Changing Guide Camera Gain whilst Guiding

[Martin Dowd]

Maybe I haven't read deep enough into the PHD2 manual, however I have the following questions in relation to changing the Guide Camera Gain settings during guiding -

1/ Would this affect the use and benefit of my Dark Library as the Library was created under one gain setting at different exposures ?

2/ If the answer to question 1 was yes , would I have to create a new Dark Library with multiple  or different gain settings and multiple exposures ? This seems almost impossible as the combinations of Gain settings and Dark exposures could end up in the hundreds.

What prompted to ask the question,  I was imaging one night under a full  moon ( at least 60 degrees away from moon ) and the Guide Star SNR was quite low with Star Profile unsatisfactory. Guiding was average.  I decided to drop the Gain setting in PHD2 ( Brain , Camera tab ) on my ZWOASI120MM ( Native driver ) from 95 down to 85 and the Guide Star SNR improved significantly with a good Star profile. Guiding improved slightly as well.

So under the above scenario changing gain settings on the guide camera did have a positive outcome.

I just wanted to know if the Dark Librqary was affected or rendered useless ??

Maybe Ive answered my own question due to the positive outcome ?

For Reference my set up is a follows -

6' f6 GSO Newt

EQ6 - R pro mount

Imaging camera ZWO2600MC

Orion 60mm Guide Scope

ZWO120MM guide camera

Ascom pulse guiding PHD2

I have another set up in a different location using a 10' Newt and ZWO290MM guide camera

Await some advice

Thanks

Martin

[bw_m...@earthlink.net]

Making frequent on-the-fly changes to gain isn’t really a good idea for guiding.  It does affect the dark library and any bad-pix maps because the noise levels in the guider images will change.  To predict how much it affects the dark library, I guess you would have to know exactly how the camera behaves. My impression is that most of the camera vendors have a recommended gain setting, probably one that optimizes signal-to-noise,  and that’s probably what you should use.  Perhaps you should run a set of experiments and find a gain setting that you can stick with. 

 

Regards,

Bruce

--
You received this message because you are subscribed to the Google Groups "Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guidi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/90f14366-e63b-480a-a08a-affacb432290n%40googlegroups.com.

[Brian Valente]

Adding to, but Bruce said, from the specs, it looks like a gain of 29 is the ideal balance of noise and signal.

>>> ZWOASI120MM ( Native driver ) from 95 down to 85 and the Guide Star SNR improved significantly with a good Star profile.

Looking at the specs again, there really isn’t much difference between 85 and 95, that’s a pretty small change so I wonder if that’s really what caused any improvements

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/004501d95c66%24dc8b8150%2495a283f0%24%40earthlink.net.

--

Brian 

Brian Valente

astro portfolio https://www.brianvalentephotography.com/astrophotography/

portfolio brianvalentephotography.com 

astrobin https://www.astrobin.com/users/bvalente/

[Martin Dowd]

Hi Bruce,

Thanks

I guess testing different Gain settings on the same night and same target is the best solution.

Martin

[Martin Dowd]

Hi Brian,

Thanks for your reply

Obviously good advice from Bruce

I have been using PHD2 for 6 years now but never really fiddled with the camera gain settings on the fly , until the other night

But in saying that I did originally briefly test the 120MM when I first bought it years ago

The ZWO120MM is used at my City suburban location under Bortle 8 Skyglow

I tried Gain 29 ( Unity ) on this camera with poor results

The ZWO specs ( graph ) on this camera has a gain range from 0 to 100 , so therefore the PHD2 Gain parameter ( Camera tab ) is a percentage of that range , so ZWO Gain 29 = PHD2  29%

Ive found PHD2 Gain 85% or 95% to be a much better option and therefore did stick with that setting

Maybe I should try Gain 29 again ?? But that would mean deleting my existing Darks and taking new Darks , correct ?

I use a ZWO290MM on my 10" cabon fibre Newt at my dark site Bortle 3 Sky ( bought the rig last year )

The ZWO specs ( graph ) on this camera has a gain range from 0 to 350 with Unity gain being Gain 110

Ive found PHD2 Gain 60% or 55% , ie ZWO Gain 210 and 192 to be the better option.

Just to confirm I am using the correct logic in regard to the PHD2 Gain parameter in the Camera tab ( ie: percentage of gain range of camera )

Martin

[bw_m...@earthlink.net]

Yes, your interpretation of the ‘gain’ setting in PHD2 is correct – a percentage of the camera-defined range from min to max.  When you are judging what’s “best”, you should be looking at the SNR of the guide camera frames, not the apparent brightness of stars.  If you’re doing testing, you might want to save the guide cam frames and look at the statistics using whatever imaging tool you like.  You could do the testing without using a dark library, then rebuild the library when you’ve decided on the gain setting you like.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/fa08c86a-e8eb-458f-a447-16a65fac6036n%40googlegroups.com.

[Dale Ghent]

Purely a sidebar on this topic...

I've been curious about the origin behind using a percentage of the gain range to specify, kind of indirectly, the gain to run the camera at. I get the feeling that this might be a hold-over from CCDs and, perhaps, CCDs from a particular vendor, many years ago in PHD2's history, where gain ranges were more or less consistent and you didn't need to have advanced knowledge of the gain to know where a percentage would put you on the scale. This is also kind of a head-scratcher these days because the camera's advertised range isn't presented anywhere in PHD2 after connecting to it. I'd say that users generally have no knowledge of this range unless they first use the camera in a separate app which makes this range apparent. Only then they are able to do the math and know which gain value the percentage they specify corresponds to.

We know now that guide camera from vendor to vendor and even model to model can have widely varying gain ranges. On top of that, guide cams of the last ~5 years have started using dual gain domain sensors, which where specific gain values switch the sensor from one domain to another, which in turn affects noise levels and dynamic range in the resulting images. Knowing a particular specific gain level to operate at is more critical, I think, than what it used to be when "somewhere in the vicinity of unity" was good enough.

In addition to this, there is offset (preventing clipped pixels on the low end) and in some cases, even a readout mode selection on some of QHY's newer guide/planetary cameras.

Has it been considered to allow a more direct prescribing of camera gain (and perhaps the ancillary knobs mentioned above) instead of the flat percentage? I would be happy to help with the camera driver side of things if a new gain control design could be pondered up.



> On Mar 22, 2023, at 00:40, bw_m...@earthlink.net wrote:
>
> <!-- /* Font Definitions */ @font-face {font-family:"Cambria Math"; panose-1:2 4 5 3 5 4 6 3 2 4;} @font-face {font-family:Calibri; panose-1:2 15 5 2 2 2 4 3 2 4;} @font-face {font-family:"Helvetica Neue"; panose-1:0 0 0 0 0 0 0 0 0 0;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {margin:0in; font-size:11.0pt; font-family:"Calibri",sans-serif;} a:link, span.MsoHyperlink {mso-style-priority:99; color:blue; text-decoration:underline;} span.EmailStyle20 {mso-style-type:personal-reply; font-family:"Calibri",sans-serif; color:windowtext;} .MsoChpDefault {mso-style-type:export-only; font-family:"Calibri",sans-serif; mso-ligatures:standardcontextual;} @page WordSection1 {size:8.5in 11.0in; margin:1.0in 1.0in 1.0in 1.0in;} div.WordSection1 {page:WordSection1;} -->Yes, your interpretation of the ‘gain’ setting in PHD2 is correct – a percentage of the camera-defined range from min to max. When you are judging what’s “best”, you should be looking at the SNR of the guide camera frames, not the apparent brightness of stars. If you’re doing testing, you might want to save the guide cam frames and look at the statistics using whatever imaging tool you like. You could do the testing without using a dark library, then rebuild the library when you’ve decided on the gain setting you like.

> To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/006701d95c78%2479a90050%246cfb00f0%24%40earthlink.net.

[Brian Valente]

Dale

as a purely sidebar to your sidebar ;) while I tend to agree with you that sensors have come a long way and that more directly controlling gain can ease setting exact gain values, I also want to caution about the practical guiding results. 

In this specific example the op suggested that changing a gain of 10% could have been the source of dramatically improved guiding results. Looking at the camera specifics this was not a dual gain sensor and not near the unity gain value (presumably the best snr value proposition). So while setting an exact gain is an interesting notion, in this specific case i doubt a more precise setting of gain or even changing the gain +/- 10% would have accomplished much (and I don't think you are proposing that, this is mostly to clarify)

In other words, i would hate to see users concluce "if I could only set my gain exactly, i would solve my problems" - I don't think that's the case, at least here.

Also I assume you are referring to just the native driver implementation? Many ascom drivers (ZWO among them) allow for precise setting of gain via their ascom camera settings.

Brian

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/D3C92AC8-D02C-4894-B7C2-36ED7E64F182%40elemental.org.

[bw_m...@earthlink.net]

Hi Dale. Generally, we want to keep the device interfaces as abstract as possible unless there is a compelling reason to get into device-specific capabilities. The recent post about gain is probably not very representative of field experience and we have gotten very few user problems that came from people getting balled up with gain settings. Getting even further into adding device-specific knobs and doo-dads is not appealing at all - the UI becomes more complicated and cluttered, we have to document all of the doo-dads, we have to deal with all the ongoing version changes to each camera interface, and we end up having to support all the misuse that comes from people diddling around with settings they don't understand. My impression is that little or none of this is important to guiding because all we need is to get usable camera exposures in the range of 1-4 seconds with an adequate SNR to find distinguishable objects.

Coming back to your point about at least supporting absolute vs. relative gain settings, that could certainly be done. It would involve changes to every camera interface, invention of one or more synthetic properties to advertise the capabilities of the driver, a forward migration of gain settings on installed systems, and then a permanent distinction between relative vs. absolute values. None of that is rocket science, it's more a matter of tedium and time. It strikes me that the work should have lower priority than other changes that are scheduled - changes that are more directly related to delivering better guiding results, new capabilities, and/or reduced user confusion. But my perspective is influenced by an assumption that setting a gain value should be mostly a one-and-done operation for any given configuration. Is this not the case?

Thanks for the input, I'd like to know where you disagree with what I've said,
Bruce

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/D3C92AC8-D02C-4894-B7C2-36ED7E64F182%40elemental.org.