[jase (Jason)]

You don’t need MaximDL to use FocusMax! It can be used independently or integrated into CCDSoft (just as it can be with MaximDL).

Wow, that’s keen. In my opinion, a microstep size of 1 is way too small. It would take the focuser way too long to move. Ultimately, such as size does not improve the focusing accuracy and would significantly slow down any automated focusing routines performed by FocusMax or @Focus. Typically a microstep size of 3 or 4 will give you the .0005” (0.0127mm) resolution per step (this can slightly vary per stepper motor hence the 3 or 4 value). The FSQ in its native focal length (530mm) and ratio (F/5) has a CFZ of 55 microns or .0021” (0.0533mm) so the focuser must hit focus within 4 steps assuming the 0.0127mm equates to 12.70 micron (12.7 x 4 = 50.3 micron) – this is not hard to achieve for the stepper or software assuming you’ve built a good v-curve model in FocusMax (not sure of the equivalent in @Focus). As a guide the size should be 15%-20% of the CFZ, so a step size between 8 to 11 micron is acceptable. Don’t get confused by microsteps and steps. A microstep is one pulse of the stepper. What you are defining how many pulses make a single step.

With a microstep size of 4, FocusMax repeatedly hits the FSQ CFZ within ~70 seconds (acquirestar, calculate position, step through HFD calculations, then move to the best focus position into the CFZ) – too easy!

Backlash can be an issue, but this is focuser specific. I haven’t experienced such issues with the new Tak 1:10 MEF (micro edge focuser) used on the FSQ-ED. I have told FocusMax that the final movement to the CFZ must be on the “in” side of focus, thus the weight of the camera is pulling on down on the focuser. Perhaps the FSQ-N is slightly different.

Temperature compensation is considered passive focusing. Once you’ve calibrated the temp slope (temp x = focus position y) you must determine a dead zone. The dead zone defines what is an acceptable temperature change before the focusing position is changed – based on the slope value. You don’t want the focuser to be changing constantly if the temp changes by a small factor such as .01 of a degree. Hence the dead zone is important. You can set a correction rate to determine how often the temperature is checked – 5 minutes is normally sufficient. It is possible to change focus during a sub exposure without ill effect. If your temp slope is not accurate you may experience soft focus if you slightly miss the CFZ. The biggest issue I experienced with this method was vibration, so I prefer to configure the corrections to be made between sub exposures.

Having stated all this, I don’t use temp compensation at the moment – may get back to it soon. I’m presently using active focusing, that is, automatically refocusing every x minutes. So for example, if I’m collecting 15min subs for luminance data and the refocus interval is 30 minutes, the set up will take two subs before the auto focus routine kicks in and establishes focus again, then automatically continues with data acquisition (as the autofocus routine doesn’t take long, I don’t lose much time and can be assured I’ve hit the CFZ again). I’ve also saved time with filter offsets. My LRGB filters are parfocal, but the 10nm Ha filter isn’t. So when the Ha filter is selected, the focuser automatically shifts a defined number of steps to reach precise focus using this filter. No need to perform the initial focus – though the filter would be refocused anyway by the refocus interval (30 mins).

EDIT: Don't believe that temperature compensation is the ultimate in focusing. There are numerous other factors (not only temperature) that can cause focus drift. Something as simple as seeing conditions will affect focus (especially if your arcsec/pixel is small). Hence, I feel the most accurate method is periodic refocusing between subs (active focusing).

Hope this helps.
Cheers