Transcript HV Adjustment Instructions: Temporary Fix

MCast Errors and HV Adjustments
Multicast Errors (seen on the DATA ERIS connection) have caused a
disruption of a HV Adjustment due to a timeout (since our replies were
delayed by the Multicast errors beyond the timeout limit). Hopefully
the Mcast errors will soon be resolved by Peter Dennet/Mike, however,
in the mean time…
– The HV Adjustment code is being modified by Bastian so that it will be less likely to fail,
and if it fails it will be easier for the shifter to see the problem and know how to
proceed.
– Corresponding documentation will be written by Karen and checked by whomever
happens to be on shift.
– Temporary instructions are now available (and attached at the end here and will be
printed in the pocc) in case something goes wrong before the code can be updated.
– Shifters should always have a separate terminal open with this command running so
that any errors can be debugged by an expert if necessary (should be open on a
terminal when you arrive, only have to hit enter before you start the adjust-trd-hv
script):
cmds_mon -g -m HOSC
– Shifters should be ESPECIALLY VIGILANT during the HV adjustment due to these
multicast errors.
HV Adjustment Instructions:
Help, it went Wrong!!! (Temporary Fix)
•
If adjust-trd-hv -2 fails, the error message you get will look something like this:
Adjusting U1 HV via JINF-U1-P
Read JINF config file... successful!
Writing JINF-U1-P config file to file /pocchome/trd/RUN/COMMANDING/TRDDAQ/JINF-U1-P_7aa1.bin...
Erase JINF config file... successful!
Write JINF config file... failed (error code: 000c)! What do you want to do? [(a)bort/(r)etry/(c)ontinue]
r
Retrying...
Write JINF config file... failed (error code: 0001)! What do you want to do? [(a)bort/(r)etry/(c)ontinue]
•
When this happens you will need to decide what to do. At the moment, the best solution is to:
(1) check with LEAD: do they have commanding still?
(2) if yes (or if no, but their commanding is repaired) then (r)etry once more.
•
If this does NOT work, then (a)bort and restart by rereading the file that was written to /pocchome/trd.
–First, ls /pocchome/trd/RUN/COMMANDING/TRDDAQ/JINF-U1-P_7aa1.bin
–If it doesn’t exist, and the Erase JINF config file… has already come up, you’ll need to use the procedure on the next page.
–If it doesn’t exist but the file hasn’t yet been erased from the JINF, then either simply restart either the entire procedure (if this was U0)
or restart the procedure for only U1 using: adjust-trd-hv -2 –only-U1
–If the file is indeed there, then restart using the file by typing exactly this:
adjust-trd-hv -2 --only-U0 --read-from-disk or
adjust-trd-hv -2 --only-U1 --read-from-disk
Depending on which side failed. If the failure was on U0, then after you recover using the above
command for U0, you can run U1 normally like this:
adjust-trd-hv -2 –only-U1
•This SHOULD work. If this still fails, then have LEAD see if they can reboot the JINF that has failed. (as per a dsp error). Then
try the line directly above again: adjust-trd-hv -2 --only-U1 --read-from-disk
•
If THAT fails, call the expert ASAP.
HV Adjustment Instructions:
Uh oh, it went WAY wrong!
OK, so it didn’t work the easy way because we lost our configuration file. So what we’re going to need to do is to first set the high voltage to a
single value for ALL channels on each side. And then we’ll have to set the individual channels using a different configuration file. So to prepare
for this, we need to :
1)
Write down the HV values prior to the HV adjustment that just failed. U0 = ___________ U1 = ____________
2)
Figure out what value you were trying to change them to. (ex: the number above -2) Let’s call this your Target Voltage:
Target U0 = ______________ Target U1 = _____________
3)
Now, we need to know how much the individual channel settings impact the AVERAGE of each crate U0 and U1. So we need to examine
the file with the individual channel voltages:
~trd/trdsoft/trd_acc_tas/TRD/adjust_trd_hv_conf_files/hv_channel_adjust_combined.conf
And to find the impact of the individual channels on the averages, we’ll issue this command :
cat filename.conf| grep -e ”^0" | awk –F= ‘{ print $2 }’ | awk 'BEGIN {SUM=0; N=0} {SUM+=$1; N++} END {print SUM/N}’
(where filename.conf is that really long filename above, and you should replace ^0 for ^1 to get U1 )
The current result for U0 is -1.14, while the result for U1 is +1.97. This means that by applying the individual channel voltages, you’ll change the
average on the side U0 and U1 by those amounts. So you need to modify your target voltage by these average voltages in order to set your crate
to the correct number. So for example: say your target for U0 was 1440 But we know that after we set it to 1440 we’re going to apply the
individual channel voltages which will raise the average by 1.97. So the voltage we should actually set is 1440-1.97.
SET TO Value for U0 is target voltage + 1.14 = ________________.
SET TO Value for U1is target voltage – 1.97 = _________________,
4)Now, once we have the voltages that we want to set U0/U1 to, we can proceed as planned: first follow the instructions in the guide book,
starting on pg 132, setting the SET TO voltages for U0 and U1 on line 117 pg 133 as the values you just calculated above. Once you reach pg 135
do NOT tell LEAD to turn on the DAQ.
5)Next, you’ll need to load the configuration file for the individual channels using:
adjust-trd-hv --conf that_long_filename_above.conf --only-U1
6)
Then, finally, it should work and you’re done. If this FAILS, definitely call the expert ASAP.