needs to be a lot smaller to make good contact.

Chances are you made a mistake and what you thought you were wiring up was not what you actually wired up. A mistake we all make.

When something does not do immediately what it should when connecting up the power I immediately remove the power and check things.

Also when trying out some new hardware always put it the software first before removing the power attaching the hardware and applying the power again. Because you never know what the software originally in it programmed the I/O lines to be.

Do you have a resistor in series with the LED?

What buzzer do you have. Mechanical buzzers take lots of current and generate nasty spikes.

In addition to what Mike mentioned you should first connect power and ground to your circuit. Measure each pin to be connected with a voltmeter and verify it reads what you expect.

Thanks all.

I have I think 440 ohm resistors on the Led as 220 I found was a bit bright still.

I had checked the LEDs as I went and that was fine, got a little confident and added the buzzer and pressure sensor at the same time. When I retry the circuit I will first check the pins of everything with the multimeter first.

This is the buzzer:

And this is the pressure sensor:

I think a simple schematic (hand drawn) would be in order. We've not identified the cause and its not cool to toast a 2nd board.

Sorry for the bad schematic, its my first time! (schematic virgin here)

So D11 had 5v when high, not connected to anything.

D10 connected to the Green led pin was 0.5v when high.

D9 connected to the speaker S pin had 0v when high.

A1 and A0 are digital write (led), A2 is analogread for the pressure sensor.

Schematic is good, nice job.

I cannot find any reason this would damage your board.

I would connect your protoboard (not the processor) to 5V.

Exercise the sensor, leds etc and use your mulitmeter to verify each input is what you think it is and each output the same.

Triple check the pin assignments and input or output assignment.


Thanks John. Can you elaborate on the connecting proto board and not processor to 5v comment?

Just connect the board to 5V and Ground with none of the other pins connected.

You will need two jumper wires to do this. Between your board and the Arduino.

Thanks for the photos but what we are missing is how you connected this to your Arduino. Especially the + signal, where was that on the board? Have you a photo of that.
Did you use header pins or did you just push the wires into the socket on the board?

Thanks Mike. Here we go. I've tried to match the colours of the wires, but have tweaked a couple to make it clearer where they go. Orange = Ground wires.

I didn't use any header pins, everything was just directly soldered onto the board. I've been using jumper wires with the pins removed for the wiring.

The Arduino was attached to the board using a double sided sticky pad. I did wonder if perhaps that might have caused a short on the underside? There aren't any real places to screw it on, so I assume I need to glue it with a non-conductive adhesive, which I'd have thought the pad was...

The only other thing I can think of is that I had to bend the pins of the pressure sesor 90 degrees so that the board would fit horizontally. I'm assuming the board wiring isn't such that the force required to bend the pins would have created an internal short? I guess I will find out when I do some troubleshooting later.

It might do.
Can you post a link to where you got that Arduino from?
I would like to check the pinout and schematic, I can’t quite make sense of the markings in the photograph because it is low resolution