***********************
Odd info about PEP feedbacks:

If you inject with the BIC, the feedback loops don't get
their bucket numbers updated, so there is a chance that they
are looking at a bucket with no beam.  (When a DIAGLIST is
executed, it updates a value in the BPMD that only the
feedbacks care about.)

You can check which bucket number the feedback BPM buffer
ring is looking at by typing dbdump, then at the dbDump>
prompt, typing:
dump bpmd.vx00.24.prms (for LER single bunch*), or
dump bpmd.vx00.14.prms (for HER single bunch*).  The third
number in the PRMS array is the bucket number that that BPMD
is looking at. 

[* the HER/LER orbit loops look at the single bunch
definitions, regardless of fill pattern]

example:

MCC::EOIC_DESK> dbdump
dbDump>dump bpmd.vx00.24.prms

BPMD:VX00,24
    PRMS    10.00000       0.0000000E+00     123.00000    
              0.0000000E+00   0.0000000E+00   0.0000000E+00
dbDump>

This tells you the LER single bunch BPMD is sampling bucket
123 for the feedback buffer rings.

If there is no beam in the PRMS bucket, you can do one of two
things to get it pointed to a good bucket number:
(1) Execute a diaglist with a good bucket number in it (no
need to actually inject beam, just execute the list.  This
will update PRMS and create new BPM buffers.  You may need
to wake the BIC up afterwards), or
(2) editdbs bpmd.dbs  (in the appropriate bpmd, force the
prms value to be a good bucket)  The mini-edit file will
look something like:

< :BPMD: VX00, 24;   ! PEP2 BPM LER
  :PRMS: = 10.0, 0.0, 20.0, 0.0, 0.0, 0.0; ! ljh 20Jul98 need decreased avging;
>

then you need to Create BPM Bufferring and cold start the
feedbacks.

If the feedback MEAS still looks bad (but the numbers are no
longer all zero) check the single bunch calibration (look
for SAT and SUMLOW BPMS) and recalibrate if necessary.