What does good guiding contribute?

[souls33k3r]

Hi All, 

Just trying to understand something here.

I've seen round stars when people are guiding at either 1.2" or 0.2" arcseconds in PHD2. 

What does a good guiding (Total RMS) in PHD2 contributes to the sub and final stacked image? 

Is it worth paying extra towards the mount to get a 0.2" - 0.3" arcseconds difference?

Thanks 

[PhotoGav]

Good question and this is something that I have been dealing with recently with my upgrade to the Mesu.

Round is only part of the story. ‘Bad guiding’ can still create round stars, but big, round stars... and that’s where the problem is... the detail is smeared across too many pixels and clarity is lost. When the tracking / guiding is tighter, the details are not smeared and consequently display with much higher clarity. So, don’t be fooled by round stars, what you really want are tiny round stars and razor sharp details. This comes from an RMS error in the lower end of the scale, down in the 0.2 - 0.3 “/px.

As far as I understand it anyway!

[souls33k3r]

4 minutes ago, PhotoGav said:

Good question and this is something that I have been dealing with recently with my upgrade to the Mesu.

Round is only part of the story. ‘Bad guiding’ can still create round stars, but big, round stars... and that’s where the problem is... the detail is smeared across too many pixels and clarity is lost. When the tracking / guiding is tighter, the details are not smeared and consequently display with much higher clarity. So, don’t be fooled by round stars, what you really want are tiny round stars and razor sharp details. This comes from an RMS error in the lower end of the scale, down in the 0.2 - 0.3 “/px.

As far as I understand it anyway!

Yeah that's exactly what i thought in my little head that it would do but wasn't sure if i was thinking this straight and you answer affirms that.

 Cheers for that Gav. Much appreciated mate.

[vlaiv]

Gav gave a short answer to the point. Here is a bit more elaborate one.

As long as guide error is equally distributed in RA and Dec and follows certain distribution - stars will be round.

However round stars don't mean good guiding. Guide error is one component in sharpness equation - level of resolved detail. Other significant parts are - seeing and aperture. These three define how sharp image on focal plane will be in long integration. Each one can be viewed as "error" to resolution. These errors add like regular random errors do - square root of sum squares of each component (linearly independent vector addition).

This means that if any one error is dominant - results won't be much larger than that particular error. Think right angle triangle size (I actually had that screen shot ready for pasting, don't ask ). If right side of triangle is much smaller than bottom one - hypotenuse and bottom side are almost of the same length (small angle approximation).

This means, for example, in very poor seeing there will not be much difference between excellent guiding and average one. You might have heard of "Guide errors being masked by the seeing", and above is explanation for that. It also means that excellent guiding is less important for very small apertures (80mm and less).

It does how ever imply that if you have steady atmosphere and larger aperture - you want your guiding to be up to task as it might be the limiting factor on resolution. It also means that better guiding will always show in the image, sometimes more and sometimes less pronounced, but it will show, and if is always better to have mount capable of higher precision. After all, in AP, mount comes first, and that is the one of the reasons why.

[souls33k3r]

40 minutes ago, vlaiv said:

Gav gave a short answer to the point. Here is a bit more elaborate one.

As long as guide error is equally distributed in RA and Dec and follows certain distribution - stars will be round.

However round stars don't mean good guiding. Guide error is one component in sharpness equation - level of resolved detail. Other significant parts are - seeing and aperture. These three define how sharp image on focal plane will be in long integration. Each one can be viewed as "error" to resolution. These errors add like regular random errors do - square root of sum squares of each component (linearly independent vector addition).

This means that if any one error is dominant - results won't be much larger than that particular error. Think right angle triangle size (I actually had that screen shot ready for pasting, don't ask ). If right side of triangle is much smaller than bottom one - hypotenuse and bottom side are almost of the same length (small angle approximation).

This means, for example, in very poor seeing there will not be much difference between excellent guiding and average one. You might have heard of "Guide errors being masked by the seeing", and above is explanation for that. It also means that excellent guiding is less important for very small apertures (80mm and less).

It does how ever imply that if you have steady atmosphere and larger aperture - you want your guiding to be up to task as it might be the limiting factor on resolution. It also means that better guiding will always show in the image, sometimes more and sometimes less pronounced, but it will show, and if is always better to have mount capable of higher precision. After all, in AP, mount comes first, and that is the one of the reasons why.

A plethora of knowledge mate and very well explained. 

One question though, we here in the UK are limited to 2 arcseconds seeing, does the guiding precision of the mount still  matter or we're just shooting in the wind and hoping for best results (sharpness)? 

[vlaiv]

I don't think that you are limited to 2" seeing.

Weather might be volatile most of the time, but I would not say that you have 2" most of the time, or in best conditions or whatever.

Seeing is sometime hard to judge. Many people rely on visual estimate. There is a good measure of seeing - 2s exposure FWHM. However, that measure is not precise unless corrected for aperture. In 2s exposure, it is generally accepted that mount won't make significant error. Any smooth running mount should be up for the task. However, mount 60mm guide scope and 8" scope side by side, take 2s exposure on both and you will end up with different FWHM values. This is because resulting star FWHM depends on above three parameters. Even if we take the mount out of the equation by using 2s exposures, there is still aperture that contributes. So when you see that someone is giving you 2" seeing - it might not be 2" - it could be less.

In city, you will probably have worse seeing due to local thermals.

Anyway, this is the website to check out for seeing forecast:

https://www.meteoblue.com/en/weather/forecast/seeing/london_united-kingdom_2643743

(I already selected London, but you can change the city / location)

Last column shows calculated / forecast seeing (one without impact of aperture) - as you see, all quoted figures for Tuesday are below 2". Actually for 3 days - every single value is below 2" - going as low as 1.19" on Wednesday / Thursday night (around midnight).

Best way to see if you need better guiding is to examine your RMS and FWHM you get in your images. Compare that to your imaging resolution. Good resolution is about x1.6 less than your FWHM. This means that if you have 2" FWHM (on average, or in final stack) - good imaging resolution for those conditions would be 1.25"/px. Or to put it more precisely - going with higher recording resolution (lower "/px) will result in oversampling.

Good rule of the thumb for needed RMS is half or less of your target imaging resolution. It will also depend on what your mount is capable of. For entry level mounts (and here I talk about Heq5/EQ6 family and alike, not EQ3 and similar - those are "toy" mounts - well not really, but better suited for wide field work with lens rather than scope), best you can hope for is 0.6-1.0" RMS range. 0.5-0.6 is achievable but only with tuning and modding and in exceptional conditions (no wind, stable atmosphere, great balance and such). Higher end mount will get you progressively closer to that "magical" sub 0.2" RMS.

If you want to play around with some numbers for your setup, here is spreadsheet in attachment, you can use it to calculate expected star FWHM for your aperture, guide rms and seeing. See how it compares to values you have from your subs.

FWHMCalc.ods

[souls33k3r]

8 minutes ago, vlaiv said:

I don't think that you are limited to 2" seeing.

Weather might be volatile most of the time, but I would not say that you have 2" most of the time, or in best conditions or whatever.

Seeing is sometime hard to judge. Many people rely on visual estimate. There is a good measure of seeing - 2s exposure FWHM. However, that measure is not precise unless corrected for aperture. In 2s exposure, it is generally accepted that mount won't make significant error. Any smooth running mount should be up for the task. However, mount 60mm guide scope and 8" scope side by side, take 2s exposure on both and you will end up with different FWHM values. This is because resulting star FWHM depends on above three parameters. Even if we take the mount out of the equation by using 2s exposures, there is still aperture that contributes. So when you see that someone is giving you 2" seeing - it might not be 2" - it could be less.

In city, you will probably have worse seeing due to local thermals.

Anyway, this is the website to check out for seeing forecast:

https://www.meteoblue.com/en/weather/forecast/seeing/london_united-kingdom_2643743

(I already selected London, but you can change the city / location)

Last column shows calculated / forecast seeing (one without impact of aperture) - as you see, all quoted figures for Tuesday are below 2". Actually for 3 days - every single value is below 2" - going as low as 1.19" on Wednesday / Thursday night (around midnight).

Best way to see if you need better guiding is to examine your RMS and FWHM you get in your images. Compare that to your imaging resolution. Good resolution is about x1.6 less than your FWHM. This means that if you have 2" FWHM (on average, or in final stack) - good imaging resolution for those conditions would be 1.25"/px. Or to put it more precisely - going with higher recording resolution (lower "/px) will result in oversampling.

Good rule of the thumb for needed RMS is half or less of your target imaging resolution. It will also depend on what your mount is capable of. For entry level mounts (and here I talk about Heq5/EQ6 family and alike, not EQ3 and similar - those are "toy" mounts - well not really, but better suited for wide field work with lens rather than scope), best you can hope for is 0.6-1.0" RMS range. 0.5-0.6 is achievable but only with tuning and modding and in exceptional conditions (no wind, stable atmosphere, great balance and such). Higher end mount will get you progressively closer to that "magical" sub 0.2" RMS.

If you want to play around with some numbers for your setup, here is spreadsheet in attachment, you can use it to calculate expected star FWHM for your aperture, guide rms and seeing. See how it compares to values you have from your subs.

FWHMCalc.ods 11.22 kB · 0 downloads

Top man. 

Much appreciated mate. 

I must admit, half of the stuff you said just went over my bald head but that's entirely my fault as I don't really understand much about the science behind it but I do follow what you mean but will read when I'm less tired

I'll take a look at the attachment when I'm next near a computer. 

[ollypenrice]

I think of it in fairly simple terms. Firstly you need to know your guide error in arcseconds rather than pixels so you give PHD your guidescope's focal length and your camera's pixel size and then it will give you the guide values in arcsecs. Then you need to know your image scale in arcseconds per pixel. As a rule of thumb your guide error should be no worse than half your image scale if guiding is not to affect final resolution.

What tends to happen is that, if the seeing is bad, it will lower the resolution of both the guide camera and the imaging camera, causing the guiding to deteriorate as well. Since I can't do anything about this I just shoot colour on such nights!

Olly

[souls33k3r]

4 hours ago, ollypenrice said:

I think of it in fairly simple terms. Firstly you need to know your guide error in arcseconds rather than pixels so you give PHD your guidescope's focal length and your camera's pixel size and then it will give you the guide values in arcsecs. Then you need to know your image scale in arcseconds per pixel. As a rule of thumb your guide error should be no worse than half your image scale if guiding is not to affect final resolution.

What tends to happen is that, if the seeing is bad, it will lower the resolution of both the guide camera and the imaging camera, causing the guiding to deteriorate as well. Since I can't do anything about this I just shoot colour on such nights!

Olly

Cheers Olly. I'm not sure i ever remembered knowing about this rule of thumb but it's super helpful now that i know

[PhotoGav]

Basically, when you get your EdgeHD 8" up and running, you will quickly want a better mount, better guiding system and better seeing!!

[ollypenrice]

12 minutes ago, PhotoGav said:

Basically, when you get your EdgeHD 8" up and running, you will quickly want a better mount, better guiding system and better seeing!!

Heheh, you're not wrong!

Olly

[souls33k3r]

24 minutes ago, PhotoGav said:

Basically, when you get your EdgeHD 8" up and running, you will quickly want a better mount, better guiding system and better seeing!!

Can someone just launch me to space and be done with it?

[jimjam11]

On 22/04/2019 at 23:50, vlaiv said:

I don't think that you are limited to 2" seeing.

Weather might be volatile most of the time, but I would not say that you have 2" most of the time, or in best conditions or whatever.

Seeing is sometime hard to judge. Many people rely on visual estimate. There is a good measure of seeing - 2s exposure FWHM. However, that measure is not precise unless corrected for aperture. In 2s exposure, it is generally accepted that mount won't make significant error. Any smooth running mount should be up for the task. However, mount 60mm guide scope and 8" scope side by side, take 2s exposure on both and you will end up with different FWHM values. This is because resulting star FWHM depends on above three parameters. Even if we take the mount out of the equation by using 2s exposures, there is still aperture that contributes. So when you see that someone is giving you 2" seeing - it might not be 2" - it could be less.

In city, you will probably have worse seeing due to local thermals.

Anyway, this is the website to check out for seeing forecast:

https://www.meteoblue.com/en/weather/forecast/seeing/london_united-kingdom_2643743

(I already selected London, but you can change the city / location)

Last column shows calculated / forecast seeing (one without impact of aperture) - as you see, all quoted figures for Tuesday are below 2". Actually for 3 days - every single value is below 2" - going as low as 1.19" on Wednesday / Thursday night (around midnight).

Best way to see if you need better guiding is to examine your RMS and FWHM you get in your images. Compare that to your imaging resolution. Good resolution is about x1.6 less than your FWHM. This means that if you have 2" FWHM (on average, or in final stack) - good imaging resolution for those conditions would be 1.25"/px. Or to put it more precisely - going with higher recording resolution (lower "/px) will result in oversampling.

Good rule of the thumb for needed RMS is half or less of your target imaging resolution. It will also depend on what your mount is capable of. For entry level mounts (and here I talk about Heq5/EQ6 family and alike, not EQ3 and similar - those are "toy" mounts - well not really, but better suited for wide field work with lens rather than scope), best you can hope for is 0.6-1.0" RMS range. 0.5-0.6 is achievable but only with tuning and modding and in exceptional conditions (no wind, stable atmosphere, great balance and such). Higher end mount will get you progressively closer to that "magical" sub 0.2" RMS.

If you want to play around with some numbers for your setup, here is spreadsheet in attachment, you can use it to calculate expected star FWHM for your aperture, guide rms and seeing. See how it compares to values you have from your subs.

FWHMCalc.ods 11.22 kB · 4 

I cannot get close to the theoretical values we discussed in another thread, so I wonder how meteoblue are calculating the seeing values, in particular for urban/suburban areas where it is almost certainly far worse than projected. 

For my setup i should be able to get some subs around 2" fwhm but I have never seen a single one < 3.5" long or short...

Mauna kea looks good though, seeing projected circa 0.15"!

[vlaiv]

7 hours ago, jimjam11 said:

I cannot get close to the theoretical values we discussed in another thread, so I wonder how meteoblue are calculating the seeing values, in particular for urban/suburban areas where it is almost certainly far worse than projected. 

For my setup i should be able to get some subs around 2" fwhm but I have never seen a single one < 3.5" long or short...

Mauna kea looks good though, seeing projected circa 0.15"!

They are integrating turbulence of different atmospheric layers. What they can't do is local thermals. These will cause trouble quite a bit.

Any sort of buildings / structures giving off heat in the night (large concrete surfaces in the summer, or heated residential areas in winder, even bodies of water like ponds / lakes) will seriously disturb seeing.

[Starwiz]

On 23/04/2019 at 00:50, vlaiv said:

Anyway, this is the website to check out for seeing forecast:

https://www.meteoblue.com/en/weather/forecast/seeing/london_united-kingdom_2643743

Thanks for the info Vlaiv, I didn't know this website existed.

[jimjam11]

I am not convinced by these numbers. Barbados, famed for its exceptional seeing is worse than the UK according to meteoblue...

https://www.meteoblue.com/en/weather/forecast/seeing/speightstown_barbados_3373505

https://www.meteoblue.com/en/weather/forecast/seeing/watford_united-kingdom_2634677

 

[vlaiv]

1 hour ago, jimjam11 said:

I am not convinced by these numbers. Barbados, famed for its exceptional seeing is worse than the UK according to meteoblue...

https://www.meteoblue.com/en/weather/forecast/seeing/speightstown_barbados_3373505

https://www.meteoblue.com/en/weather/forecast/seeing/watford_united-kingdom_2634677

 

Quite a bit of difference between the jet stream between the two - UK around 6mph, Barbados - 24mph

Maybe that would explain it?