Difference between revisions of "HV HowTo for Users"

From Hall A Wiki
Jump to: navigation, search
(TroubleShooting for Users)
 
(29 intermediate revisions by 2 users not shown)
Line 1: Line 1:
 +
<big>High Voltage Instructions</big> -- the basics
 +
 
The High Voltage is controlled by LeCroy 1458 crates and associated modules like the 1461N.  A trailing "N" means negative, and a "P" means positive HV.
 
The High Voltage is controlled by LeCroy 1458 crates and associated modules like the 1461N.  A trailing "N" means negative, and a "P" means positive HV.
  
 
<font color="red">Simplified Instructions :  To control the HV in Hall A, do the following</font>
 
<font color="red">Simplified Instructions :  To control the HV in Hall A, do the following</font>
  
Login as "adev" on the "adaql1" or "adaql2" computers.
+
Login as "aslow" on the "adaqsc" computers.
  
cd ~adev/slowc
+
cd ~aslow/slowc
 
+
To run both HRS:
+
./hvs HRS
+
  
 
To run the Left HRS alone:
 
To run the Left HRS alone:
Line 19: Line 18:
 
./hvs BEAMLINE
 
./hvs BEAMLINE
  
======================================================
+
== Links to more HV help ==
  
I haven't gotten around to adaping this yet, but here are some old but useful instructions
+
These two links provide some old but useful information.
  
 
http://hallaweb.jlab.org/equipment/daq/HVhelp.html
 
http://hallaweb.jlab.org/equipment/daq/HVhelp.html
Line 27: Line 26:
 
http://hallaweb.jlab.org/equipment/daq/gen_slow_control.html
 
http://hallaweb.jlab.org/equipment/daq/gen_slow_control.html
  
=====================================================
+
For experts
 +
 
 +
https://hallaweb.jlab.org/wiki/index.php/HV_HowTo_for_Experts
 +
 
 +
== TroubleShooting for Users ==
 +
 
 +
After installing the rpi boards as servers, it seems that less goes wrong.  If you have a setup that doesn't use rpi, you should upgrade.
 +
 
 +
1. Restarting the server sometimes is necssary.  See the expert page https://hallaweb.jlab.org/wiki/index.php/HV_HowTo_for_Experts.  Simplest way to restart the server is to power-cycle the rpi and wait 5 minutes.  The server gets restarted by "cron".
  
Troubleshooting
+
2. Is the HV crate on ?  There are two power supplies which must be turned on.  Turning "on" in software will only turn on the HV if the hardware switches are in the right state and the electrical cables are plugged in.  Sorry, I know it's obvious, but I've seen this mistake ...
  
A number of things can go wrong. If these solutions don't work for you, call an expert.
+
3. If multiple cards seem "bad", I've found that power-cycling the HV crate helps.  And what I mean by power-cycling is you go to the crate and manually turn it off (cutting power), then on.  NOT just via the GUI but actually the hardware.  And there is both a mainframe to power-cycle and the 24-Volt supply.  Plus the rpi has a source of power.  Altogether 3 things that can be power-cycled (mainframe, 24V, rpi).
  
1. HV crate is on ? There are two power supplies which must be turned on.
+
4. Need the sign ! Most of our cards are negative HV, and when you enter the target HV, you must also enter a minus sign. Like "-852" and not just "852".  Beware !
  
2. The GUI connects to a Perl server on a PCThe PC/server then talks to the HV crate via a serial connectionIf the server isn't working for some reason, rebooting the PC may help.   The name of the PC is discovered in the tab corresponding to that crate; e.g. intelha3 on Left HRS.
+
5. Make sure that HV cards are all well-seated in the sockets, and no stress on the cables that can unseat the cardsAlso, it has been observed before that a "bad" card can cause inexplicable problemsFinding and and replacing the bad card solves this. One way to find a bad card is to observe what the lecroy servers print when run by hand. If one card is persistently missing, it's probably bad. But this might be too "expert" an instruction.

Latest revision as of 16:39, 19 August 2022

High Voltage Instructions -- the basics

The High Voltage is controlled by LeCroy 1458 crates and associated modules like the 1461N. A trailing "N" means negative, and a "P" means positive HV.

Simplified Instructions : To control the HV in Hall A, do the following

Login as "aslow" on the "adaqsc" computers.

cd ~aslow/slowc

To run the Left HRS alone: ./hvs LEFT

To run the Right HRS alone: ./hvs RIGHT

To run the beamline crate: ./hvs BEAMLINE

Links to more HV help

These two links provide some old but useful information.

http://hallaweb.jlab.org/equipment/daq/HVhelp.html

http://hallaweb.jlab.org/equipment/daq/gen_slow_control.html

For experts

https://hallaweb.jlab.org/wiki/index.php/HV_HowTo_for_Experts

TroubleShooting for Users

After installing the rpi boards as servers, it seems that less goes wrong. If you have a setup that doesn't use rpi, you should upgrade.

1. Restarting the server sometimes is necssary. See the expert page https://hallaweb.jlab.org/wiki/index.php/HV_HowTo_for_Experts. Simplest way to restart the server is to power-cycle the rpi and wait 5 minutes. The server gets restarted by "cron".

2. Is the HV crate on ? There are two power supplies which must be turned on. Turning "on" in software will only turn on the HV if the hardware switches are in the right state and the electrical cables are plugged in. Sorry, I know it's obvious, but I've seen this mistake ...

3. If multiple cards seem "bad", I've found that power-cycling the HV crate helps. And what I mean by power-cycling is you go to the crate and manually turn it off (cutting power), then on. NOT just via the GUI but actually the hardware. And there is both a mainframe to power-cycle and the 24-Volt supply. Plus the rpi has a source of power. Altogether 3 things that can be power-cycled (mainframe, 24V, rpi).

4. Need the sign ! Most of our cards are negative HV, and when you enter the target HV, you must also enter a minus sign. Like "-852" and not just "852". Beware !

5. Make sure that HV cards are all well-seated in the sockets, and no stress on the cables that can unseat the cards. Also, it has been observed before that a "bad" card can cause inexplicable problems. Finding and and replacing the bad card solves this. One way to find a bad card is to observe what the lecroy servers print when run by hand. If one card is persistently missing, it's probably bad. But this might be too "expert" an instruction.