Lakeside Auto focuser

Hi

Any one noticed the following issue,

Using the Lakeside and Bahtinov Mask with Bahtinov grabber I was able to get the focus spot on the LCD on the LSide showed a value of 2150 , ready for a first run with the auto focus routine . it produced a nice V curve a significantly determined the vale of the focus point to be 2150

So 1 star align etc and on to the first image. Later in the session I ran AF , it adjusted the focuser to a slightly .different value… on to collect more subs. on inspection when zoomed on the smaller stars seemed to be doughnut in shape ie slight darkness in the center , definitely out of focus ,

In a nut shell , found slight improvement by changing the binning from 2x2 to 1x1 , but still getting differing values when testing AF over a short period of time. The seeing was very good. I’m using the SB8300M and Skywatcher Esprit 80mm.

Checking the focus with the Bahtinov Mask will show the AF routine typically didn’t get it right

Can the numbers be tweaked to to improve accuracy and reliability .

Clear skies

Trevor

I also use the Lakeside motor focuser and don’t have the issues you have encountered. Some suggestions I can think of:

Double check that the stepper motor is securely attached to the focuser shaft and that there is zero slippage.

Configure backlash in the SGP focuser parameters - I have mine set to 100 IN which was another user’s recommendation.

Calculate the Critical Focus Zone (CFZ) for your setup. There are formulae online to do this (I’m too lazy to google for you). On my 110mm f6 refractor the CFZ is just 70 microns for Blue light and 100 microns for Red.

Work out how many microns each step of the Lakeside moves your focuser - e.g. move the Lakeside 5000 steps and measure the change in position of the focuser. On my scope each step is 4.5 microns.

So on my setup the CFZ is just 15 steps wide in Blue and 22 steps wide in Red. In the autofocus setup, you have to select the number of steps between data points such that there is a good chance at least one will land in or close to the CFZ. But if the step size is too small then you risk the autofocus routine taking too long to complete if the initial point is too far away from the best focus. I have done some experimenting on my setup dropping the step size down to 15 and upping the number of data points to 15 but have settled on a step size of 25 and 9 data points.

Are you cropping the image before measuring the HFD? If your frame is not perfectly flat then the stars at the edges will be distorted . I have set mine to crop 20%.

Even so, no two autofocus runs are the same for me. Sometimes I get a nice V, sometimes its more of a U and sometimes a flattened W. I haven’t had to discard any frames for being out of focus yet so have stopped worrying about it. At the start of each new session the focuser is usually at its last position from the previous session and that is close enough to launch AF and find the new focus. If the focuser has been moved in the meantime then I’ll use SharpCap to get approximate focus before launching SGP.

cheers
John

John

Thanks for the info , I have the formula for the correct step size , it will be the first exercise next session, The scope has a good flat field and all it tight. Backlash needs setting , I thought I had saved a value in the equipment profile. Will update when its configured.

Trev