ZWO Seestar 50

[Cosmic Geoff]

59 minutes ago, Elp said:

How does it cope with tracking at these exposures?

I couldn't see a difference between 10 sec and 20 sec exposure stacks of the same object & field, except that the 20 sec stack was brighter. 

[Geoff Lister]

9 hours ago, Elp said:

Now that is interesting. How does it cope with tracking at these exposures? Also has anyone kept a log of hours imaged and exposure time lost per session. Seriously thinking of getting one as a no nonsense AP setup rather than spending time setting up the "serious" stuff.

I found a couple of YouTube videos comparing 10/20/30 second exposures on M5, M42 & M45. The presenter noted that at the 30 second setting, the stacking software was rejecting a significant number of images from the stack, and stars were, as expected egg shaped. A histogram comparason showed that the 10 & 20 second exposures were similar, but there was significantly more spread on the 30. The conclusion was that, for M42, the 20 may give slight advantages for nebulosity, but that the 10 gave better definition for the stars in the Trapezium.

As field rotation with an Az/Alt mount is minimum at azimuth 90 and 270 and low altitude, perhaps some targets in these 2 directions will benefit from the 30 option.

Geoff

[LDW1]

So on dimmer / fainter dso's ie M101, M33 etc. the 20 sec and maybe the 30 sec should provide a better end result if you can find / hit on the correct duration ?

[Elp]

I don't think it makes a difference for per sub length as long as the total time remains constant. Speed of light is the speed of light. The benefit is if it's windy whilst doing it or for less occurrence of tracking error in the final stack when keeping exposures short.

[LDW1]

15 minutes ago, Elp said:

I don't think it makes a difference for per sub length as long as the total time remains constant. Speed of light is the speed of light. The benefit is if it's windy whilst doing it or for less occurrence of tracking error in the final stack when keeping exposures short.

But on the great nights wouldn't each take have more data inbedded in it if you were using 20 sec. or maybe 30 sec.wouldn't the sum total be all round more enhanced, of higher quality.  Lets disrgard wind, stray light etc. affects in this case. I am just trying to get my mind around the purpose of 20 / 30 sec exposures because up til now my 10 sec. results are excellent for the SS and what it is.

[Elp]

No. Total time is total time. Photons are flying through space regardless and stacking is an averaging (addition) process. The main difference will be the amount of read noise in the image, more shorter images and the greater the read noise signal will be in the image in comparison to a stack with longer exposure but less images.

 

[Elp]

People expose longer doing narrowband because the filters block out some signal, so a longer per sub time is required for the pixels to register received signal. You'll see this straight away if you ever try to polar align using an o3 filter vs no filter.

Edited January 12 by Elp

[LDW1]

7 minutes ago, Elp said:

No. Total time is total time. Photons are flying through space regardless and stacking is an averaging (addition) process. The main difference will be the amount of read noise in the image, more shorter images and the greater the read noise signal will be in the image in comparison to a stack with longer exposure but less images.

 

I thought maybe it was the way the SS, its electronics handled the photons in each longer take that would make a difference in end product, a better photo if you will ?

Edited January 12 by LDW1

[Elp]

You'd think it. But one ten minute image will be similar to one made up of ten one minute images, but the latter will be a bit more noisy due to containing an average of ten lots of sensor read noise opposed to the formers one.

Think of it a bit like this, if you ran a tap of water into a large bucket, didn't change the tap rate and repeated the same time collecting water over multiple buckets, the amount of water collected will be similar if not the same.

A difference occurs though when you begin to change pixel imaging scales, as the Seestar is fixed this doesn't apply in this case.

Edited January 12 by Elp

[powerlord]

ooo, I'm no expert Elp, but don't think that's correct. @vlaiv will explain it better than I can though so I best not try.

 

[Elp]

I'm only going by this as I've asked the question previously:

https://stargazerslounge.com/topic/413390-bortle-7-skies-should-i-take-180s-or-300s-exposures/?do=findComment&comment=4410585

 

[happy-kat]

The IMX462 has really low read noise. Do we know what ISO the seestar generally uses?

[vlaiv]

6 hours ago, Elp said:

You'd think it. But one ten minute image will be similar to one made up of ten one minute images, but the latter will be a bit more noisy due to containing an average of ten lots of sensor read noise opposed to the formers one.

This is correct.

Question is - just how much more noise will be in 10x1minute vs 1x10minute exposure, and that depends on how high is read noise compared to other noise sources - most notably the highest one which is usually light pollution noise.

Noise adds like linearly independent vectors - square root of sum of squares. If one component is much smaller (just to get the idea of what we are talking here - if read noise is x3-x5 smaller than any other noise level in that single exposure) - you simply won't see the difference.

Humans can't tell the difference of say 5% in noise increase. This 5% corresponds to ~x3 larger light pollution noise.

In another words - if you have some read noise and per exposure you have x3 more LP noise - that is the same as having 5% more LP noise and no read noise at all. When you have 0e read noise then 10x1 minute is equal to 1x10minute as far as SNR goes (tracking and other artifacts not included - just signal quality).

Only issue is that LP noise depends on LP signal - which depends on sky conditions but also sampling rate of the camera as it is amount of photons from the sky that end up hitting each pixel.

General rule:

Faster the optics - shorter exposures can be used

Brighter the sky - shorter exposures can be used

[Geoff Lister]

Charging the Seestar - The USB Lead Matters

As a recent SS 50 purchaser, I was interested in the charging time for the internal battery. As the SS is supplied with a USB C lead, but no mains adaptor, it is reliant on power from other sources.

I was using an old 5V 2A plug-top supply provided with a redundant tablet, and a generic, budget, USB lead, but the SS seemed slow to illuminate more bars. I plugged the power supply into an in-line mains power meter, and the reading was about 5W, with the PSU case close to ambient temperature. I replaced the lead with the one supplied with the SS, and the input power showed close to 15W, the PSU was slightly warm to the touch. The power input and PSU case temperature later dropping as the charge approached 100%.

I can only conclude that the budget cable's conductor resistance was fine for successful USB data transfer and a 500mA slow charge, but to get closer to 3A, I needed to use the SS-supplied lead with its (noticeably) thicker cable.

Geoff

 

[powerlord]

You're nearly right.

The proper cable may or may not have thicker wires, but the secret is, it has MORE of them.

The reason is that with USB, the data lines are used to tell the PSU what voltage and current the device can take (USB-C in addition can supply all sorts of voltage and current levels). In absence of these, a PSU will only deliver 5v at 1a max at 5v usually. most devices will play it safe talking to a 'dumb' PSU (which is how they now see it thanks to the cable), and will draw only 500ma or 1a  incase they cause the PSU to stop working (most cheap 500ma ones will just drop to low voltage if you try to draw more than 500ma from them or worse - do something nasty). I'm simplifying here - usbc actually spreads the current across multiple different wires in the cable - 12 in total for data and power)

Your cheap usb lead undoubledly had only 2 wires inside (Earth and +ve), hence you got 5v and 500ma or 1a. it's a 'dumb cable'

The proper lead will have a minimum of 4 wires and  lets the S50 negotiate and probably get its 2a.

A more sophisticated negotiation happens with things like new phones and laptops where USB4/C features are used, and they might get 12v and 2a or something - and then even more than 4 wires are required!

USB does not help itself here, in stuffing 'USB' marks on cables whether they are 2 wire or more (usbc can be 12 wire!)

To avoid this sort of thing happening (and this is true to a lesser extent with usb2 too where 4 wires are required for full power negotiation), I always bin usb leads that come with devices for charging. They will nearly always (unless its something like a laptop or phone), be rubbish 2 wire things. Before you bin it, pull it apart, and wonder at the 2 wires, or the fact you've just destroyed a 4 wire one 🙂

Stick to using good quality leads with are advertised as 'data leads'. For USB-C just to confuse things further, you can instead just search for 'PD' leads (power delivery). But data cables will also work fine (though not for very high 40w+ delivery required for laptops unless specced accordingly).

The thing that really annoys me, is this is there 5th go at getting usb right - and still its a bloody mess.

[Geoff Lister]

6 hours ago, powerlord said:

The proper cable may or may not have thicker wires, but the secret is, it has MORE of them.

The reason is that with USB, the data lines are used to tell the PSU what voltage and current the device can take

I agree that USB C has greater capability than earlier versions of USB. The PSU that I used has a standard USB A, 4-pin, output socket, and both cables have the standard 4-pin A plug. The PSU was manufactured long before USB C was marketed.

I had a squint into the C connectors on both cables. The budget (thin) cable had all 24 pin positions populated, whereas the Seestar cable has only 12: 4 power, 4 ground, a USB2 data pair on one side only, and (I may be wrong here) the Configuration Channel CC1 & CC2 pins. It is clear that the only way for the average punter to judge the quality of a cable is to destroy it; but it is more likely that a thicker cable contains more useful copper.

Use the Seestar's supplied cable if you want to recharge the internal battery quickly, or extend operations with an external supply.

Geoff

[Cosmic Geoff]

Re the 10/20/30 second exposures on the Seestar S50: My experience of doing EVAA with a 102mm refractor is that the longer the sub exposure time , the brighter the image, and  live-stacking more subs mainly reduces the noise.

With the Seestar, my initial impression on trying 10 and 20 second exposures on the same object was that the 20 sec version was brighter.  My provisional conclusion is that a lot of objects would benefit from using the longer exposure.  If one examines the raw 10 sec. subs, a lot of them require extreme stretching to make anything at all visible.

[squipper]

my next attempts to  image the sun think i am getting the hang of it at last

[vlaiv]

2 minutes ago, Cosmic Geoff said:

My experience of doing EVAA with a 102mm refractor is that the longer the sub exposure time , the brighter the image

This is often believed and it is so - but solely because of the way software works.

There is no 1:1 correspondence between captured light and emitted light. You can simply - increase brightness on your computer screen or other viewing device - just enter settings of that device and fiddle with brightness / contrast. Also - different display devices have different brightness.

Process when capturing image up to showing it on screen is always the same, and can be simplified as this:

- camera captures certain number of electrons of signal in exposure

- those electrons get converted into ADU - by conversion factor we know as Gain / ISO - which is expressed in e/ADU units

- those ADU units get scaled to display units by using some conversion which is sometimes known as STF - or screen transfer function. Basic version of that is to set black and white point to appropriate values.

Now, if you always use the same physical units in this process - you will get equally bright image every time.

For example - if you convert electron count in your exposure to electrons per second instead of using electrons per 10 seconds or electrons per 30 seconds, and similarly if you use the same gain settings and if you set white and black point equally - you will get the same image.

On the other side of things - once you have captured certain number of photons / electrons - no amount of above math manipulation afterwards can change that and image stays the same - it is just emitted from the screen differently.

This is why we say that only thing that is really important is SNR.

Difference between 10, 20 and 30 second subs is not the brightness - as that is something you can adjust without changing the contents of the image (increasing brightness does not change the amount of noise for example) - difference is SNR - which you can understand as - if you adjust parameters to get equal output for all three images - 10 second one will be noisiest, 20 second one in between and 30 second one - the least noisy.

On the other hand, if you pay attention to read noise and swamp it with LP noise - then 10, 20 and 30 second subs stacked to same total time (Say 5 minutes worth of each) - will produce the same looking images if you adjust output properly. There will be no difference in noise for equally bright images.

[LDW1]

1 hour ago, vlaiv said:

This is often believed and it is so - but solely because of the way software works.

There is no 1:1 correspondence between captured light and emitted light. You can simply - increase brightness on your computer screen or other viewing device - just enter settings of that device and fiddle with brightness / contrast. Also - different display devices have different brightness.

Process when capturing image up to showing it on screen is always the same, and can be simplified as this:

- camera captures certain number of electrons of signal in exposure

- those electrons get converted into ADU - by conversion factor we know as Gain / ISO - which is expressed in e/ADU units

- those ADU units get scaled to display units by using some conversion which is sometimes known as STF - or screen transfer function. Basic version of that is to set black and white point to appropriate values.

Now, if you always use the same physical units in this process - you will get equally bright image every time.

For example - if you convert electron count in your exposure to electrons per second instead of using electrons per 10 seconds or electrons per 30 seconds, and similarly if you use the same gain settings and if you set white and black point equally - you will get the same image.

On the other side of things - once you have captured certain number of photons / electrons - no amount of above math manipulation afterwards can change that and image stays the same - it is just emitted from the screen differently.

This is why we say that only thing that is really important is SNR.

Difference between 10, 20 and 30 second subs is not the brightness - as that is something you can adjust without changing the contents of the image (increasing brightness does not change the amount of noise for example) - difference is SNR - which you can understand as - if you adjust parameters to get equal output for all three images - 10 second one will be noisiest, 20 second one in between and 30 second one - the least noisy.

On the other hand, if you pay attention to read noise and swamp it with LP noise - then 10, 20 and 30 second subs stacked to same total time (Say 5 minutes worth of each) - will produce the same looking images if you adjust output properly. There will be no difference in noise for equally bright images.

As a bit of an aside, what is the purpose of the Adjust +/- slider on the enhancing screen,  on the right side near the AF. Does / will it change the light / dark / contrast of the photos  end result ?  As you darken it you lose outer detail but its sharper, if you lighten it you get a broader detail but its very faded, does that carry through to the final product ?  Everything that I have read from ZWO about the SS I haven't found it mentioned unless I have missed it.

[vlaiv]

11 minutes ago, LDW1 said:

As a bit of an aside, what is the purpose of the Adjust +/- slider on the enhancing screen,  on the right side near the AF. Does / will it change the light / dark / contrast of the photos  end result ?  As you darken it you lose outer detail but its sharper, if you lighten it you get a broader detail but its very faded, does that carry through to the final product ?  Everything that I have read from ZWO about the SS I haven't found it mentioned unless I have missed it.

No idea. I was speaking in general when imaging, but have never worked with SS so I have no idea what each particular command does.

It could be either contrast enhancement - which is just adjustment of black / white point or it could be gamma setting - which is really a type of non linear stretch.

[LDW1]

23 minutes ago, vlaiv said:

No idea. I was speaking in general when imaging, but have never worked with SS so I have no idea what each particular command does.

It could be either contrast enhancement - which is just adjustment of black / white point or it could be gamma setting - which is really a type of non linear stretch.

Thanx, what strikes me is that they don't seem to mention its purpose very specifically, very clearly

[powerlord]

So, this chap says he has it working in EQ mode. He says since latest firmware it does 3 star calibration.

https://www.cloudynights.com/topic/904547-say-no-more-field-rotations-for-the-seestar-s50/

live stream last night which I hoped he would actually walk through the setup - but no - just a repeat of a (very unhelpful) picture.

I will say I tried it last night 5 times. And though I got it to pass calibration in then failed to goto anywhere properly (mainly tried M45 which was well within range). And manually going to M45 area showed weird movement looking like it was still running an ALT AZ word model

He does appear to be the sort of person who cannot actually explain anything well.. so hopefully there is just some trick to getting it to calibrate right.

Obviously it can't see all targets even if it did work due to limitation in what the DEC arm can do. I personally hold no time for the 'strain on gears" argument - its got bearings, etc - its not an issue imho.

Maybe now it's been tried others will try and someone who can actually properly explain the app steps to get this to actually work once you've pointed it at polaris - mine just kept going to somewhere close to where it would if it was in AZ mode.

 

[ObscuredView]

So I bought one and finally got clear sky 

To address some earlier negative comments, I had fun last night watching the image stack.

The image attached, was a one hour stack of IC434, Cropped, rotated and adjusted in OSX Photos.

Great fun and as it was 3 degrees below zero last night the fact I could do all this on my IPad sitting in the warm by the fire.

[Cosmic Geoff]

On 15/01/2024 at 08:00, powerlord said:

So, this chap says he has it working in EQ mode. He says since latest firmware it does 3 star calibration.

Since the latest upgrade, it does spend a couple of minutes doing something called horizon adjustment (whatever the heck that is) unless it is turned off in the settings.