Test error checking on rotating sample changer

[DominicOram]

As of #4287 there is now some code to test against a real rotating sample changer the two main things that need to be tested are:

Sample dropped

Code written as part of #4287. Additionally, to potentially tidy up this work we should test:

• What the status bits give when in the retrieve sample position.
• Confirm that you can do a ma whilst in the retrieve sample position.
• When does the sample drop error clear

Automatically re-initialize on power cycle

The code in #4028 seemed to do the right thing eventually, but needed several attempts to do so. The scientist usually reinitializes the sample changer manually. It may be that the command is unreliable and needs to be sent multiple times. This should be tested thoroughly on all 3 sample changers.

We will also need to test on HRPD, as the electronics are sufficiently different that we may need different logic on that sample changer.

The expert for electronics in Graham Burgess.

---

A similar issue was seen in #5027 . When doing this ticket, we should be careful to consider all of the known error cases and react appropriately to them. We may be able to get a list of the possible errors from Electronics.

The Polaris sample changer Maclennan control crate indicated an error -
position 3 requested, but the sample had failed to lower. However, the
Move_Finished block indicated "1" (i.e. it though the move has been successful)
and started the next run with no sample in the beam.

[DominicOram]

Putting into impeded as I have emailed the instrument scientist to get RAMS going so that we can do the test.

/remind me chase scientist in one week

[reminders]

@DominicOram set a reminder for Aug 5th 2020

[DominicOram]

Instrument scientists highlighted in the demo today that the most important change was that the Move Finished PV should reflect that the move finished successfully. If testing the rest goes badly make sure at least this is merged.

[reminders]

👋 @DominicOram, chase scientist

[DominicOram]

Went down and started this testing on POLARIS today. The logic for retrieving a dropped sample did not work due to the following unforeseen quirks of the sample changer:

• Sending any new commands to the rotating sample changer will clear all errors
• Sending commands whilst the sample changer is "doing a move" has no affect
• There are some points where the sample changer is "doing a move" but the motors aren't moving e.g. when the sample arm has been raised but the carousel is yet to rotate, we can't send commands at this time

Unfortunately I also ran out of time to test the automatic re-initialisation on power cycle.

To close this ticket I will do the following (as agreed with @Tom-Willemsen and @JamesKingWork):

• Remove the logic for retrieving dropped samples
• Add additional logic for propagating alarms and highlighting errors
• Deploy these as hotfixes to
  • GEM
  • POLARIS
  • HRPD
• Contact scientists of above instruments to see if there's any alerts etc. we could set up to help them in the coming cycle
• Make a new ticket for further testing