Misc notes
problems

---

Miscellaneous notes...

• maintenance mode
• 05apr16
  
• switch inside the cabinet
• Only read at boot up, so cycle the power after changing the switch.
• remote access (even read) is not supported in maintenance mode.
  
• it will:
• encoders not used
• only velocity mode. slower than normal mode
• az slave motor not used (back driven by the master).
• The amp encoder display is garbage in maintenance mode.
  
• software limits not respected.
• To read hardware limit:
• enter maintenance mode
• drive to limit switch
• switch back to normal mode and read the encoder.
• limit switches
• 05 apr16
• software limits.
• in tracking mode, can't go beyond
• in velocity/rate mode. sets velocity to 0. 
• el: min 5.8, max: 88
• az; min -180, max:360..  (from acceptance test..)
  
• hardware limits
• 1st switch is hardware limit.
• Should normally not hit it (software limit protects .. unless 0 velocity creep)
• can only drive out in velocity mode.
• 2nd switch. backup limit. should never hit
• manual (operation II says you need special attention to get out.
• we manually flipped the switch. ended up rebooting system to get back.
• turn count system.
  
• See page 18 of op&maint manual#2 page 18
• The az position is az = encoder_reading + fixedOffset  + 360*turncount
  
• we have an absolute encoder (the azimuth,elevation fixed offsets offsets are not used #20.21).
• The soft limits we are using : -180 to +360 #20.22,23 (page 152 of opManual #1)
• So the turn count should -1 (-180 to 0) and  0 (for 0 to 360).
• Looks like there are 2 turn counts: 
  
• the relay 11K3 located at az=90.
  
• It is 0 when less then 90, 1 when greater than 90
• Software turn count that is driven by the value of the encoder
• This changes when the encoder crosses 0  deg north.
• this value is stored in #20.09 and moved to nvram on reboot. this location is read only, you can't change.
• What we found out testing the relay
• you can manually put the relay in the wrong state by pushing a button on the relay. This does not change the value in #20.09. So 20.09 must be written by the software when it sees the encoder value cross 0 deg.
• az Software limit errors.
• We've set the azimuth software limits to -180 to 360.
• Changing parameters via the keypad
• we've go the security turned off  #00.34 =0
• after changing a parameter you need to save it
• go to xx.00  enter 1000  and push the red reset button, then reboot
• -->This has to be done with the drive powered (contactor closed). Putting 1001 in xx.00 doesn't work

---

Problems

• 29jan21: Software high limit error, turn count error
• we had driven the az through the  360 soft high limit in maintenance mode.
  
• It stopped at the hardware limit (which was > than 360deg)
• we then went back to normal mode and  we couldn't move the telescope. softHigh limit, turncount error
• the encoder must have read > 360
• the software run count was 1. 
• going back to maint mode we ran the encoder ccw, then back to normal . now had about 700deg.
• This was because the software turncount of 1 was adding 360deg  to azimuth. Still no positioning in normal mode.
• We spent lots of time messing around trying to get the software value run count to be 0  or -1 with no luck
• We finally got it to work by driving in maint mode into the cw limit, switching to normal, back to maint, then drive out a little and switch back to normal. when we then drove back through 360 deg, the software turn count went back to 0.
• 14mar23:  trouble with turn count error..
  
• had -357 az
  
• tried the above proceduare (with reboots) and didn't work.
• this was probably from a power dip rather than a maintenance mode driving beyond 360.
• --> looking at the logs
• 11:30 the telescope was at -180 and there was a power dip
• when coming back up the azimuth read 0 deg
• i've seen this a before after power dips. i've had to command to to a non zero azimuth to get a reading
• this caused a turn error.
  
• solution.was at 360, drive ccw back around through 0 (in maintenance) and then switch back to normal mode. turn count was then ok.
• --> dump turncnt (-1 az  -180 to 0)  (0  0<az<360)
• /share/megs/phil/svn/aosoft/p12m/Test/turncnt.sc   (run from galfas2)
  
• 01feb 21: barely moves in normal mode, moves normally in maint mode.
• This occurred in normal mode. In maint mode it moved normally.
• The pid loop values had changed.
  
• See page 42 of opManual#2
  

• menu	was	correct
  20.01 controller gainA	0	150
  20.02 controller gainB	232	150
  20.03 el main profiler max speed (user units)	5	15000

• We reloaed the correct values, then 1000->20.00, reset , reboot and it worked ok