cancel
Showing results for 
Search instead for 
Did you mean: 

A word of warning... (and a request for explaination)

dciliske
Associate
Posted on September 23, 2013 at 21:45

So, I'm currently working on a side project at home for a LED driver/controller board. In the process of this, I've decided to use a STM8 for some additional A/D and to offload the averaging. I got the eval board on Friday and spent some time getting things up and running on Saturday. When I got to the point of programming the board, I couldn't communicate with the board; ''Ok, somethings wrong with the driver/software tools setup''. So, I get out my STM32 discovery board which I've played with some before and had working previously and try to get that to work. 

In the process of doing all this, I drop the STM32 board onto my oscilloscope's ground connector, and temporarily short the power to earth ground. And my PC turns off. Ok, so I shorted out my USB port, that was dumb. Guess I'll reboot. Oh, my computer is still off. And it won't POST. And it says somethings wrong with the memory. And maybe the motherboard is fried. Really? I thought this wasn't supposed to happen with USB devices?

Oh, the ST eval boards don't have input fuses? How? I thought that was a requirement of USB devices? Be very careful with eval boards and I guess use an externally powered USB hub if you need to hook them up and they don't have an Vbus fuse.

TL/DR

: Accidently dropped an STM32 dev board onto my oscilloscope ground probe and fried some major component(s) on my home PC, apparently because ST didn't put an input fuse on the Vbus line per spec. It's also probable that there was one missing on the motherboard too for this to happen but... -.-

Now, I'll say this, I'm really liking ST's parts. The STM32 is a great lineup. I just wish they put the same thought into the dev/eval board that they did into the chips themselves...

#problem #pc #fried
3 REPLIES 3
Posted on September 24, 2013 at 08:42

>

Oh, the ST eval boards don't have input fuses? How? I thought that was a requirement of USB devices?

No.

It's the ''responsibility'' of device not to draw more than ''unit load'' (100mA) until configured, and the requested current after that. Your ''device'' (the combination of module plus the dropped probe) clearly violated that.

Overcurrent protection is specified in chapter 7.2.1.2.1/USB2.0 , which says nothing on the devices. However, it says this:

''If an over-current condition occurs on any port, subsequent operation of the USB is not guaranteed, and once the condition is removed, it may be necessary to reinitialize the bus as would be done upon power-up. The over-current limiting mechanism must be resettable without user mechanical intervention.''

So, it's clearly your PC's designer fault, that he did not implement proper overcurrent protection as per specs.

>

TL/DR

:

What is TL/DR?

JW

dthedens23
Associate II
Posted on September 24, 2013 at 17:02

I have never seen a fuse on a USB device.

I once wrote drivers for USB host and I have seen hundreds of devices.

The host (PC) must provide 500ma over current protection (generally 700 to 800ma trigger)

I have seen some cheap PC that use a 1 Amp re-setable polly fuse (often to much, to late)

but it would have to be real cheap to offer no protection

Posted on September 24, 2013 at 17:26

What is TL/DR?

Too Long, Didn't Read - Executive Summary

Sorry about the PC, but it probably would have got nuked if you stuck a USB cable from it in a can of Cola too.
Tips, buy me a coffee, or three.. PayPal Venmo Up vote any posts that you find helpful, it shows what's working..