Re: [Discuss-gnuradio] Sensitivity of USRP to power glitches

[Marcus D. Leech]

Juha Vierinen wrote:
> I got a feeling of deja vu when reading your post. I haven't had any
> problems for a long time now, but I never really found the true cause
> of these problems. Still, in my mind I have hypothetically correlated
> the following with protocol and  usb_reap_async errors:
>
> 1. Loose power supply connector
> 2. Bad USB cable
> 3. Using a non-USRP under-rated power supply
> 4. Old PC with a piece of shit USB controller
> 5. Fiddling around with daughterboards while USRP is live ;)
> 6. Static electricity during the winter
>
> But as I said, I haven't seen these problems for a long time now. I
> have managed to avoid causes 3, 4 and 5 during the last year or so,
> but as I said, I don't really know the exact cause for the errors as
> they were fairly sporadic.
>
> juha
>   
My power supply is more than adequate: 7.5V at 4A   But I've seen this
with other power supplies as well (even the as-shipped one).
Many different USB cables over the years.
Modern PCs
I'd never fiddle with a daughtercard live!
Seen static problems, too.   My usrp is inside a large aluminum chassis,
completed surround by ground (the chassis is grounded).
  Still static is a problem sometimes.

The vacuum cleaner incident this morning gave me a strong correlation
event--USRP had been running just fine, and then
  a little big of AC line "sag", and USRP runs off to never never land.

When I had the system deployed at the SBRAC site, it ran for several
days without a hiccup.  They have well-conditioned
  power at that site, so that might explain the "longevity".

My suspicion is that USB is more sensitive than you might expect to
ground noise, but that's just wild speculation on my part.

-- 
Marcus Leech
Principal Investigator, Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org