Jupiter

[mirrorgirl1980]

Hi there

 

i took a short video of Jupiter and then ran it through Autostackert then processed it in Regis tax 6 and did final colours in photoshop here it is all comments most welcome

 

 

 

[Space Cowboy]

What scope and camera did you use?

[mirrorgirl1980]

14 hours ago, Space Cowboy said:

What scope and camera did you use?

I used my nexter 4se and my zwo 385 mc colour planetary camera

[Space Cowboy]

2 hours ago, mirrorgirl1980 said:

I used my nexter 4se and my zwo 385 mc colour planetary camera

Ok, there is quite a lot of noise on the image which suggests you need to stack more frames. At least a 3 min video would be ideal. Auto colour balance in Registax would be beneficial too as your colours are a bit out. Considering you have a small scope its a very good effort!

[bosun21]

Did you use ROI to increase the frame rate? This allows you to capture many more frames increasing the chances of capturing clearer frames in moments of good seeing.

[mirrorgirl1980]

15 minutes ago, Space Cowboy said:

Ok, there is quite a lot of noise on the image which suggests you need to stack more frames. At least a 3 min video would be ideal. Auto colour balance in Registax would be beneficial too as your colours are a bit out. Considering you have a small scope its a very good effort!

To tell you the truth the nexter 4 se is not very good at all it always leaves blurry images and i don't know why but yes next time i will do a 3 minute video thanks for that

[mirrorgirl1980]

3 minutes ago, bosun21 said:

Did you use ROI to increase the frame rate? This allows you to capture many more frames increasing the chances of capturing clearer frames in moments of good seeing.

no but i will next time thanks for that bit of info

[vlaiv]

Color is a bit weird which suggests that you did not use UV/IR cut filter with that camera?

Camera only has AR coated window (anti reflection coating) but passes full spectrum of light and sensor is sensitive in IR part of spectrum as well. If you want proper colors out of that camera you should filter the light to 400-700nm range - which means using UV/IR cut filter.

You don't have to use it if you don't mind strange color cast.

Here are some tips:

- use high FPS and low exposure time. Something like 5-6ms will work good. Don't be alarmed if video looks under exposed and planet looks very dark in video - that is ok as stacking will sort it out

- record for at least 3-4 minutes. You should get about 40000-50000 frames in total with these settings (if everything is ok - your computer is capable of recording at those speeds and you use USB 3.0). Do use ROI - no need to shoot at higher resolution than say 800x600px

- stack only 5-10% of best frames. With above frame count - that will give you plenty of frames and smooth image in the end.

[Kon]

10 minutes ago, mirrorgirl1980 said:

To tell you the truth the nexter 4 se is not very good at all it always leaves blurry images and i don't know why but yes next time i will do a 3 minute video thanks for that

You might want to check the collimation. Seeing conditions will also affect the quality of the image.

4 minutes ago, vlaiv said:

record for at least 3-4 minutes.

For Jupiter 120s max as you have fairly fast planet rotation that will blur things. If you derotate the video then go for Vlaiv's suggestion.

[Bugdozer]

It's way better than I can get using my 5SE so gets a thumbs up from me! 

[vlaiv]

2 minutes ago, Kon said:

For Jupiter 120s max as you have fairly fast planet rotation that will blur things. If you derotate the video then go for Vlaiv's suggestion.

Well, we can do some calculations to see what would happen so that we get a good idea of what is limiting total time.

Circumference of Jupiter along the equator is about ~439300Km (diameter times pi).

Rotation period of Jupiter is 9h 55m = 35700s

Speed of point in equator is thus ~12.3Km/s

Critical sampling rate for  0.4126"/px. Let's see what time it takes for a point on equator to move half pixel?

At present, Jupiter is 602.34 million Km away (according to google). Half pixel will be 0.2 arc seconds. What length will subtend angle of 0.2 arc seconds at distance of 602.34?

With speed of 12.3km, this distance will be covered in only ~47.5s!

Above suggests that we can have motion blur even under one minute!

However, we are using Autostakkert for stacking and one of its features is ability to correct for lowest order atmospheric disturbance - namely tilt. In average to very good seeing star profile will present FWHM of say 1.5-2".  That translates into 0.64 - 0.85 arc seconds of RMS displacement from true position. We can see that when we observe this slow motion recording of lunar surface:

Parts of the image "jump around" by at least half of arc second if not more and if we "freeze" the seeing - we won't get motion blur because of this - only geometric distortion. AS!3 handles geometric distortion with alignment points.

This means that software can "return" part of the image that displaced up to 1 arc second from its true position.

Now, if we have this feature in software - this means that it will also "derotate" image automatically if it moves up to say 1 arc second (or even more, depending on size of alignment point).

So we can put 1" instead of 0.2" in above calculation and get x5 higher duration. 47.5s x 5 = 237.5s or ~240s or up to 4 minutes.

In fact - if AS!3 takes middle of the recording to produce reference frame - it can correct for rotation for +/- 4 minutes around that point.

 

 

[dweller25]

@mirrorgirl1980

Your Jupiter image is a little blurry.

How do you thermally acclimate your scope ?

And how does collimation look at x200 at perfect focus ?

[Kon]

1 hour ago, vlaiv said:

Well, we can do some calculations to see what would happen so that we get a good idea of what is limiting total time.

Circumference of Jupiter along the equator is about ~439300Km (diameter times pi).

Rotation period of Jupiter is 9h 55m = 35700s

Speed of point in equator is thus ~12.3Km/s

Critical sampling rate for  0.4126"/px. Let's see what time it takes for a point on equator to move half pixel?

At present, Jupiter is 602.34 million Km away (according to google). Half pixel will be 0.2 arc seconds. What length will subtend angle of 0.2 arc seconds at distance of 602.34?

With speed of 12.3km, this distance will be covered in only ~47.5s!

Above suggests that we can have motion blur even under one minute!

However, we are using Autostakkert for stacking and one of its features is ability to correct for lowest order atmospheric disturbance - namely tilt. In average to very good seeing star profile will present FWHM of say 1.5-2".  That translates into 0.64 - 0.85 arc seconds of RMS displacement from true position. We can see that when we observe this slow motion recording of lunar surface:

Parts of the image "jump around" by at least half of arc second if not more and if we "freeze" the seeing - we won't get motion blur because of this - only geometric distortion. AS!3 handles geometric distortion with alignment points.

This means that software can "return" part of the image that displaced up to 1 arc second from its true position.

Now, if we have this feature in software - this means that it will also "derotate" image automatically if it moves up to say 1 arc second (or even more, depending on size of alignment point).

So we can put 1" instead of 0.2" in above calculation and get x5 higher duration. 47.5s x 5 = 237.5s or ~240s or up to 4 minutes.

In fact - if AS!3 takes middle of the recording to produce reference frame - it can correct for rotation for +/- 4 minutes around that point.

 

 

I like how you always back your argument with a calculation. Very interesting. I used to go for 3 min based on this feature but I was told that the 3 min is way too long for as!3 to handle this dropping to 2min. I haven't done a side by side comparison to see the effect of 2 min vs 3 or 4. I know others are down to 30s to 1min with the new  low noise cameras.

But based on what you posted it should be doable. We just need some excellent seeing to put in practice.

[vlaiv]

4 minutes ago, Kon said:

I like how you always back your argument with a calculation. Very interesting. I used to go for 3 min based on this feature but I was told that the 3 min is way too long for as!3 to handle this dropping to 2min. I haven't done a side by side comparison to see the effect of 2 min vs 3 or 4. I know others are down to 30s to 1min with the new  low noise cameras.

But based on what you posted it should be doable. We just need some excellent seeing to put in practice.

Yes, there is easy way to test this - do a 4 minute video for example (or even 5 minute) and stack that using best 10% or whatever percentage of frames, then simply take the same video and split it into 2 parts of 2 minutes and then stack those using same percentage of frames.

SNR difference won't be huge and you will see if there is issue with rotation. If two short videos produce images of same quality that means that seeing was similar during whole session and that 4 minute video won't be poorer due to seeing alone. If there is a difference it will be due to rotation.

[vlaiv]

@Kon

By the way, above calculation for 8" telescope suggests that one should limit their video to half of that of 4" (because it can resolve twice as much, and resolved blur will be smaller) - so only about ~23 seconds for rotation blur to kick in without alignment points - yet you probably use videos 2-3 minutes long without issues as AS!3 deals with any rotation that happens on those scales.

[Kon]

41 minutes ago, vlaiv said:

@Kon

By the way, above calculation for 8" telescope suggests that one should limit their video to half of that of 4" (because it can resolve twice as much, and resolved blur will be smaller) - so only about ~23 seconds for rotation blur to kick in without alignment points - yet you probably use videos 2-3 minutes long without issues as AS!3 deals with any rotation that happens on those scales.

Thanks. Can you share the link to the calculator please? I don't think I have seen it before.

[vlaiv]

9 minutes ago, Kon said:

Thanks. Can you share the link to the calculator please? I don't think I have seen it before.

Here it is:

https://www.1728.org/angsize.htm

[Space Cowboy]

5 hours ago, mirrorgirl1980 said:

To tell you the truth the nexter 4 se is not very good at all it always leaves blurry images and i don't know why but yes next time i will do a 3 minute video thanks for that

Could be collimation is out. Have you done a star test before imaging a planet?

[mirrorgirl1980]

21 hours ago, Space Cowboy said:

Could be collimation is out. Have you done a star test before imaging a planet?

On 15/10/2023 at 13:53, Space Cowboy said:

Ok, there is quite a lot of noise on the image which suggests you need to stack more frames. At least a 3 min video would be ideal. Auto colour balance in Registax would be beneficial too as your colours are a bit out. Considering you have a small scope its a very good effort!

I think i did

To tell you the truth the nexter 4 se is not very good at all it always leaves blurry images and i don't know why but yes next time i will do a 3 minute video thanks for that

[mirrorgirl1980]

On 15/10/2023 at 14:58, dweller25 said:

@mirrorgirl1980

Your Jupiter image is a little blurry.

How do you thermally acclimate your scope ?

And how does collimation look at x200 at perfect focus ?

A bit blurry

[Cosmic Geoff]

You should use a filter as suggested above to get the colours right.  The image could be a little blurry for various reasons - slightly out of focus (what do you focus on?) or shooting over your neighbours' chimney, or just bad seeing.  

The 4 SE is a Maksutov, so the collimation should remain good unless you are unlucky.

Yesterday I took a series of images of Jupiter with an 8" SCT through thin cloud, and it wasn't any better than the above.  The setup can do far better. I have never used extremely long videos - in reasonable conditions 5000 frames seems enough - sometimes acquired in only 18 seconds.

[Space Cowboy]

1 hour ago, Cosmic Geoff said:

You should use a filter as suggested above to get the colours right.  The image could be a little blurry for various reasons - slightly out of focus (what do you focus on?) or shooting over your neighbours' chimney, or just bad seeing.  

The 4 SE is a Maksutov, so the collimation should remain good unless you are unlucky.

Yesterday I took a series of images of Jupiter with an 8" SCT through thin cloud, and it wasn't any better than the above.  The setup can do far better. I have never used extremely long videos - in reasonable conditions 5000 frames seems enough - sometimes acquired in only 18 seconds.

Why restrict yourself to 5000 frames? The more frames you have more more selective you can be in Austostakert without worrying about noise kicking in. 10-20000 frames gives you far more options.

[dweller25]

@mirrorgirl1980

If possible leave your scope outside  for an hour to let it cool down.

Then check a slightly out of focus star at x200 a star should look like this with the circles all concentric around the central dot…..

 

 

[mirrorgirl1980]

1 minute ago, dweller25 said:

@mirrorgirl1980

If possible leave your scope outside  for an hour to let it cool down.

Then check a slightly out of focus star at x200 a star should look like this with the circles all concentric around the central dot…..

 

 

THANK YOU FOR THAT INFO

[dweller25]

And if it looks like this then your scope needs collimating to get sharp images….