Mobile unguided, feasibility discussion

Started 5 days ago | Discussions thread
Trollmannx Senior Member • Posts: 3,766
Re: Mobile unguided, feasibility discussion

RudyPohl wrote:

1llusive wrote:

RudyPohl wrote:

elgol20 wrote:

elgol20 wrote:


maybe a new thread is good which deals with the question of how far we can go being mobile and use no guiding (for different reasons).

It would be nice to keep this experienced based without gearfights. I have used all kinds of equipment but like to be fast and light and still do deep sky astrophotography. Rudy and others gave nice examples of how precious time under clear skies is to them and we all have different circumstances, reasons and needs to get there. so how good does ist get, more with less?

with astrotracks and lighttracks (II, which I have) there are effects like

PE (periodical error)



stellar aberration,

differential flexure,

polar alignment method (accuracy),



which might play a role. the measured result can be a distorted star shape. softwares are able to "measure" FWHM, maximum and minimun. means if max and min are the same the star is round but may be out of focus ... the aim is the get as small and equal FHWM as possible with dependency on focal lenghts, wheight (aperture, sensorsize etc)

The goal is of course to gather as much light as possible so you have limited chances to do this. Increase aperture, exposure time, sensor capacity (area and QE). Then to be able to make DS pictures you easily must go beyond 150mm foal length. some say 400mm is still wide field. with 300 or 400 mm there a lots of objects possible and pixelpitches lead to resolutions of 2 to 3 arcsec per pixel.

For me I want to be able so be as precise as 2 pixels of my D810A using 400mm, talking 5 arcsec. Using a New Atlux in Namibia using 400mm and 10 min exposure were easily done guided even though there were still lost frames due to whatever. I did 2 to 6 hours total exposure per object time with the dslr. and still I see noise...

so for me the question is, how good can it be and this I want to find out with tests. or just do 30 sec or 1 min exposures forever?!

the effects namend above can be discussed, like Roger has already done, me too a bit.

here are some examples of how things can be mounted.

due to bad weather I had 2 chances so far to test things. with a nikkor 300 f/4 PF pics were super sharp doing 2 minutes with no wind (cirrus). so 300mm 2 min is easy but the corners are no good so another lens (used gimbal)!

next time I tried my 400 f/2.8 which wheighs 4.6 kg, with counterwheights, see above and another thread. Did this before a storm so no real test, but 6 min had "OK" results not quite so sharp though, round stars so I guess due to seeing and wind.

guessed or known quantity of the effects:

PE: Lighttrack 2 arcsec, Astrotrack 5 arcsec

wind: unpredictable, but could be that using wheights is less prone to blur

refraction: depends on height obove horizon and exposure time, increases and limits accuracy, like 1 to 2 arcsec per minute possible not too close to horizon. astronomical refraction is 0 in zenith and about 60 arcsec at 45 degress above horizon, 5.3 arcsec at 10 degrees and can be approximated to 20 degrees above horizon with R=tan (angle obove horizon). for example when doing exposure you start at 45 degrees and end at 44 you get 2 arcsecs of error. hope that is correct ;), for tan(45) is 1 so 1 arcmin. tan(45)-tan(44)=0.034 arcmin=2.0... arcsec

stellar aberration, can be ignored, small enough

differential flexure, well... could be a big problem and depends on many things. I saw it when using the lighttrack when moving the scope, it was quite strong with the gimbal head, less with the counter wheights. that means i saw a change in position of the polar star which can only be a part of the total diff. flexure of the setup. limits of wedges used, bar for wheights...lots of contributions...

polar alignment method (accuracy), how good is the polar aligment, depends... for me I calculated for the rotational calibration of the polar scope plus viewfinder scale: it can reach to like 10 arcmin accuracy which would result in like 1 arcsec per minute for my setup. better I think is impossible or pure luck. alternative buy a polarmaster...

seeing: widens the stars and can be checked, Pickering steps go up to more than 5 arcsec to no test when seeing no diffration rings in the telescope at all.

etc., ...

-- hide signature --

sorry for the typos

Hello, snow and mud, so, time to check and process:

Refraction drift: from the data I derived an refraction drift of 3.6 arcsec per minute in the end. From formula 3.2. Conclusion for me is really to think before tracking without guiding: which refraction drift do I get with my setup ... Have to admit before I did not think about it out there for either I guide or I do Polari with short exposure time, shorter focal length.

One possibility is using your smartphone with software like Stellarium and look at the beginning and ending height and calculate the difference using the refraction formula (not valid with angle less than 20 degrees above horizon): tan(startangle)-tan(endangle)*60, result in arcsec, check with tan(45)=1 arcmin

-- hide signature --


Hi Stefan,

In order to help those of us who hate math and were never any good at it, ( and that certainly includes me from my earliest days in grade school right through university ), could you explain your conclusions in simple layman's terms and the reasons for those consclusions?

For example, can you say something like.."There is no benefit in trying to do unguided tracking for more than x-minutes because refraction drift (whatever that is??) will cause the stars to look like little elongated eggs." In this way, I can understand your conclusion without having to understand the math. Or perhaps you can say something similar to this so that I and others who cannot follow the math can still understand the conclusions, the reasons for them, and the consequences.

By the way, I am not saying don't include the math in your posts, please do so for others to benefit from, I'm just saying that perhaps you could also include a simple summary statement at the end of your post explaining your findings for math-morons like me.

Thanks so much,

I'm no math whiz, but 3.6 arc-seconds per minute (in what part of the sky - it's very specific?) would point to keeping exposures to 30 seconds I think.

Thanks but I need a little more info than that...

30-seconds at which focal length?

Your intention is to use a 300 mm lens. Right?

Also, what happens when 30-seconds is exceeded? Is the result star elongation and if so how much in percentages of the original round star?

For example, in a complex math calculation given here last the poster was able to tell us, while doing the math, that a 105-second exposure on a Star Adventurer tracker with a periodic error of 5 arc-sec/minute (average for the SA), using a camera with 4.22 microns pixels (Nikon 24 MP sensor), would result in a 20% elongation for stars 15-pixels in diameter, which would make them 1.2 x 15 = 18 pixel longer than wider.

The effect of refraction increase as astronomical objects get closer to the horizon.

If taking images starting 10 degrees above the horizon and ending 5 degrees above the horizon (and that is pretty low) the difference in refraction is about 5' (arc minutes) or 300" (arc seconds). And this is what matter - the difference in refraction during the exposure. Not the value of refraction itself.

At latitude 45 degrees north or south a 5 degree difference in latitude corresponds to something like a 30 minute exposure (objects setting at a 45 degree angle relative to the horizon) - during this exposure refraction will increase 300". Something like 3" per minute.

Use 1 min sub exposures with the 300 mm lens and the effect of refraction is far less than the width of a pixel per minute.

Modern cameras have pixel sizes around 4 micron or so. My Canon 7DII have 4.1 micron pixels, and coupled to a 300 mm lens this corresponds to 2.8 arc seconds per pixel.

Numbers are ok, they tell the entire story. If we get them right, or interpret them right that is. Hope my numbers do not stray to far from reality...

Refraction decrease higher in the sky and the practical problem simply - vanish...

This way of presenting complex data is very helpful to the mathematical-challenged among us. I realize this means more work and more time for the folks who do understand the math, but when the info is also presented with simplicity and in a meaningful context it makes this valuable information available to many more.

The practical answer:

Would settle for a quite ok polar alignment and let loose. More likely periodic error, flex and vibrations are all more more severe issues. Do not bother with what does not matter.

Anyway, atmospheric refraction will cause stars to be slightly elongated anyway when close to the horizon. No eggs, tiny elongated stars only.

And atmospheric extinction will limit how faint you can go when close to the horizon. After all we can usually look directly into the setting sun, quite a few magnitudes of extinction.

Here's the original post:


Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow