Question about tracking accuracy

[dazzystar]

Hi All,

I understand that generally the more expensive the mount the better the build quality will be and therefore the better its tracking accuracy. Fundamentally I understand what tracking accuracy means but what I don't understand is all this talk about sub-arc second stuff. What does that mean to the lay person trying to do some astrophotography?

My rig is yet to see first light but should do very soon and I know that the mount I have (an EQ3 goto) will need to be upgraded but what I don't know is what I should upgrade to and what benefits (apart from generally a heavier payload capacity) I will get.

Can anyone offer some advice or point me in the right direction to a video / document that can help explain it to me and perhaps others?

Cheers
Daz 

[vlaiv]

What do you want to know exactly?

Tracking performance is best characterized by couple of parameters:

- drift rate.

- periodic P2P together with period.

Two are connected, but periodic error is just in RA while both RA and DEC can have a drift rate.

Drift rate due to polar alignment error is mainly constant and is only in DEC, while drift rate in RA changes and is mostly due to periodic error. Very rough calculation of drift rate in RA can be made by P2P * 2 / worm period. For actual peak and average drift rate - it is best to record periodic error and then analyze it in software like PecPrep

Drift rate is expressed in arc second / second and helps determine max exposure length without significant trailing.  Say you have 0.1"/s drift and you image for one minute or 60 seconds - then you will have 60 * 0.1 = 6 arc seconds of movement between start and end of exposure.

If you working resolution is 3"/px - this means elongation of 2px in direction of drift.

Guiding performance is characterized by total guide RMS and RA and DEC RMS errors. Those are connected and total RMS is telling you how much your image will be additionally blurred over having perfect mount. Difference in RA and DEC values will tell you of any star elongation.

Number is calculated as standard deviation of measured star position with respect to where it should be. Measurement is performed on guide star of course.

When guiding, if you have round stars - that does not tell that you are guiding good - it only tells that RA and DEC RMS figures are about the same.

What tells you that you are guiding good is total RMS error in arc seconds.

As a rule of thumb you want total RMS guide error to be about half of your working resolution, but actual impact on final image resolution (star FWHM) is a bit more complex.

Here is list of values and how they fare:

0.2-0.3" RMS - top tier mounts

0.5" RMS - very good performance

0.6-0.7 RMS - good performance

0.8-1.0 RMS - average performance

>1.0 RMS - poor performance, acceptable only on low end mounts like EQ3 / EQ5 and AzGTI which are used to do low resolution / wide field work.

If I'm using short FL lens and doing wide field image with say 8"/px - then I really don't care if I have 2" RMS error - as it really won't affect image at all (remember that half working resolution rule of the thumb).

[SimonHutch]

In the simplest of terms an arcsecond is an Astro term for a measurement of angle. In the case of a mount think about either how accurately a mount is made and how steady it is. Any wobble caused by design/manufacture/gear tolerances means a higher arcsecond guiding number (less accurate guiding). The lower the number, the better the mount and the bigger your pockets need to be! 😀

https://astronomy.com/magazine/bob-berman/2016/10/just-an-arcsecond

HTH

[dazzystar]

Thanks guys for the reply. A lot of it went over my head to be totally honest but I need to learn more about this stuff. 

[dazzystar]

30 minutes ago, vlaiv said:

What do you want to know exactly?

Tracking performance is best characterized by couple of parameters:

- drift rate.

- periodic P2P together with period.

Two are connected, but periodic error is just in RA while both RA and DEC can have a drift rate.

Drift rate due to polar alignment error is mainly constant and is only in DEC, while drift rate in RA changes and is mostly due to periodic error. Very rough calculation of drift rate in RA can be made by P2P * 2 / worm period. For actual peak and average drift rate - it is best to record periodic error and then analyze it in software like PecPrep

Drift rate is expressed in arc second / second and helps determine max exposure length without significant trailing.  Say you have 0.1"/s drift and you image for one minute or 60 seconds - then you will have 60 * 0.1 = 6 arc seconds of movement between start and end of exposure.

If you working resolution is 3"/px - this means elongation of 2px in direction of drift.

Guiding performance is characterized by total guide RMS and RA and DEC RMS errors. Those are connected and total RMS is telling you how much your image will be additionally blurred over having perfect mount. Difference in RA and DEC values will tell you of any star elongation.

Number is calculated as standard deviation of measured star position with respect to where it should be. Measurement is performed on guide star of course.

When guiding, if you have round stars - that does not tell that you are guiding good - it only tells that RA and DEC RMS figures are about the same.

What tells you that you are guiding good is total RMS error in arc seconds.

As a rule of thumb you want total RMS guide error to be about half of your working resolution, but actual impact on final image resolution (star FWHM) is a bit more complex.

Here is list of values and how they fare:

0.2-0.3" RMS - top tier mounts

0.5" RMS - very good performance

0.6-0.7 RMS - good performance

0.8-1.0 RMS - average performance

>1.0 RMS - poor performance, acceptable only on low end mounts like EQ3 / EQ5 and AzGTI which are used to do low resolution / wide field work.

If I'm using short FL lens and doing wide field image with say 8"/px - then I really don't care if I have 2" RMS error - as it really won't affect image at all (remember that half working resolution rule of the thumb).

Do you know what the performance of my EQ3 goto mount is? I'm planning on upgrading it using the OnStep system shortly too.

[vlaiv]

27 minutes ago, dazzystar said:

Do you know what the performance of my EQ3 goto mount is? I'm planning on upgrading it using the OnStep system shortly too.

Thing is - those mass produced mounts have quite a bit of sample to sample variation so there is no telling what sort of mount you have.

Positive side of things is that you can usually substantially improve mount performance by adjusting / tuning it.

I've seen reports of EQ3 being 2" RMS down to sub 1" RMS. Just recently someone reported having second hand mount that works great. Maybe previous owner did adjusting / tuning of it?

There is also a thread started recently on tuning EQ3 type mount by replacing some nylon washers with proper roller bearings for better performance. Look it up.

[dazzystar]

Yes, I commented on that thread but the OP hasn't posted any replies yet.

[900SL]

1 hour ago, vlaiv said:

What do you want to know exactly?

Tracking performance is best characterized by couple of parameters:

- drift rate.

- periodic P2P together with period.

Two are connected, but periodic error is just in RA while both RA and DEC can have a drift rate.

Drift rate due to polar alignment error is mainly constant and is only in DEC, while drift rate in RA changes and is mostly due to periodic error. Very rough calculation of drift rate in RA can be made by P2P * 2 / worm period. For actual peak and average drift rate - it is best to record periodic error and then analyze it in software like PecPrep

Drift rate is expressed in arc second / second and helps determine max exposure length without significant trailing.  Say you have 0.1"/s drift and you image for one minute or 60 seconds - then you will have 60 * 0.1 = 6 arc seconds of movement between start and end of exposure.

If you working resolution is 3"/px - this means elongation of 2px in direction of drift.

Guiding performance is characterized by total guide RMS and RA and DEC RMS errors. Those are connected and total RMS is telling you how much your image will be additionally blurred over having perfect mount. Difference in RA and DEC values will tell you of any star elongation.

Number is calculated as standard deviation of measured star position with respect to where it should be. Measurement is performed on guide star of course.

When guiding, if you have round stars - that does not tell that you are guiding good - it only tells that RA and DEC RMS figures are about the same.

What tells you that you are guiding good is total RMS error in arc seconds.

As a rule of thumb you want total RMS guide error to be about half of your working resolution, but actual impact on final image resolution (star FWHM) is a bit more complex.

Here is list of values and how they fare:

0.2-0.3" RMS - top tier mounts

0.5" RMS - very good performance

0.6-0.7 RMS - good performance

0.8-1.0 RMS - average performance

>1.0 RMS - poor performance, acceptable only on low end mounts like EQ3 / EQ5 and AzGTI which are used to do low resolution / wide field work.

If I'm using short FL lens and doing wide field image with say 8"/px - then I really don't care if I have 2" RMS error - as it really won't affect image at all (remember that half working resolution rule of the thumb).

You should write a book on this Vlaiv, I'd buy it

[Same old newbie alert]

In a simplistic way of describing it..the target that you want to image will move across the sky at a measured rate( 23 hours 56 mins ) this is known as sidereal rate.. in a ideal world you can have perfect polar alignment and perfect tracking in perfect skies... Unfortunately all of those will have errors.. so you could buy a top end mount  with perfect gears and encoders with less errors than a budget mount which will have more errors ... Most people will have the budget end( nothing wrong with that as they're still expensive enough as it is)  and use a guiding software like phd2 which will track a guidestar and tell the mount to speed up or slow down and correct this errors on both axis, the less it has to correct the better your guiding RMS will be..this is where people get their sub arc sec measurements from

Also it's to do with image scale, if your image scale is say 4 arc secs per pixel you won't need sub arc sec guiding, it's only when your image scale is at 1 arc sec that you need to have the guiding better... But also depends on your sky limitations as you can't image under what your skies allow

[dazzystar]

Thanks @Same old newbie alert for the reply too. How do I find out my image scale? I have a ZWO ASI183MC camera.

[dazzystar]

According to astronomy.tools CCD resolution calculator my resolution is 0.99" / pixel if that helps?

[Same old newbie alert]

3 minutes ago, dazzystar said:

According to astronomy.tools CCD resolution calculator my resolution is 0.99" / pixel if that helps?

Yep that's the one, so your scopes focal length is 500mm

[dazzystar]

47 minutes ago, Same old newbie alert said:

Yep that's the one, so your scopes focal length is 500mm

That's right. It's a Startravel 102. I'm waiting to find a used guide scope and camera and will be using PHD2.

[vlaiv]

2 hours ago, dazzystar said:

According to astronomy.tools CCD resolution calculator my resolution is 0.99" / pixel if that helps?

Two important things to note here:

- use super pixel mode for debayering - that will make recording at actual 2"/px vs 1"/px - as is normal for OSC cameras. Color/OSC (short for "one shot color") cameras have something called bayer matrix of pixels instead of regular pixels. This means that each other pixel records different color - they are arranged in grid of 2x2 pixels and each grid element records - one red, one blue and two green color (each pixel has different filter applied to it).

You probably heard of RGGB or similar being mentioned as bayer order - that just explains how pixels are arranged in 2x2 matrix.

In any case - this means that there is only single red and blue pixel per group of 2x2 pixels and that actual resolution of color camera is less than of equivalent mono camera (where each pixel records the same information - depending on filter used for whole sensor).

Super pixel mode acknowledges this and treats image accordingly.

- you are using achromatic refractor and level of chromatic aberration will be severe in comparison to all other things in the image.

I took following image with ST102 and 3.75um pixel size camera:

Star bloat is evident and since you have smaller pixels - it will be even more severe. You can lessen it by using super pixel mode, but even then, best results will be if you really under sample.

While above image has poor stars - if I presented it like this:

Then it would not be as bad. That is about 4.5"/px - so if you use super pixel mode and then additionally bin x2 before processing - it will be more pleasing and less bloated.

All of this also means that at 4"/px - you don't really have to worry that much about mount performance. If it guides at 2" RMS -  you'll be fine.

@Same old newbie alert

What happened to your old account?

[Same old newbie alert]

5 minutes ago, vlaiv said:

 

@Same old newbie alert

What happened to your old account?

I changed my phone, and couldn't remember my password( years ago) seems like I tried too many times and it locked me out

Grand suggested doing the same thing that I'd had been and was just going around in circles

One of the other mods suggested I spoke to Grant

I can't access my account on my laptop as I'm waiting on a new charger, but that's coming from Singapore...

So I had no other choice

It's still me though

[Same old newbie alert]

2 hours ago, dazzystar said:

That's right. It's a Startravel 102. I'm waiting to find a used guide scope and camera and will be using PHD2.

You can get an adapter for your finder to convert it to a guider if you have a 9x50 finder... Be on the lookout for a cheap CMOS ASI 120.. or a mini are about 150... Mono is better but a colour one should do the trick

[vlaiv]

6 minutes ago, Same old newbie alert said:

I changed my phone, and couldn't remember my password( years ago) seems like I tried too many times and it locked me out

Grand suggested doing the same thing that I'd had been and was just going around in circles

One of the other mods suggested I spoke to Grant

I can't access my account on my laptop as I'm waiting on a new charger, but that's coming from Singapore...

So I had no other choice

It's still me though

I guess admins should be able to merge accounts or something to sort it all out for you

[Same old newbie alert]

Just now, vlaiv said:

I guess admins should be able to merge accounts or something to sort it all out for you

You would think wouldn't you

[dazzystar]

Yes, looking for a ZWO mini guide and an ASI120MM camera. Not sure how long the ST102 will be with me for as everyone has slated it as a scope for AP work so will probably look for something to change it to. Any ideas and a value on the ST102? The ASI183MC will stay for quite some time as I don't have the money to upgrade both!

[vlaiv]

1 minute ago, dazzystar said:

Yes, looking for a ZWO mini guide and an ASI120MM camera. Not sure how long the ST102 will be with me for as everyone has slated it as a scope for AP work so will probably look for something to change it to. Any ideas and a value on the ST102? The ASI183MC will stay for quite some time as I don't have the money to upgrade both!

You can produce some good images with said scope - but it requires some skill and tricks to do it.

I would not dismiss it straight away - you can use it to learn few tricks and get the hang of imaging and then move to something better.

I took this image with my ST102:

And this was with ASI185 - which is camera with much smaller sensor than yours (x3.5 times smaller).

[dazzystar]

Great Pic @vlaiv