Advice needed please, re guide camera...

[Stuart1971]

Hello,

I recently bought a Takahashi FSQ 85 imaging scope, f5.3 and an Tak FS60 f5.9 for use as a guide scope amongst other jobs...

My imaging camera is an SXVR M25c Camera with 7.8 micron pixels, so all that being said what would be a good guide camera, ie pixel size to look for..?

Thanks in advance. 👍😀

Edited July 28, 2020 by Stuart1971

[Stuart1971]

Ok, so I found an online calculator, and this is what is gave me, putting in the SX lodestar x2 as a guide cam choice, as I was considering one....but don’t know if this is good or not... 🤔

 

[michael8554]

3.5 arcsecs/pixel for the Imaging camera is Undersampling.

Try a pixel size that gives you a resolution of 1.0 arcsec/pixel or thereabouts.

The pixel size of the SXVR M25c camera would be suitable for the Guidescope.

Michael

[Stuart1971]

9 hours ago, michael8554 said:

3.5 arcsecs/pixel for the Imaging camera is Undersampling.

Try a pixel size that gives you a resolution of 1.0 arcsec/pixel or thereabouts.

The pixel size of the SXVR M25c camera would be suitable for the Guidescope.

Michael

Well I have no choice with that, as I have my imaging scope and imaging camera already....And seeing In the U.K. does not allow much below 2.0 arcsec/pixel anyway...

My question was about the guiding part of it.... but thanks anyway...👍

[michael8554]

39 minutes ago, Stuart1971 said:

My question was about the guiding part of it

A 7 to 8 um pixel size would be suitable for guiding a 355mm FL guidescope.

Or consider a 200mm FL guidescope and a guidecam with 3.75um pixels, plenty of choice with that size.

The M25C has the right size pixels for guiding, but best to go for a Mono guide camera, for the extra sensitivity.

41 minutes ago, Stuart1971 said:

seeing In the U.K. does not allow much below 2.0 arcsec/pixel anyway...

Agreed, for which you need 1.0arcsec/pixel resolution.

Michael

[Stuart1971]

33 minutes ago, michael8554 said:

Agreed, for which you need 1.0arcsec/pixel resolution.

Michael

This does not make sense, why do you need a resolution that is more than the seeing will allow, thats pointless... 🤔 if your seeing will only allow a max of 2 arcsec/pixel why should you try and image at 1 . it’s the other way round 

[tooth_dr]

58 minutes ago, Stuart1971 said:

This does not make sense, why do you need a resolution that is more than the seeing will allow, thats pointless... 🤔 if your seeing will only allow a max of 2 arcsec/pixel why should you try and image at 1 . it’s the other way round 

You should aim to guide at 1/2 your imaging resolution.

[tooth_dr]

One guidecam of choice these days is the 290MM, I wouldnt get too hung up on resolution of the guidecam though

[Stuart1971]

15 minutes ago, tooth_dr said:

You should aim to guide at 1/2 your imaging resolution.

Yes, but he was saying I need to image at 1 arcsec / pixel when seeing only allows 2 max... not guiding...

[michael8554]

7 minutes ago, tooth_dr said:

if your seeing will only allow a max of 2 arcsec/pixel why should you try and image at 1 . it’s the other way round 

Ah, if only life was that forgiving.

Nyquist Theory, you sample at twice the required quality.

Audio example - to get 20KHz highest frequency, Digital Audio is sampled at 44.1KHz or more.

So to get 2 arcsecs pixel result, you sample at twice that resolution, 1 arcsec/pixel.

9 minutes ago, tooth_dr said:

You should aim to guide at 1/2 your imaging resolution.

That could well work as the relationship between Guide image scale and Imaging image scale has never really been well pined down, but 3 to 4 times Imaging scale seems to be common.

Michael

 

[tooth_dr]

18 minutes ago, Stuart1971 said:

Yes, but he was saying I need to image at 1 arcsec / pixel when seeing only allows 2 max... not guiding...

I believe you should aim for a guiding ratio of 1:4, but I've seen people guiding at 1:10.  If you buy a sensitive small pixel camera like the 290MM or 120MM, you cant go far wrong surely?

Edited July 29, 2020 by tooth_dr

[Stuart1971]

3 minutes ago, tooth_dr said:

I believe you should aim for a guiding ratio of 1:4, but I've seen people guiding at 1:10.  If you buy a sensitive small pixel camera like the 290MM or 120MM, you cant go far wrong surely?

Yes I get that, but we were talking about imaging pixel scale not guiding....what is the point in imaging at less than 2 arcsec / pixel if the seeing will not allow, that was my point...? @michael8554 said I should be....

[tooth_dr]

On 28/07/2020 at 12:54, Stuart1971 said:

what would be a good guide camera,

290MM

[Stuart1971]

27 minutes ago, tooth_dr said:

290MM

Ok, you missed my point about the individual discussion with the other person on the thread....no worries....👍

[vlaiv]

Let's try it this way:

- re sampling resolution - depends on what scope you are using, quality of your mount and how wide you want to go.

1"/px is very high resolution and in most cases unattainable. You need mount that guides at least at 0.5" RMS or lower. You need very steady skies and you need at least 8" of aperture to get there.

For telescope of 80-100mm of aperture, realistic maximum sampling rate is at about 2"/px. This does not mean that you have to go that high - if looking for wide field setup - simple fact is that you have limited size of sensor / corrected field and you won't be able to fit that many pixels. 3.5"/px is fine sampling rate for wide field.

If we want to be specific about max sampling rate and Nyquist - there is simple rule to follow - measure your FWHM and go with sampling rate that is about 1.6 less than that. This means that one needs 1.6" FWHM stars in order to fully exploit 1"/px sampling rate.

- re guiding resolution. Well depends on the mount you have and what is realistically achievable in terms of guide RMS.

My advice would be to sample at about x3 per best possible RMS in terms of centroid accuracy. Centroid accuracy is about 1/16 - 1/20 of single pixel.

To give a bit better explanation, here is how to calculate guider resolution.

Let's say that you have mount that is capable of 0.5" RMS guiding under best circumstances. You want your centroid accuracy to be about 0.5/3 = 0.167". That will be 1/16 to 1/20 of a pixel so pixel size needs to be 0.167 * 16 to 0.167 * 20 = 2.67"/px to 3.34"/px

You have guide scope that is 60mm F/5.9 or about 350mm of FL. You need a camera that has less than 6um pixel size to use as a guiding camera.

With most common pixel size of 3.75um and 350mm of FL, you'll get 2.21"/px - which is slightly better than you need.

With 290mm camera, which is a good choice for guide camera, you'll have 1.71"/px and if you choose that camera, then use x2 bin to further improve it's sensitivity as it will still provide you with 3.42"/px - close enough for above criteria.

Hope this helps?

 

[Stuart1971]

15 minutes ago, vlaiv said:

Let's try it this way:

- re sampling resolution - depends on what scope you are using, quality of your mount and how wide you want to go.

1"/px is very high resolution and in most cases unattainable. You need mount that guides at least at 0.5" RMS or lower. You need very steady skies and you need at least 8" of aperture to get there.

For telescope of 80-100mm of aperture, realistic maximum sampling rate is at about 2"/px. This does not mean that you have to go that high - if looking for wide field setup - simple fact is that you have limited size of sensor / corrected field and you won't be able to fit that many pixels. 3.5"/px is fine sampling rate for wide field.

If we want to be specific about max sampling rate and Nyquist - there is simple rule to follow - measure your FWHM and go with sampling rate that is about 1.6 less than that. This means that one needs 1.6" FWHM stars in order to fully exploit 1"/px sampling rate.

- re guiding resolution. Well depends on the mount you have and what is realistically achievable in terms of guide RMS.

My advice would be to sample at about x3 per best possible RMS in terms of centroid accuracy. Centroid accuracy is about 1/16 - 1/20 of single pixel.

To give a bit better explanation, here is how to calculate guider resolution.

Let's say that you have mount that is capable of 0.5" RMS guiding under best circumstances. You want your centroid accuracy to be about 0.5/3 = 0.167". That will be 1/16 to 1/20 of a pixel so pixel size needs to be 0.167 * 16 to 0.167 * 20 = 2.67"/px to 3.34"/px

You have guide scope that is 60mm F/5.9 or about 350mm of FL. You need a camera that has less than 6um pixel size to use as a guiding camera.

With most common pixel size of 3.75um and 350mm of FL, you'll get 2.21"/px - which is slightly better than you need.

With 290mm camera, which is a good choice for guide camera, you'll have 1.71"/px and if you choose that camera, then use x2 bin to further improve it's sensitivity as it will still provide you with 3.42"/px - close enough for above criteria.

Hope this helps?

 

Why do I need a pixel size of 6 micron to guide with my guide scope, which is 60mm 355mm, as the x2 lodestar with 8.2 still gives me a good sampling compared to my imaging scope, 1.33x..
see the example I posed above of my set up...

and why Bin the ASI290 as that doubles the pixel size, so why not just use a camera like the X2 lodestar to start with unbinned..?

 

[vlaiv]

8 minutes ago, Stuart1971 said:

Why do I need a pixel size of 6 micron to guide with my guide scope, which is 60mm 355mm, as the x2 lodestar with 8.2 still gives me a good sampling compared to my imaging scope, 1.33x..
see the example I posed above of my set up...

and why Bin the ASI290 as that doubles the pixel size, so why not just use a camera like the X2 lodestar to start with unbinned..?

 

Sampling rate of main camera and sampling rate of guide camera are only related by bunch of rules of thumb but in principle are not related.

We could say that there is two schools of thought at work here.

1. Good enough

2. Best given circumstances

Using first school of thought you can get to relationship between guide resolution and imaging resolution and it goes something like this:

- under reasonable circumstances you need your guide RMS to be at least half of your imaging resolution. You probably heard about this rule of the thumb

- There is 1/16 centroid accuracy thing and x3 of that accuracy per guide RMS which generally give about x5 guiding resolution vs guide RMS.

Put those two together and you get our general rule of thumb that says something - make your guide resolution about x2.5 that of imaging resolution.

This approach is "good enough" kind of approach, and using this approach you are fine with Lodestar as it gives you x1.33 ratio to your imaging resolution.

Let's now try a bit different approach - approach number two that will try to do the best given circumstances.

Rationale behind this approach is simple - if you can, why not go for best result given your circumstances?

To explain it a bit better - your imaging resolution is 3.55"/px. Does this mean that you should settle for 1.77" RMS guide error just because rule of thumb says that it should be (at least) half of imaging resolution? Notice that - at least part in parenthesis.

If your mount is capable of doing 0.8" RMS guide precision - why not go with that. If you compare two images - both sampled at 3.55"/px and one being guided at 0.8" RMS and other at 1.7" RMS you will see the difference in sharpness between the two - regardless of the fact that you are sampling at rather low rate.

Why not aim for the sharpest possible image given your setup?

In this approach - we don't look at guide resolution in relation to imaging resolution, we look at guide resolution with respect to what your mount can achieve at its best.

This is your starting point. What mount do you have and what sort of guide performance does such mount usually have?

I gave you an example for mount capable of 0.5" RMS. But we don't have to do it like that - we can do reverse. Given your guide scope at 355mm and lodestar with 8.2um pixel size - what would be the best mount performance this combo is capable of guiding.

This works out to 4.76"/px, so if we take 1/16 of that, that is about 0.3". Multiply that with at least x3 to get resulting RMS - and that is 0.9" RMS. So that setup is only good for mounts that on average do above 1" RMS and only in exceptional circumstances can achieve 0.9" RMS.

HEQ5/EQ6/AZEQ6 and all of those mounts are capable of better guide performance. Why limit them with guide setup that can't measure star position with enough accuracy to be able to instruct the mount to track better?

23 minutes ago, Stuart1971 said:

why Bin the ASI290 as that doubles the pixel size, so why not just use a camera like the X2 lodestar to start with unbinned..?

ASI290 has 2.9um pixel size, while X2 lodestar has 8.2um pixel size. One is too large and does not provide enough resolution to be able guide mount like HEQ5/EQ6, while other has too much resolution with said guide scope - you don't need that much precision and smaller pixel size is less sensitive (part of guide performance comes from good SNR on guide star so you want that as well).

You can't split pixels on Lodestar, but you can bin pixels on ASI290 to get to pixel size that both balances sensitivity and resolution needed to determine star position to a good precision needed to properly guide mount like HEQ5/EQ6.

If you on the other hand have mount like Mesu200 or similar that can guide below 0.3" RMS - then I would say - use ASI290 but don't bin it - as you'll need that much precision in order to guide mount with such low error.

 

[alacant]

On 28/07/2020 at 13:54, Stuart1971 said:

a good guide camera

Hi

No theory. Just hands on... We guide a DSLR from a 55mm lens to a 1000mm f5 reflector with a 60mm f4 refractor using a zwo120mm or it's clone, the t7m. 

Works fine 

So, to answer your question from a practical point of view, the asi120mm. Save some money that way too;)

Cheers and HTH.

[vlaiv]

2 minutes ago, alacant said:

Hi

No theory. Just hands on... We guide a DSLR from a 55mm lens to a 1000mm f5 reflector with a 60mm f4 refractor using a zwo120mm or it's clone, the t7m. 

Works fine 

So, to answer your question from a practical point of view, the asi120mm. Save some money that way too;)

Cheers and HTH.

But then again - you are not actually guiding either 55mm lens nor 1000mm F/5 reflector - you are guiding the mount holding either of those

 

[alacant]

6 minutes ago, vlaiv said:

guiding the mount

Ahhgghh!!!

[Stuart1971]

1 hour ago, vlaiv said:

Sampling rate of main camera and sampling rate of guide camera are only related by bunch of rules of thumb but in principle are not related.

We could say that there is two schools of thought at work here.

1. Good enough

2. Best given circumstances

Using first school of thought you can get to relationship between guide resolution and imaging resolution and it goes something like this:

- under reasonable circumstances you need your guide RMS to be at least half of your imaging resolution. You probably heard about this rule of the thumb

- There is 1/16 centroid accuracy thing and x3 of that accuracy per guide RMS which generally give about x5 guiding resolution vs guide RMS.

Put those two together and you get our general rule of thumb that says something - make your guide resolution about x2.5 that of imaging resolution.

This approach is "good enough" kind of approach, and using this approach you are fine with Lodestar as it gives you x1.33 ratio to your imaging resolution.

Let's now try a bit different approach - approach number two that will try to do the best given circumstances.

Rationale behind this approach is simple - if you can, why not go for best result given your circumstances?

To explain it a bit better - your imaging resolution is 3.55"/px. Does this mean that you should settle for 1.77" RMS guide error just because rule of thumb says that it should be (at least) half of imaging resolution? Notice that - at least part in parenthesis.

If your mount is capable of doing 0.8" RMS guide precision - why not go with that. If you compare two images - both sampled at 3.55"/px and one being guided at 0.8" RMS and other at 1.7" RMS you will see the difference in sharpness between the two - regardless of the fact that you are sampling at rather low rate.

Why not aim for the sharpest possible image given your setup?

In this approach - we don't look at guide resolution in relation to imaging resolution, we look at guide resolution with respect to what your mount can achieve at its best.

This is your starting point. What mount do you have and what sort of guide performance does such mount usually have?

I gave you an example for mount capable of 0.5" RMS. But we don't have to do it like that - we can do reverse. Given your guide scope at 355mm and lodestar with 8.2um pixel size - what would be the best mount performance this combo is capable of guiding.

This works out to 4.76"/px, so if we take 1/16 of that, that is about 0.3". Multiply that with at least x3 to get resulting RMS - and that is 0.9" RMS. So that setup is only good for mounts that on average do above 1" RMS and only in exceptional circumstances can achieve 0.9" RMS.

HEQ5/EQ6/AZEQ6 and all of those mounts are capable of better guide performance. Why limit them with guide setup that can't measure star position with enough accuracy to be able to instruct the mount to track better?

ASI290 has 2.9um pixel size, while X2 lodestar has 8.2um pixel size. One is too large and does not provide enough resolution to be able guide mount like HEQ5/EQ6, while other has too much resolution with said guide scope - you don't need that much precision and smaller pixel size is less sensitive (part of guide performance comes from good SNR on guide star so you want that as well).

You can't split pixels on Lodestar, but you can bin pixels on ASI290 to get to pixel size that both balances sensitivity and resolution needed to determine star position to a good precision needed to properly guide mount like HEQ5/EQ6.

If you on the other hand have mount like Mesu200 or similar that can guide below 0.3" RMS - then I would say - use ASI290 but don't bin it - as you'll need that much precision in order to guide mount with such low error.

 

So you are saying that the lodestar x2 has too larger pixels to guide an HEQ6  and EQ6 with a 60mm guide scope, yet many people do, and very well..so I am confused...

If it helps I have the EQ8...

[vlaiv]

9 minutes ago, Stuart1971 said:

So you are saying that the lodestar x2 has too larger pixels to guide an HEQ6  and EQ6 with a 60mm guide scope, yet many people do, and very well..so I am confused...

If it helps I have the EQ8...

No, I'm not saying any of that and honestly, I'm failing to see what in my post lead you to that conclusion.

First - I'm not saying that Lodestar X2 has too large pixels to guide HEQ5/EQ6 class mount. That depends on focal length used - not on aperture size.

I'm also not saying that one can't use Lodestar X2 with certain focal length to guide HEQ5/EQ6 class mount even if guide resolution is too coarse - you can certainly guide your mount, only question is how well?

In first post you asked what would be a good guide camera. I understood term good in a certain way - for me good guide camera will one that will enable mount to perform to its best (given choice of guide scope).

I've shown that you can't reliably guide below about 1" RMS with Lodestar X2 and 60mm F/5.9 scope simply because it lacks precision to measure star position accurately. If you want to use Lodestar to guide to better precision - you certainly can - just change focal length of guide scope so that it has better precision.

Alternatively, if you don't want to change guide scope - then select guide camera that has smaller pixels - sufficiently small to enable good centroid precision for your mount.

Btw, I would expect EQ8 class mount to guide to at least 0.5" RMS or less and above calculation and recommendation is valid.

[Stuart1971]

49 minutes ago, vlaiv said:

No, I'm not saying any of that and honestly, I'm failing to see what in my post lead you to that conclusion.

First - I'm not saying that Lodestar X2 has too large pixels to guide HEQ5/EQ6 class mount. That depends on focal length used - not on aperture size.

I'm also not saying that one can't use Lodestar X2 with certain focal length to guide HEQ5/EQ6 class mount even if guide resolution is too coarse - you can certainly guide your mount, only question is how well?

In first post you asked what would be a good guide camera. I understood term good in a certain way - for me good guide camera will one that will enable mount to perform to its best (given choice of guide scope).

I've shown that you can't reliably guide below about 1" RMS with Lodestar X2 and 60mm F/5.9 scope simply because it lacks precision to measure star position accurately. If you want to use Lodestar to guide to better precision - you certainly can - just change focal length of guide scope so that it has better precision.

Alternatively, if you don't want to change guide scope - then select guide camera that has smaller pixels - sufficiently small to enable good centroid precision for your mount.

Btw, I would expect EQ8 class mount to guide to at least 0.5" RMS or less and above calculation and recommendation is valid.

You said exactly that here....

“ASI290 has 2.9um pixel size, while X2 lodestar has 8.2um pixel size. One is too large and does not provide enough resolution to be able guide mount like HEQ5/EQ6”

so would a 1.5x barlow on the guide scope help..?

I have to say your replys are far to advanced for me to understand....sorry but true....but thanks for taking the time to try and explain....but I need lamens terms... 😩👍

Edited July 29, 2020 by Stuart1971

[vlaiv]

14 minutes ago, Stuart1971 said:

You said exactly that here....

Yes, you are right, my mistake. That is not quite what I wanted to say. I was talking about best performance of the mount rather than ability to guide at all, but you are right, I should have been more specific and clear about that.

15 minutes ago, Stuart1971 said:

so would a 1.5x barlow on the guide scope help..?

Indeed - barlow will provide additional focal length and additional resolution. That will make centroid algorithm more precise. It is not about pixel size, nor focal length alone - combination of the two is important.

16 minutes ago, Stuart1971 said:

I have to say your replys are far to advanced for me to understand....sorry but true....but thanks for taking the time to try and explain....but I need lamens terms... 😩

Ah, ok, sorry about that.

I'll try to be more straight forward with my explanations. It is about how precise something can be measured.

Imagine you are trying to park a car in a garage where door is 3.2m wide, but I give you directions in whole meters - so I say either stay on course or shift left/right but one whole meter each time. Odds are - you will miss the garage door and hit the wall.

Mount can be directed quite precisely in the sky but you don't tell it in precise units how much to move - using large pixels gives large "units" of move left/right (it depends on focal length).

If you want mount to move precisely where its supposed to be - you need to tell it to move in small enough units.

All the math above just calculates how small units you need depending on how fine your mount is. You can still use larger "units" in your directions but, and mount will respond but it simply won't be as precise as it can be - due to "coarse directions" rather than anything else.

All of that is related to sharpness of your image. If mount moves a lot causing larger RMS guide error - that adds to blur and image becomes less sharp. It is one of contributing factors of overall resolution achieved - other two being aperture size and seeing.

It is therefore straight forward to see that you want as low (true) guide RMS as possible.

As pointed above, ASI120mm will do excellent job - both in pixel size and in general. It will also be cheaper than Lodestar X2. Do you have any particular reason to prefer Lodestar + Barlow option?

 

 

 

[Stuart1971]

8 minutes ago, vlaiv said:

Yes, you are right, my mistake. That is not quite what I wanted to say. I was talking about best performance of the mount rather than ability to guide at all, but you are right, I should have been more specific and clear about that.

Indeed - barlow will provide additional focal length and additional resolution. That will make centroid algorithm more precise. It is not about pixel size, nor focal length alone - combination of the two is important.

Ah, ok, sorry about that.

I'll try to be more straight forward with my explanations. It is about how precise something can be measured.

Imagine you are trying to park a car in a garage where door is 3.2m wide, but I give you directions in whole meters - so I say either stay on course or shift left/right but one whole meter each time. Odds are - you will miss the garage door and hit the wall.

Mount can be directed quite precisely in the sky but you don't tell it in precise units how much to move - using large pixels gives large "units" of move left/right (it depends on focal length).

If you want mount to move precisely where its supposed to be - you need to tell it to move in small enough units.

All the math above just calculates how small units you need depending on how fine your mount is. You can still use larger "units" in your directions but, and mount will respond but it simply won't be as precise as it can be - due to "coarse directions" rather than anything else.

All of that is related to sharpness of your image. If mount moves a lot causing larger RMS guide error - that adds to blur and image becomes less sharp. It is one of contributing factors of overall resolution achieved - other two being aperture size and seeing.

It is therefore straight forward to see that you want as low (true) guide RMS as possible.

As pointed above, ASI120mm will do excellent job - both in pixel size and in general. It will also be cheaper than Lodestar X2. Do you have any particular reason to prefer Lodestar + Barlow option?

 

 

 

I have the chance to buy an x2 for a very good price, (£200)  a bargain really, hence banging on about it... 👍😀