MDM Trouble Report for 2007 Dec 11

Telescope: 1.3m
Observer(s): Yee & Eastman (OSU)
Instrument: 4k

Problem(s) Encountered:
During the day, I was running the filter wheel and noticed it 
failed, which gave me something to test tonight, since there's 
no chance we can open the dome.

After extensive testing, it has failed many times in a somewhat 
repeatable manner. When it happens, the gear stalls, it doesn't 
fully engage into the slot to which it was commanded to go, and 
the xmis briefly flashes an error  FWMove left filter wheel out 
of position , followed immediately by  Move failed - Check 
Filter Wheel & Reset . Upon issuing a new filter wheel command 
move command, we get an error at the bottom of xmis  Filter 
Wheel out-of-position RESET , and when we issue a  status  
command to prospero, then the status in Prospero is set to 
 -1 [] 

To fix it, we go to  Setup->Filter Wheel Reset  on the xmis.

These are what we were doing when it failed:
[semi-random sequence of ~150 moves]
?? to 4
9 or 10 to 4
2 to 4
?? to 4
2 to 3
2 to 3
1 to 3
1 to 3 [manual]
1 to 3 [manual]

[fully random sequence of 600 filter changes between positions 1 and 11]
2 to 4
2 to 5
1 to 7
2 to 4
1 to 4
*8 to 4
1 to 4
2 to 7 (error but no reset required)
2 to 7 (error but no reset required)
2 to 5 (error but no reset required)
1 to 7 (error but no reset required)
1 to 4
11 to 4
11 to 4
1 to 3
3 to 8 (error but no reset required)
9 to 1
1 to 5
9 to 2
10 to 3
10 to 3
1 to 3
1 to 6
2 to 6 (error but no reset required)
2 to 7 (error but no reset required)
11 to 4
*9 to 6 (error but no reset required)
1 to 5
11 to 4
1 to 6 (error but no reset required)
10 to 3
3 to 8 (error but no reset required)
10 to 3
11 to 2
11 to 2
11 to 3

[100 random moves between 1 and 12]
3 to 6
*11 to 6
8 to 1
1 to 6 (error but no reset required)
3 to 9 (error but no reset required)
11 to 4
11 to 3
11 to 4

[100 random moves between 1 and 5]
3 to 4
1 to 5
2 to 5
1 to 4
1 to 5
1 to 3
1 to 5

* denotes the filter wheel was moving in the other direction

The outside temperature was between -1.5 and 0.5 degrees C for 
all the tests. Do you think a combination of balance and 
temperature cause these errors?

A couple notes about the software handling these errors:

After it fails, no error is printed in the Prospero terminal, 
and scripts happily continue with the next command. It is also 
impossible to catch these errors with the  error  command 
inside scripts. As the filter status is unknown (and thus won't 
be recorded in the header) and it often vignettes the field, 
this should immediately halt the script with a big warning.

When doing a  status  command in Prospero after the filter 
wheel is lost, it happily reports  System Status check found no 
errors or warnings  and changes the filter status to  -1 [] . 
It should recognize this is an error and tell the user to reset 
the filter wheel.



------------------------------

Submitted on 2007 Dec 12 [3:09:52]