Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

More autofocus questions


kirkster501

Recommended Posts

I had my first chance to play with the autofcouser (Lakeside) under the stars (albeit moonlight) last night.  After my issues on Saturday I now have this all set up and running and most impressive it is too.

However, I have a numpty question.  The Lakeside has 65535 potential positions.  However, when calibrating my focuser with the Lakeside calibration routine my full range of my TEC Feathertouch (4 inches of travel) is only 3890 steps.  That means most of the steps are completely "wasted" miles out of focus and that my actual critical focus "zone' is only within about ten steps.  Consequently, when using SGP autofocus routine last night I was finding my focus point was always pretty much identical in terms of the step count (within a couple of steps) even though I know my Blue and HA are not quite parfocal with Lum, Red and Green.

I think I am missing something.  How can I get a greater resolution on my focuser so I use a bigger total number in my focus range to give me more granularity in the critical focus zone please?  Something to do with the step size in the controller I think.  I'll read up but am i on the right lines?

Thanks, Steve

 

Link to comment
Share on other sites

First of all, 3890 steps equates to around 4" of travel.  So each step equates to around 0.001 of an inch.  There is no way that you could possible focus that accurately by hand.  Unfortunately I am not familiar with SGP, but certainly using Focusmax it is possible to set both the step size and the start point for each focus run, and I would have thought that SGP has a similar routine.  I also believe that you need to run a series of "V-curves" so that the focus software learns the particular set-up.  Hope this helps.

Chris

Link to comment
Share on other sites

Hi Steve,

First of all did you get a V curve in SGpro. If so, how many steps did it take to get from the top of your right hand V. You should be aiming for 3 to 4 steps. Any smaller or bigger you should then change the step size you specified in your auto focus set up. Don't worry about the final position SGPro throws up when it has finalised it's focus routine. they will be ever changing as temperature and seeing fluctuates. If the image is in focus the that's all that matters. As you have a temp probe on your auto focuser, ensure temp compensation is activated.

Steve

 

Focuser temp comp.JPG

Link to comment
Share on other sites

Thanks guys,

Chris, point taken about 0.001 inch. Yes, I hadn't thought of it that way.  Indeed, when thinking in those terms it is very accurate even if "only" 3890 steps.

Steve, Yes I got a nice V-curve and it was a "V" as well, not a "U".  I pretty much guessed the step size.  From approx focus at 3114 it went "out" four times my step size of ten.  So it went out to 3154 and was clearly out of focus.  It then came in in steps of ten and then back out of focus "inside" of focus to 3084 before coming back to settle on 3114.  I was getting HFR averaging 0.8.  Does that sound about right?

One more thing.  I notice that I have a backlash three steps before the focuser starts to move in the opposite direction when I play with moving the focuser one step at a time.  How is this accounted for in the algorithm?

Link to comment
Share on other sites

According to Peter at Lakeside backlash setting isn't critical as long as you've got some, I've got mine set to 20.

Setting mine up I run it inwards past focus and before it runs out of travel, set that point to zero, wind out to 1500 steps and set as outwards travel, wind back in to zero, wind out a thousand steps, measure with calipers and divide by 1000 to get step size to enter into software, some software claims to be able to automatically work out the step size but I don't trust them.

I'm using Maxim so no idea about SGP.

Dave

Link to comment
Share on other sites

I am pleased with the focuser so far.  Manual focusing is one of the biggest PITA of AP IMO.  Manual fine for visual but no so for AP.  The fact that SGP can focus on your subject without having it slew away to a bright star is fantastically useful.

A lot of learning I am doing (or trying to do) with a new MESU 200, Autofocuser, SGP and Polemaster.  Taken weeks and weeks on end of cloudy skies to get that precious few hours last night to do some experimenting.

Link to comment
Share on other sites

2 hours ago, kirkster501 said:

Thanks guys,

Chris, point taken about 0.001 inch. Yes, I hadn't thought of it that way.  Indeed, when thinking in those terms it is very accurate even if "only" 3890 steps.

Steve, Yes I got a nice V-curve and it was a "V" as well, not a "U".  I pretty much guessed the step size.  From approx focus at 3114 it went "out" four times my step size of ten.  So it went out to 3154 and was clearly out of focus.  It then came in in steps of ten and then back out of focus "inside" of focus to 3084 before coming back to settle on 3114.  I was getting HFR averaging 0.8.  Does that sound about right?

One more thing.  I notice that I have a backlash three steps before the focuser starts to move in the opposite direction when I play with moving the focuser one step at a time.  How is this accounted for in the algorithm?

Most auto focusers will have backlash of some kind. I set mine to 50 steps with compensation value as IN. This means that if the focuser is moving out it will overshoot the step distance by 50 steps and then move in to it's allotted step position, therefore eliminating any backlash. It's amazing to watch it just takes less than a second :)

Steve

 

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.