Chomping at the bit for a new aquisition

[RobertM]

Fay,  The Sky90@f/4.5 has a f/l of approx 405mm, the ED80 that I guide with is 600mm so as you can see guiding is less critical.

Just looking at your M33 from Les G. and you have pin sharp stars there.  Maybe the problem is focusing or seeing - I've noticed that's difficult to achieve and keep with the supplied focuser.

[MarkS]

Robert,

Do you set your focus once for a whole imaging session?

I am becoming more and more aware that focus (on my C11 SCT) drifts and I need to refocus.  It seems to be consistent with falling temperature though I haven't quantified yet - this will be the next stage.

I reckon I need to refocus every 30 minutes or so under present (Wintry) conditions.  I also give the scope 90 minutes to adjust to ambient temperature even before I being imaging.

Mark

[Mike]

Focus on all scopes will meander through the night, especially if the temperatures changes a lot. Even more so if you bring your scopes from a warm room to outdoors where they can take hours to get down to ambient temperature. better to sstore them in a cold garage or shed.

Do you suffer from the same fate in summer also Mark? Is mirror shift a cause?

[RobertM]

Mark, I actually set it then only refocus if on change of target or filter.  My imaging setup is always at ambient temperature when I start (it's out all the time under wraps) and there isn't much change at the moment between  +4 and -2 degrees day to night.  I certainly don't notice the difference though MaximDL often identifies one of the images near the beginning of the session as the one with the best FWHM.  If you work out the coefficient of expansion of the aluminium over 5 degrees C you could probably have a stab at guessing how much the position changes.

If I was automating the imaging session then I would refocus after every 30-60 mins.  It's so quick as long as there's a bright enough star, which isn't always the case.

[Mike]

If anyone is thinking of buying a motorised focusser, a lot of the modern designs have automatic temperature compensation built into them.

[RobertM]

This is the one I use: http://www.lazyfocus.com/#easyfocus-stepper

[Fay]

So, Robert, you have to buy a Robofocus motor & Easy Focus Stepper interface? Is that everything? Do you just connect the interface to the motor? Do you then have to make something to replace one of thescope focusing wheels?
I am confused as to how all this works & goes together. 

[Fay]

Ok, I have read the manual, now.

[RobertM]

The Stepper motor is connected to the EasyFocus via a ribbon cable which is the same as Robofocus.  The EasyFocus unit needs 12v power and the output is rs232 which is plugged into my 4-way Serial to USB unit so it just looks like another serial device.

There are various ways of connecting the stepper motor shaft to the focuser shaft but I use a aluminium shaft coupling so it's a direct drive.  The motor is held in place by a bracket made out of some spare Dural plate.  The couplings can be obtained in almost any size and material and vary in price from £5 to £150, the one I got was just over a tenner.  Will take a picture tonight so you can see how it all holds together.

posted anyway as I'd written it ...

[MarkS]

I don't think it is mirror shift.. The mirror is very stable - I can even do a meridian flip with no need to refocus.   However, I do take the scope outside from a warm house - maybe that is part of the problem - maybe it takes longer to reach ambient temperature than I am allowing for.  But I don't want to store it in a damp garage.

So just how big an effect is thermal expansion/contraction?

Warning: This next bit contains scary maths

My scope is steel which has a coefficient of expansion of 13x10e-6/deg C
It is 550mm long, therefore for every 1 deg C it expands by 7microns (13*550/1000)
Therefore, naively, the image plane also shifts by 7 microns
But, you might say, the light path is folded, therefore the image plane might shift by 2 or 3 times that.

So I did some rough optics calculations using approximate focal lengths of the mirrors.  It turns out that the image plane shifts by about 5x the expansion in the tube length i.e. it is behaving as if it were a simple telescope 2750mm long (which is suspiciously similar to the focal length of the C11).  I suppose that makes sense really.

So if the glass elements do not expand (i.e. they retain their original focal length) then the image plane shifts by 5x7 = 35microns.  At F10 this means that the image of a star widens by 35/10 = 3.5microns i.e. approximately one pixel for every deg C.

But the glass elements do expand.  If they had the same coeff. of expansion as steel then the whole apparatus would expand uniformly and there would be no shift in the image plane.  But glass expands at probably half the rate of steel so it compensates for approx half the shift caused by the tube expansion.  So we can divide the 3.5microns in half giving 1.85 microns - let's call it 2 microns.

In sum, for my C11, the image of a star bloats (defocuses)  by half a pixel in diameter for every deg C change in temperature.  So a temperature drop of 5C (which is quite typical) can lead to stars bloating by 2.5 pixels. This is pretty consistent with the defocusing I see over the course of an evening.

Mark

[Fay]

Mark, that is exactly the answer I came up with after doing my mathematical calculations   

[RobertM]

I didn't bother to work it all out like Fay and yourself but those figures sound very plausible.

[Mike]

Well that pretty much explains it then 

Is your attic badly insulated? if so, keep it up there or do you have a larder? My tiny larder has a big air brick at the back meaning it is bloody freezing in there all the time and must be pretty close to the outside temperature. Would be agreat place to store my scopes if it wasn't full of a washing machine, microwave and all kinds of other crap.

[Ian]

priorities Mike, priorities...

[Fay]

Mine are all up the shed, nice & dry & cold.