Problem w/ Nano - puff of smoke & goodbye.

Other Hardware 3rd Party Boards
1

Hello -

I am using Windows 7 and UNO R3 Arduino IDE 1.6.5 my NANO's have the micro USB on-board

I got my chicken door sketch running fine with my UNO, and got it working on the breadboard. Then I decided it was time to replace the UNO with a NANO to save space (and because my UNO is the genuine article and I want to keep it for other projects).

So I ebayed a knock off nano, which eventually came in. I tried to upload the blink sketch to it (selecting board:nano, com1), but, while it did blink, it gave me an avrdude: stk500_getsync() error.

So I googled and forumed and went thru the process of burning a bootloader using the UNO. It said the burn was successful, but also gave me the same error. And I still couldn't get it to accept any sketch.

I figured I had just balled things up so bad that I might as well start over. I bought another NANO on ebay, making sure it said that the bootloader was already installed. It came today. I tried to load the blink sketch, and maybe I did because it blinked, but I was getting the same avrdude error. So I decided to upload another sketch. Soon after I hit upload, there was a puff of smoke and the usb cable got very hot very fast (so i jerked it out of the computer).

When i connect my UNO, and I select tools>port, it shows a new port, Com5 (arduino uno), and things are fine. In device manager under ports it shows : Arduino Uno (COM5) & Communications Port (COM1).

But I never got anything but COM1 when I had the Nano plugged in. And that's all it shows in the device manager.

I figured there must be a reason that I am not getting something in the ports when I connect the Nano. So I downloaded the driver that was referenced in the ebay item description. but the driver wouldn't install. So I tried to find another driver, but they all seem to be the same, and won't install.

So I think I have a pretty good idea what the problem is (the IDE isn't seeing the Nano), but don't know what to do about it.

I suppose there is the slim chance that it is not something I did wrong, that it is bad boards. I had tried to get a real Nano from adafruit or sparkfun, but they don't seem to sell them anymore.

Any suggestions ? Is there a way to tell if my remaining board is good at least ?

2

http://www.ftdichip.com/Drivers/VCP.htm

did you install the drivers?

EDIT: If the nano smoked, I wouldn't feel to comfortable plugging back in to pc...

3

Yes, and I would also get rid of the cable the smoked Nano was connected with.

Now, what are you trying to do, get your non-smoked Nano working? With a fresh cable?

If so, the steps I recommend are:
Look on the bottom of the Nano to see what USB chip you have, and install the drivers for that chip.
Do the Loop Back test which is described in one of the pinned posts at the top of the installation and troubleshooting forum.
Depending on which bootloader you put onto the Nano with your Uno, you may need to select the board Uno, or the board Nano, from the Tools, Board menu. Try one, then the other. Let us know how it goes.

In the Arduino menu in Preferences you can checkmark show verbose output during upload, and if you have troubles that info it displays may help us help you.

4

My guess is you slipped while handling the nano and it got shorted out by debris on the table. I've lost prototypes like that.

Most nano clones don't use the FTDI serial adapter, they use the CH340G, so you need the driver for that. Despite the incredibly unpromising installer, the driver and chip is rock solid, and I prefer them over FTDI's chips.

Also, your computer's USB ports are supposed to limit the output current... It's unfortunate that they didn't! You'd have still seen the smoke and lost the nano, but the cable should not have gotten warm...
Mine will let me short the output and in all but one case (out of way too many), have shut the power off before I let the smoke out.

5

Forget about drivers. That is easy enough.

In respect of the chip and the "puff of smoke", the question that matters is - from where did the "puff of smoke" come? Which exact component?

That is the only important question.

And a USB cable getting hot fast is quite absurd. It just isn't going to happen unless some other mains-powered device is plugged into the Nano and you are somehow feeding major current into the ground wire. You clearly should not proceed any further until you have all that sorted out.

Something that annoys me about the plethora of so-called "UNO R3" clones mendaciously advertised on eBay and other sites.

By definition many or most are not UNOs. A UNO is defined by using an ATmega16U2 (or 8U2 in the rev1 and rev2 boards). That is what UNO is. If it does not have an 8U2 or 16U2 then it is simply and undeniably not a UNO. Whether using a FT232 or CH340 or the preferred CP2102, it is essentially a Duemilanove and nothing more. The bootloader software - which may be UNO - is irrelevant to the hardware.

6

Thanks for the reply's.

As far as the smoke & heat. My USB to micro USB cable is 8"L, and it won't reach the desktop, so I was holding the cable (with the Nano connected on the end) in my left hand, away from the computer case so it didn't get shorted. I was watching the IDE progress when either a small sound or maybe a small flash of light made me look down. Then came the smoke and stink. I would say I stared at it for maybe 1 second when I felt the cable getting damned hot in my hand. That is when I jerked it out of the computer.

After maybe 10 seconds, while I mentally re-grouped, I felt the cable again and it was still warm.
I have located the component that burned and took a picture (attached). It is in the upper left.

Whether the cable getting hot is possible or not, it did. Apparently it didn't hurt my computer. I was able to connect my UNO back up (using the standard cable) and successfully upload a sketch to it.

I bought the UNO about 2 years ago directly from Arduino. It cost about $30.

Yes, I am trying to get the remaining Nano to work. I do have another 8"L cable. The chip says CH340G. The driver I tried to install was for that chip, but maybe I went to a bad source. Can you suggest a known good source ? I will try another install.

Thanks for all the comments and suggestions.

7

having trouble getting the .png attached.

8

There must be a short between 5v and gnd on the Nano. Check with tester for shorts.

9

There is a short, but whether it was there before the component burned, I don't know.
Still can't get the blasted image attached.

burned nano 2.JPG
burned nano 2.JPG518×973 192 KB

10

53a391b99dafcb7116506bcd3cd19bfb09a9e3c0.jpg
53a391b99dafcb7116506bcd3cd19bfb09a9e3c0.jpg518×973 192 KB

OK, that is the 4.7 µF tantalum capacitor on the USB power input. This is of course, not an original Nano, but close enough.

Just unsolder the fried component and replace it and you should be good to go - when you get the drivers sorted (or take the hint and convert to Linux :grinning: )

It's not your fault - the mug who assembled the board soldered the capacitor in backwards!

800×463 88.1 KB

11

Non-smoked Nano:
Driver available from the manufacturer's web site

I believe you are looking for ch341ser.exe

12

Paul__B that is a great catch on the orientation of the cap. We can still see the stripe on the blown cap. I have the same variety of board, mine with the cap oriented OK.

13

Lol, that was a good observation Paul. Lets hope the smoke came from the cap and not from other component.

14

yes, this is the driver I already tried to install. But in any case, I tried again, and got the same rather confusing warnings ("it is already pre-installed", etc).

After I wrote that sentence I thought, what the heck, let me see if it sees the Nano. In port, there is now a "COM6". And it works. I successfully uploaded a sketch. hallelluya (or however it is spelled).

Thanks to you all.

15

mart256:
Lol, that was a good observation Paul. Lets hope the smoke came from the cap and not from other component.

Well, it clearly did. That capacitor is directly across the USB power input, so no other component should be involved.

My point in any case - the board should be perfectly repairable.

16

That still leaves the question of the hot USB cable. This guy must be using some really 'Brand X' equipment. How come the USB buss didn't blow?