Best guide camera for absolute sensitivity / QE?

[Stratis]

As topic really I have searched about this many times, even comparing QE curves as best as one can, but I've not been able to confidently select a guide camera capable of accurate, rapid corrections (0.25sec or less). Time was the SX Lodestar was the undisputed champ of guide sensitivity, but the CMOS era is more complicated it seems. I currently use an Altair GPCAM2 290M.

As the happy owner of an RST-135, the only complication of the mount is the necessity for rapid guiding exposures; the unofficial manual (authored by my mount's first owner, as it happens...) makes this clear and it's my experience that 0.5sec is as fast as I can achieve with the 290M / 30mm guidescope setup. As I plan switching to OAG and possibly longer focal lengths, I fear I will need a camera that can make better use of faint signal.

Are there any guide cameras that can be recommended on raw QE? Almost every other metric from expense to field of view to image quality to transfer rate I can work around... it's that 'ultimate' performance that I need.

Edited October 29, 2023 by Stratis

[vlaiv]

Don't know where to start with reply

Ok, let's address QE question first.

Compared to other parameters of sensor - it is really not that important. QE of modern sensors are very close to each other - they differ by 10-20% in peak performance. For example:

Simple ASI120mm has QE of 80%

ASI178mm has peak QE of 81% - so (81 - 80) / 80 = 1.25% increase

ASI290mm has peak QE of 80% - no difference

ASI462mm has peak QE of 89% - so (89 - 80) / 80 = 11.25% increase

ASI432mm has peak QE of 79% - so less than ASI120

The latest ASI220mm (I did not even know this one existed) - has peak QE of 92% so that is (92-80)/80 = 15% increase

So each sensor is within 15% QE of basic ASI120

However, let's compare some other metric - like pixel size since number of photons per exposure depends on both QE, but also on photon gathering surface - which is pixel area.

Let's compare ASI120 and say ASI290 with pixel size. One has 3.75um and other has 2.9um. That is 3.75 * 3.75 / 2.9 * 2.9 = x1.67 increase in sensitivity on account of pixel size alone. Much more than any difference in QE that we observed above.

When using OAG - you can bin your camera (and you probably should) to increase sensitivity. Sensitivity of guide camera with OAG vs small guide scope will be much much greater if you have reasonable resolution because large aperture creates smaller star image and larger aperture gathers more light - so star image will be much brighter.

Now onto very short exposures. That will simply have you chase the seeing. If your mount requires corrections every 250ms - then change the mount. Best mount is the one that is smooth and that has very slow varying error.  Best guide exposures are in range of 4 to 8 seconds if your mount can cope with that.

If mount needs to be corrected quicker - that just signals serious mechanical issues like mechanism roughness or some sort of very fast period error (I once had ~11s period error on my HEQ5, but that was due to issues with belt / pulley teeth meshing and I had issue on each tooth - which is roughly 11.4 seconds if I remember correctly).

Guiding should not attempt to solve mechanical issues - it is there to correct for periodic error (or declination drift due to inaccurate polar alignment).

Anyways, if you are guiding with faster guide cycle than a few seconds, in my opinion, you are doing it wrong.

To answer your question anyway, I'd say that ASI220 seems to be very good combination of factors for guiding. It has 4um pixel size which is good size for OAG (you might still need to bin if you guide scope with over one meter focal length), it has 91% QE and it has very low read noise of less than 1e (close to 0.6 and best gain setting).

[Elp]

HD mounts respond better to faster guiding but it doesn't necessarily mean the guiding is better (look at your subs as that is what matters). My hem responds to 0.5s guide exposures getting rms down to 0.6 or so, but it's pointless if the seeing is average (which is usually is) as it'll be correcting falsely to the seeing. Last session left mine on 1s exposures but usually guide at 2-3s.

[david_taurus83]

Is the "chasing the seeing" argument really relevant anymore considering that we use multi star guiding now?

[vlaiv]

7 minutes ago, david_taurus83 said:

Is the "chasing the seeing" argument really relevant anymore considering that we use multi star guiding now?

With guide scope, I'm guessing that it probably is less important than it used to be, but with OAG - it might be different story.

Don't know how much guide stars one can pick up with OAG as FOV is much smaller and also, I'm not sure how much seeing is varied across such a small FOV.

From planetary imaging and adaptive optics systems - we know that isoplanatic angle is not that large - so seeing disturbance seems to be the same over say 20ish arc seconds if I remember correctly. That is still much smaller than FOV of OAG even at very long focal length with small sensor (say you have ASI120 which has 1280 x 1024 and you use 2 meters of focal length so you end up at 0.4"/px - you still have more than 500 arc seconds across the sensor). However, I have no idea how much tilt component of wavefront error alone changes with angular distance.

I'm inclined to think that selection of stars on small FOV such as one gets with OAG won't have enough of diversity in star position to average them out, but I could be wrong.

In any case - more exposure, more average star position approaches true star position.

There is also "SNR approach" to guiding.

I've read once in some document a very sensible argument (and I tent to agree with it) that we could observe guiding in similar way to imaging - by examining signal to noise ratio. In this instance signal is actual mount error and respective correction and noise is well - noise, inaccuracy in both determining star position, but also in issued correction as no mount is perfect and will not respond accurately to guide pulse - there will always be some backlash, some inertia to overcome, some oscillations due to weight on the mount being moved and its inertia and so on ...

Conclusion of the paper is that we increase SNR in part by reducing number of corrections issued and that corresponds to long guide cycle. As long as mount does not accumulate significant error in that time (smooth mount) - long guide cycle is better than short because of this, even if we don't chase the seeing.

[ONIKKINEN]

I have had issues with chasing seeing with my 220MM in an OAG at 1018mm focal length f/5 scope even when there were multiple stars. Typically there are a good number of stars too, like 7 or 8 so multi-star guiding has not fully gotten rid of the issue.

Pushing exposures to 4s and above does solve that issue somewhat, but in the cases where this happened the seeing was just so bad that the improvement was of little value.

But on the topic of guide sensors, the 220MM seems like a really good guide camera. 4 micron pixels, which im running binned x2, high QE, low read noise. SNR of a guide star is always very high, in the thousands or at least hundreds.

[Stratis]

7 hours ago, vlaiv said:

Anyways, if you are guiding with faster guide cycle than a few seconds, in my opinion, you are doing it wrong.

Generally I would agree, but with the RST-135 in particular this seems to be a good thing. It has a very tight response and no backlash to speak of on either axis, but like most harmonic mounts does need to correct sometimes quite rapid drifts. 1sec is about as high as I like to go with it, 0.5sec improves matters if the seeing is decent, this is something many owners of the mount have noted and it's in the 'unofficial' manual for the mount.

Multi-star seems to mitigate chasing the seeing, and the subs are good at least at the shorter FLs I usually contend with. Part of my hopes in this upgrade plan are that I'll be able to make better use of my big scopes which tend to be relegated to visual work.

Quote

To answer your question anyway, I'd say that ASI220 seems to be very good combination of factors for guiding. It has 4um pixel size which is good size for OAG (you might still need to bin if you guide scope with over one meter focal length), it has 91% QE and it has very low read noise of less than 1e (close to 0.6 and best gain setting).

Thank you for this @vlaiv, I had no idea this camera even existed and even if I had, I'd likely have dismissed it due to my overall poor impression of its predecessor. There's a review on FLO from a fellow 290MM user that clearly states the 220M is a superior camera specifically for OAG use, so that's fantastic news for me.

Quote

But on the topic of guide sensors, the 220MM seems like a really good guide camera. 4 micron pixels, which im running binned x2, high QE, low read noise. SNR of a guide star is always very high, in the thousands or at least hundreds.

Thanks also @ONIKKINEN for the similar recommendation!

Can I ask which other guide cameras you've used in the past and how it might compare, especially in terms of faint star SNR in OAG usage? I've been working towards a truly portable but powerful imaging setup that can fit into hand luggage on a flight, and OAG/filter/rotator combo unit is a big part of that I'd rather not waste by falling back to a guide scope.

Also I've long held the dream of using an OAG to guide an SCT at f/6.3 or even f/10 instead of having to bolt an achro to the top rail with all that added mass. The 290MM works fine on a guide cam but haven't had the clear nights to test it on the OAG.

Quote

HD mounts respond better to faster guiding but it doesn't necessarily mean the guiding is better (look at your subs as that is what matters). My hem responds to 0.5s guide exposures getting rms down to 0.6 or so, but it's pointless if the seeing is average (which is usually is) as it'll be correcting falsely to the seeing. Last session left mine on 1s exposures but usually guide at 2-3s.

@Elp you're right in what you say. Part of the plan for a more sensitive camera is the use of a broad NIR filter to reduce seeing interference with the guider. At present the 290MM just isn't sensitive enough for both fast exposures and NIR.

Edited October 29, 2023 by Stratis

[ONIKKINEN]

28 minutes ago, Stratis said:

Can I ask which other guide cameras you've used in the past and how it might compare, especially in terms of faint star SNR in OAG usage? I've been working towards a truly portable but powerful imaging setup that can fit into hand luggage on a flight, and OAG/filter/rotator combo unit is a big part of that I'd rather not waste by falling back to a guide scope.

I used a 120MM before, i switched because it because i thought it developed an issue (seems more like USB issues in hindsight, cable probably), and because it wasn't quite good enough to guide trough an Antlia Triband filter with short guide exposures. With the 120MM i had some issues finding suitable guide stars trough the filter, and had to use longer exposures than ideal for my AZ-EQ6 and its rough RA worm with 2 sharp peaks per period which really needs less than 5s exposures to deal with. With the 220MM its been smooth sailing with the Triband or with UV/IR. Never rotated the camera in the OAG either, since there is no point with how every field of view has at least a few stars to guide on.

[windjammer]

I use a 120MM on a 90mm F10 guide scope, running 10-20ms exposures at 100ms intervals. (It is a homemade GEM so it needs extreme treatment to behave).  I don't really buy the 'following the seeing' argument.  Biggest problem is making sure the mount has responded before the next guide exposure - exposing while moving can lead to positive feedback problems (corrections get wilder and wilder!)

The 120MM/90mm F10 combo needs a guide star at magnitude 7 or brighter to get a good signal (5k to 10k counts) for PHD.  The 120MM sensor is quite small (1/3 inch diagonal) so the cam is on an XY stage so the sensor can move around the guide scope FoV to locate a guide star if need be.

I also have a 120MM on an OAG for the main OTA (150mm F5).  It needs 4-5 secs exposure to get a guide star but at this length exposure it will pick up a star in almost random field.  So given what you say about RST-135 guide frequency requirements the OAG might be difficult.  The point about rig portability is well taken.

I have just been thinking of a new guide cam coincidentally:  a larger sensor (hence greater FoV) and greater sensitivity would be nice.  I thought the ASI 432 was promising: 9u pixels, 1608x1104, sensor 14.5 x 9.95mm (1/1.1 inch diagonal).  Expensive though.  I will look at the 220MM - never heard of it either! The ZWO product line is a bit of a wonder...

Simon