Some thoughts on Temperature and Automation

After the issues I had the other night with focus running away I got to thinking about focus in general. Based on that, I have a few thoughts and suggestions. These are not intended to be a solution to the issues of a few nights ago - those have a known cause that the authors are already addressing. Rather, I think that the ideas might be helpful for improving automation, especially cold start-up, in the future.

Note that these would only apply to those systems that have both automated electric focusers and temperature sensors as well as reasonably predictable thermal behavior. Almost all refractors with electric focus and many reflectors as well would fit this description.

There are two issues this would address:

  1. Focus that somehow manages to “get lost” during a mid sequence autofocus
  2. Most important, cold startup situations where SGP is set up well ahead of time to start a sequence at time “X”, often hours in the future, usually at astronomical twilight.

First I should note that two critical pieces of data are already present in SGP logs. These are position at each focus and temp at each focus (remember this whole idea only applies if one has that sensor). This would make it very easy for SGP, after a few normal sessions, to automatically compile a thermal profile and attach it to that equipment profile. In fact, the graph below was made in a few minutes by doing a search for the data within one of my logs and entering it into a spreadsheet…

This data could be used during a cold timed startup to give an initial temperature based position for the first focus. Just using the single position in the filter setup just plain does not work unless the temp happens in to be just the right place - an unlikely event. In my experience focus is almost never near enough to that position when one starts up.

It could also be used in a focus recovery scheme since it would know where the focus “ought to be” at a given temperature. In fact, if this turns out to be reliable enough (position wise) for a given system, one might optionally let the user specify this point as a start point for each and every autofocus.

Finally, many folks use a cool down period where they open their roof or dome for an hour or two. This “pre-event” might be given some thought as well.