I have identified the cause of the shorting issue I was experiencing and
thought I would share. The bottom-left corner mounting hole of the USRP
passes very close to the internal heat sink pad for the 3.3V regulator;
this pad is at 3.3V due to the regulator design (tab is hot). Some of
the green board mask had scraped off when the mounting post for the Side
B daughtercards was installed, which brought that mounting post, and by
extension the enclosure, to the 3.3V potential. Photos of the board site
are at http://www.likesgadgets.com/gr/ When the SMA jumpers connected
the ground to the case, a short occurred.
I was also able to reproduce the short out of the enclosure just by
tightening down the mounting screw for a daughterboard into that same
mounting post; the lock washer bit through the daughterboard mask and
shorted the daughterboard ground plane through the mounting post to the
3.3V heatsink pad.
Even though there is no lock washer between the mounting post and the
3.3V regulator heatsink pad, the raised edge of the pad caused the mask
to rub off when the mounting post was tightened (or perhaps earlier, who
knows?)
The short term solution is to insulate the side B mounting post from the
3.3V pad with a very thin insulating washer, scotch tape, or perhaps to
not mount that post. This prevents the enclosure from being connected to
the +3.3V buss when we really do want it grounded. If the washer is too
thick it will cause the daughterboard to bend or not fully seat in the
nearby connector.
IMHO, the long term solution for the next batch of USRPs is to increase
the clearance between the mounting hole and the 3.3V regulator heatsink
pad that the hexagonal mounting stud does not cover or touch the pad.
That will prevent problems for folks with overenthusiastic fingers. 
Best,
Mike
Out of curiosity, what generation USRP do you have? My Rev 3 board has
quite
a lot of clearance on the TX-B side (1/4"ish- not measured)
Weber, Michael J. (US SSA) wrote:
That will prevent problems for folks with overenthusiastic fingers.
To: George B.; [email protected]
the enclosure metal which causes a short. My solution was to loosen
Thanks,
occurs, the regulator gets very hot, and the voltage at the input jack
[email protected]
Discuss-gnuradio Info Page
Discuss-gnuradio mailing list
[email protected]
Discuss-gnuradio Info Page
–
View this message in context:
http://www.nabble.com/RE%3A-USRP-%2B-case-%3D-electrical-short---EXPLAINED-tf3358833.html#a9350646
Sent from the GnuRadio mailing list archive at Nabble.com.
Folks, please trim your posts.
Do not top post like I just did (to illustrate what I’m talking
about). For that matter, don’t blindly bottom post either.
Note the 5 copied messages.
Thanks,
Eric
2007/3/7, Eric B. [email protected]:
Folks, please trim your posts.
Do not top post like I just did (to illustrate what I’m talking
about). For that matter, don’t blindly bottom post either.
Note the 5 copied messages.
I’ve added some notes regarding this to the wiki:
http://gnuradio.org/trac/wiki/MailingLists
–
Trond D.
Eric B. wrote on 2007-03-07 16:11:
Folks, please trim your posts.
Do not top post like I just did (to illustrate what I’m talking
about).
Sometimes I think to myself: No, you don’t need to send me back all that
quoted text, I know it already.
I call that Tofu: Top-quote, full text underneath. (acronym works even
better in German)
For hints and rationales behind quoting, have a look at
http://www.netmeister.org/news/learn2quote.html
Sometime new messages are actual replies with changed subject. That is
quite convenient for people writing messages, but results in new topics
attached to existing threads, which is inconvenient for readers.
[ n+1 fullquotes removed]
Patrick
ps: Eric, can you change the mailing list footer so that it has a line
with "-- "? This would at least get us rid of this footer cascades.
Engineers motto: cheap, good, fast: choose any two
Patrick S.
Student of Telematik, Techn. University Graz, Austria