Axon USB not recognized anymore

[robinson69]

Hello,

I use an Axon (great product) since about 6 months without any problem. But since yesterday, I have the following error message when I connect to the USB (on Windows XP SP3): "USB device not recognized. Once of the USB devices attached to this computer has malfunction, and windows does not recognize it."
The last time it was working was 2 weeks ago. The downloaded program was working also (everything was working in fact). Yesterday, I tested again my program (I didn't use the Axon during 2 weeks). It was still working. Then I wanted to get some data on my PC. I stopped the power supply of the Axon, I connected it to the USB of the PC and I got this error message. It is the only thing that doesn't work, the program still works.
I tested on another PC, with another cable, same problem. So it seems to be a problem on the Axon.
Between 2 weeks ago and yesterday, nothing has changed in the connections in the application.

Some informations on my configuration:
- the Axon is powered by a 12 V power supply (not a battery)
- I have a gyroscope connected to the 3.3 V supply of the Axon (http://www.sparkfun.com/commerce/product_info.php?products_id=8955). The output of the gyroscope is connected on ADC2, the 2 other inputs (PD and ST) are connected on the ground of C1 and C2.
- a H-bridge (http://cache.national.com/ds/LM/LMD18200.pdf) is connected to A3, A7, H5 (PWM) and ADC8 (current measurement)
- a quadrature encoder (http://www.robotshop.ca/encodeur-quadrature-easy-roller-300-1.html) is connected to 5V of Axon, ground of Axon, E7 and C3

Please tell me if you need more informations.
Thanks in advance for your help.

[SmAsH]

have you tried connecting the axon with nothing attached?

[robinson69]

I forgot to tell it, but yes, I tried with absolutely nothing attached.

[SmAsH]

http://www.societyofrobots.com/robotforum/index.php?topic=6436.0
do a continuity test on the usb pins, this may be your program.

[robinson69]

I've just made the continuity test.
If I start with the 2 pins that are shorted (the 2 at the right when you look at the connector on the Axon): they are shorted, OK. But when I put current to these pins, I get also some current in the 3 other pins.
When I test the 3 pins at the left: the continuity is OK, but I get also some current (less than in the selected pin) in all the other pins.
To summarize: if I put some current in a pin, the continuity is OK but I get also some current in all the other pins.

[SmAsH]

okay... so what you are saying is that there is a possible short?
sorry, its a hard thing for me to understand... ive never used an axon...

[Admin]

I have some bad news - chances are the USB is permanently fried. I've seen this error on some prototype Axons that I badly hand soldered, and wasn't able to revive the USB afterwords. You can try and add solder to the 5 pins on the USB connector port, but doubt that'll fix it.

Likely a power surge somehow went to it through the 3.3V connection and damaged it. Or possibly mechanical or thermal shock knocked the USB chip loose but is unfortunately near impossible to resolder.

You have several options beyond buying a new Axon:
- buy a serial to USB adapter and use a different UART port, ~$15
- buy bluetooth, ~$60

[robinson69]

More precisely, did I fry the CP2102 ?
Another question: where does the 3.3V supply come from ? It is still working. Can I still use it or is it dangerous for something else on the board ?

[SmAsH]

yup, admin thinks your cp2102 is fried...
i think the cp2102 provides the 3v3 regulated output... and gets its voltage from the regin?
ive looked at both the axon and cp2120 datasheets and am confused?
the onboard 3v3 regulator gets its 5v from "regin" (pin 7) so could it power itself from its onboard regulator?

[Admin]

Yea, the 3.3V reg comes from the CP2102 USB chip. Odd how the 3.3V still works!

The ATmega640 is buffered from the USB chip, meaning this malfunction won't hurt the rest of your board. So there isn't any reason you couldn't still use the 3.3V. But you may want to make sure the 3.3V is still clean (multimeter/oscope it) and won't damage anything using it.

If you want I can mail you a free replacement CP2102, but you'd need a hot air soldering pen and some skill/patience to swap it out. My first try took 40 minutes and a fried CP2102 to get right


smash, the CP2102 takes 5V from either the USB line or the 5V Axon reg (whatever is higher) and converts it to 3.3V.

[robinson69]

Thanks for your proposition but I don't have the tools and the skills to replace the CP2102 chip. I will use a serial to USB adapter on another UART.
My skills are in control, once I have finished my application, I will publish it. It is a rotary inverted pendulum controlled by a LQG regulator (state feedback + Kalman filter), it may be interesting for some people.

[robinson69]

Should I use a new bootloader.hex that works with another UART ?

[Admin]

Unfortunately you will need a hardware programmer to burn in a new bootloader (which yes, can work through a different UART).

Or you could solder your tx/rx wires from your external USB adapter to the pins of the ATmega640 directly. (if you have steady hands; just don't get tx and rx mixed up or shorting)

If I remember right, I posted up Axon bootloaders for each UART somewhere on the forum . . .

[awally88]

As a side note isn't Windows XP SP3 bad for Axon connectivity too? I remember reading something about it somewhere on here..

[Admin]

As a side note isn't Windows XP SP3 bad for Axon connectivity too? I remember reading something about it somewhere on here..

http://www.societyofrobots.com/axon/axon_FAQ.shtml#boot_slow