Another smoking Leonardo

Projects General Guidance
1

Very much a disillusioned new kid here.

I connected as per numerous getting started/first time users , a single l.e.d. and typed in the code to make it blink. Uploaded it via USB and it worked fine. I then disconnected the USB cable, and connected a 12v wall wort psu. Again, it worked fine. So far so good, and onto something else for a newbie.

Wired up a Hitachi compatible 16x2 LCD display as per Arduino cookbook, and uploaded the code from this website, (which is identical to the book). After a couple of false starts, (contrast on pot needined winding all the way up), it was working fine. The Hello World and seconds appeared and stayed running for a couple of hours.

Wanting to switch my computer off, I unplugged the USB cable, and plugged in the 12v wall adapter again.

The program did not start, but smoke started to appear in the area of the 12v connector socket on the Leonardo board. I can see no scorch marks on the board, but I wasn't leaving it plugged in to find out where it was coming from exactly. No wires were moved or shorted, and I did check.

I'm now unable to communicate with the board at all, and suspect it's cooked. Not impressed since I only followed instructions. (Yes I know they all say that...)

So, what have I supposedly done wrong?

I ask because although I'm prepared to buy another, I'm only going to do it once before moving onto something more robust.

2

Did you have the board sitting on something metal? Maybe it touched a screwdriver or something?

3

it could be the regulator, hopefully it burnt out protecting the rest of the board..

take a picture?

4

Thanks for the replies.

I'd point out that although I'm a newbie to Arduino, I'm not that new to either electronics or this type of microcontroller, having previously owned a Basic Stamp, and built stepper motor controllers with it. I gave up with it some years ago, and only decided to have another go a few weeks ago. It was a choice between a Raspberry Pi, or Arduino. For no particular reason, I chose Arduino, but am beginning to wonder if that was a mistake, in light of my experience.

What I'm trying to put into context here is that I know enough not to leave a board on a conducting surface, and what precautions to take when assembling and running electronic items.

So it's particularly galling that on only my second attempt at using the Leonardo, the thing should start smoking and now fail to operate. The board functioned perfectly on my USB cable, but one of my objectives is/was to have the thing running independantly powered, driving motors and displays. If I can't even power the thing according to instructions, then to me, it's useless.

I'd happily take a photograph of the board, but it would only appear as a catalogue photograph. There is no evidence of from where the smoke was coming from. Even if I knew which part had cooked, I couldn't replace it since the whole board is surface mount.

The wall wort (or whatever you want to call it) is a 12v supply for a router/modem. I returned it to the router just to see if that would work, but it powered up perfectly.

If I now attempt to plug in the Leonardo via USB, the ON led comes on, then fades to nothing. I'm pretty sure it's cooked, but still have no idea why.

I've ordered a couple more (1 x Leonardo and 1x UNO), but if either of those follow suit, then I'll have to look at the alternatives.

Howard

5

I'd never use 12v on an Arduino unless the load is very light, otherwise the linear regulator will turn into a stove.

To keep heat down, supply 7v to the dc barrel, and stick to low current, anything over 30ma use a transistor to power your loads (unless you need more power, fets, relays/triac etc...

My concern is that the atmega fried..

I'll be honest I always use a 7805 (from a 12v supply) and feed it directly in, I'm yet to see that setup fail... but this is the catch the real arduino board should have been fine, the cheap clones use some el cheapo chinese regulator which is useless for any decent amount of current.

6

I have quite a large number of Arduinos, particularly the Uno variety, and I think I only ever zapped one, and that was probably with static because of the damn carpet. Assuming you know enough to be cautious with electronics (like with stray metal, and not confusing an AC power supply with a DC one) then I can't really explain what happened. However I have to say it would be very rare, in my experience.

I've used the stock boards, soldered together my own, and made them with breadboards. Catastrophic failure is practically non-existent. And I'm the sort of person that makes a new circuit every day or so in order to answer a forum question.

Here's one example, and there are quite a few others on my forum:

My advice is, unless you happen to want the USB connectivity (eg. emulating a keyboard) the Uno is easier to work with. Also it (usually) comes with a socketed chip, so if you happen to blow it up you just replace the ($5) chip.

7

Nick!

That alarm clock is exactly why apps for mobile devices are used... even though this is 1000x easier than using ic counters and logic chips :slight_smile:

(P.s love the alarm clock)

8

Well, it's a learning exercise too. :slight_smile:

9

Many Walwarts are unregulated power supplies which return a rated voltage when plugged into the device they were intended for
so a walwart rated at 12V which is designed for a modem / router will produce 12V when powering the router.
Routers draw far more current that a Arduino does, so using such a power supply on a Arduiono or Leonardo will result in more than 12V being delivered.
The Leonardo is rated at 12V max , and thats the board itself, not including any extra devices plugged in , such as LCD displays etc.
If the LCD display has a backlight, then theres a good chance the on board regulator is fried.

10

mauried:
The Leonardo is rated at 12V max

20V according to the documentation. The 12V supply would have to be way out of spec to get that high even when lightly loaded, althiough it's not impossible. With the LCD backlight running from the 5V supply and drawing perhaps 100mA, the regulator would certainly get hot when running from 12V or more. However, it should go into thermal shutdown before any smoke appears.

11

If you did everything you said then it would not have happened. You are missing something. Exactly what is hard to say because we have nothing else to go on.
As for Raspberry PIs I had two blow up on me, I have never had an Arduino blow.

12

If it's only the voltage reg that's gone you way be able (just for fun) to get it going again by using Vin/5v (well you could if it was an Uno)

Mark

13

Thanks everyone for your input.

As is nearly always the case, the problem lay not in the board, but in the chair. Yes, I had previously plugged in the 12v supply and had it working perfectly. What I hadn't realised was that the second time I tried this, with the lcd display, I used a different wallwart. (Any fool could make this mistake, and clearly I fit the mould). This one was centre ground. Clearly, there doesn't appear to be a bridge rectifier on the Leonardo to save people from this issue, but I'm not blaming the makers.

Doubtless I'll have a zillion questions on the "how do I..." front, but for the time being I'm working my way through whatever examples I think are relevant to my not yet defined goal. Trying to remember all I forgot about electronic components, and learn the syntax and commands of the Arduino's language in the process.

Again, many thanks,

Howard

14

I've just looked at the Leonardo schematic, and I was surprised to find that instead of having a diode in series with the input from the barrel jack (which is what the Uno has), it has a diode between the input and ground. This seems to me to be a strange thing to do.

So if you connect the power backwards, that diode will burn out. If you are lucky, it will go short circuit and protect the mcu. In that case, you can unsolder the diode (the 2 pin black component behind the barrel jack, between the two capacitors) and your Leonardo should work again. If you are unlucky, the diode will have failed open circuit and the mcu will have been toasted too.

15

The RBBB board has that, and I suppose it is supposed to be a crowbar protection. Someone suggested that it would work better with a thermal fuse in series with the supply. The idea of the diode wired that way is no doubt to eliminate the voltage drop (and power consumption) in a series diode.

16

Quite right, a crowbar must always be used in conjunction with a fuse or some form of current-limited power supply.

When the series diode voltage drop on the Uno is too much, you can avoid it by feeding power to the Vin pin instead.

EDIT: also, the series diode voltage drop takes a small amount of power dissipation away from the regulator, which is often a good thing.

17

If the OP often uses different "wall warts" to power up his Arduino, why not take out the diode dc42 referred to earlier then use an external bridge rectifier.
Yes there will be a 1.4v drop, but might be beneficial in some cases and any worries about wrong polarities are then a thing of the past.

Bridge.jpg
Bridge.jpg760×549 37.6 KB