Sharpcap and Sequence Generator Pro

[admin]

Hi,

Could I ask the question about the backlash compensation that is built into the focusers that you have? Does the focuser Ascom driver hide the backlash compensation completely from SharpCap?

What do I mean by that?

Suppose that your backlash compensation figure is 25. You start with the focuser position 19,000. You press the large step outwards button which is set to +100, so the position is now 19,100. You now press the small step in which button which is set to step size of five. Obviously at this point the focuser should move in the five points you have requested plus the 25 required for backlash compensation. What I want to know is what does the focuser position read as after pressing the small step in button? If it reads as 19,095 then the backlash compensation is completely hidden from SharpCap. If it reads as 19,070 then SharpCap is aware of the backlash compensation movement because the compensation is included in the new readout. If it reads something else then who knows what is going on!


Cheers, Robin

[chongo228]

I'm 99% sure it's hidden and would read 19,095 in your example....I will confirm next time I'm out which won't be for another week or so. That's one of those details I never paid attention to until you asked.

[remyastro]

Hi,

ZWO EAF manual recomments to set focuser backlash to 0 if backlash is set in capture software.

My understanding of this is that the focuser backlash compensation is hidden from the capture software.

Examples of software backlash setting are also given for Sequence Generator Pro and FocusMax in EAF manual.

Am I wrong ?

Cheers, Remy

[remyastro]

HI Robin,

I have performed the test on my focuser ZWO EAF, as you have specified it.

Backlash compensation figure = 25
Start position of the focuser = 16000

Below the readouts from Sharpcap user interface:

16000 -> move out 100 -> 16100 -> move in 5 -> 16095 -> move out 5 -> 16100 -> move in 100 -> 16000

I have carried out this sequence several times with the same results each time.

This confirms that the focuser backlash compensation is hidden from the capture software.

Cheers, Remy

[admin]

Excellent, thank you for testing that.

I am now thinking that I will add an option to the hardware settings in SharpCap where you can choose backlash compensation for your focuser – the options will be something like

* Handled by Ascom driver
* Handled by SharpCap - with an adjustment for the amount of compensation
* None

If you select the second option and set up the backlash compensation in SharpCap then the control will behave in the same way that yours does from the point of view of the user.

If the focuser is backlash compensated in either of two ways then in due course I will make extra functionality available in the focusing routines.

Cheers, Robin

[Obelisk]

One question that does not seem to be addressed is what is the achievable resolution of a single focuser step if the OTA system is subject to a temperature change?
Aluminium has probably the worst coefficient of thermal expansion of common OTA materials. So if we assume, for example, the gap from the objective lens to the camera sensor is replaced with a single tube of aluminium we can calculate the difference in length for a given temperature change. To calculate this we need the length of the tube, the temperature change and the linear expansion coefficient. For my refractor the nominal distance between the objective lens and the camera sensor (in focus with field flattener) is about 360mm, temperature change, say 1°C and the coefficient for aluminium alloys, typically 23.5x10-6/°C. This comes out as 0.00846mm/°C (i.e. for every 1 °C temperature change. So if the temperature change is -10°C then the overall length will reduce by 0.0846mm). With brass and other materials factored in this number will be smaller – you could calculate this.
For the next part, the OTA was mounted vertically with the camera at the bottom. The focuser was driven out (downwards) by 11000 steps. It was then driven in by 1000 steps to ensure there was no backlash in the system. The distance between the static and moving components was measured (in this case the end of the optical tube and the shoulder of the field flattener). The focuser was driven in 10000 steps and a second measurement taken. The two measurements were 68.12mm and 27.42mm. So for 10000 steps the focuser tube moved 40.7mm which is to say that 1 step of the focuser will move the tube 0.00407mm.
If we combine the two we have a focuser resolution of about 2°C/step. So unless the metal temperature changes by at least 2°C there is no point in refocusing.
My guess is that if you are concerned about temperature effects on focus you may wish to consider fixing the temperature of the OTA (extend the dew heater back as far as the camera), insert a temperature probe into the metal work providing a feedback loop to a PWM driver. Once thermally stable there would be no expansion/contraction - focus once and forget.
Best, Colin

[admin]

Hi Colin,

don't you mean about 0.5C per step?, since your numbers above indicate a length change of 0.008mm per C and a movement of 0.004 per step, meaning it would take about two steps to compensate for a 1C change.

If you wanted to consider this in a little more detail you could also think about how far out of focus you can afford to be before the effects become noticeable – this would depend on the relative sizes of in focus style images, the circle of confusion at the out of focus position and the pixel size of the camera.

Assuming your telescope is 360mm at f/4 (90mm diameter) then the resolving power at the focal plane is about 2.44 µm (1.22 f λ / D). A common pixel size is 3.75 µm, whereas the size of the circle of confusion from being 8 µm out of focus in an f/4 scope is only 2 µm, so it's unlikely that you would notice a significant effect from being that far out of focus.

Cheers, Robin

[Obelisk]

Robin, many thanks for this.

Yes, I did mean 0.5C.

I had not come across circle of confusion (though given the above, it is no doubt a space I inhabit). I have trawled the internet and now understand what it is however, I am at a loss to understand how you came to determine the values you suggest. I would welcome instruction.

The telescope is 66mm diameter with a pixel size of 2.4µm.

Best, Colin

[admin]

Hi Colin,

think of the cone of light that starts off are the objective lens of your telescope as a circle 90 mm across and finishes at the focal plane as a point. Because your telescope is (I'm assuming) f/4, at every point along the way between the lens and the focal plane, the diameter of the cone at that point will be 1/4 of the distance from the point in question to the focal plane. When you are at the focal plane the diameter of the cone is zero, when you are at the lens, the diameter is the full lens diameter. This explains why I divide the focal error of 8 µm by 4 to get the size of the circle of confusion as being 2 µm. If your telescope was f/8 it would have a narrower light cone and you would divide by eight instead..

Hope that makes sense, Robin