Back to forums

RC8 Collimation using Innovations Foresight SkyWave

Equipment Ritchey-Chrétien Collimation
18 replies•820 views
Jérémie avatar
Jérémie avatar
Hi,

I will soon resume AP using my GSO RC8, that hasn't been used for 2 years and travelled a bit by car in between.

Last time I checked it wasn't collimated anymore. I doubt it will be when I'll get it out of retirement (never works that way…).

I will therefore need to collimate it again and I would like to know if anyone here has ever done it using the software SkyWave from Innovations Foresight ?

This software is used on high-end telescopes for collimation (Planewave CDK, high end RCs etc.) and more (I understand the core functionalities offered by this - these ? - software(s) are tracking,  focusing, and probably planning).

I have a Cheshire, lasers etc., but in the end you need to be under the stars and tweak everything "by eye", zooming on the stars on your screen… It seems the software allows you to "measure" the alignment of the mirrors, the uniformity of the astigmatism across the field, and even optimize the distance between mirrors (though I am not sure the GSO RC8 allows you to do so without screwing the first two tunings…). I like being able to "measure" things properly instead of relying on my untrained eye.

The cost of licence seems OK, though the difference between "pay per use" or "lifelong use" is not very clear (I didn't figure out if in the 'pay per use' option you still need to buy the weights of the neural net used to model the relationship between wavefront and defocused image of stars… or if you pay a model only when your telescope is not among the "standard" models already trained… and how it's related to the 'ppu' / 'lifelong' options…).

I was wondering if some people here had test it for small, low-end telescopes, such as the GSO RC8 ? Is it like using a hammer to kill a fly ? On the contrary, is it worth ? Does it also allows you to align the focuser with the primary mirror (in the case of the GSO RC8, the focuser is directly linked to the primary mirror, and may not be properly aligned : you have to buy and mount a tilt plate to adjust that).

Any feedback if welcome on your experience using SkyWave !
John Hayes avatar
John Hayes avatar
SkyWave (SKW) is a revolutionary method for improving the accuracy of optical alignment.  Full disclosure:  I worked with Gaston at IF! to write a paper about SKW that we gave at SPIE a couple of years ago.  I’ve used it to align my ASA600 RC scope and it is equally easy to use on any scope—including your 8” RC.  It is important to understand that aligning a RC is fundamentally more involved than aligning a SCT or CDK system.  You should first carefully geometrically align the system to insure that the secondary is well centered and tilted to retro-reflect a centered laser beam precisely down the mechanical axis of the scope.  SKW can then be used to adjust the tilt of the primary to minimize coma.  Then you have to tweak the tilt of the secondary to balance aberrations in the field.  There are numerous documents online that describe the process well.  SKW greatly reduces uncertainty in knowing when things are properly aligned so it speeds up the process quite a bit.  Furthermore it shows the quality of the wavefront when you are all done.  I give it a big thumbs up.

John
Well Written Helpful Engaging
Bill McLaughlin avatar
Bill McLaughlin avatar
Agreed. I have found, at least with my CDK (the only scope I have tried it with) that is has the effect of giving one confidence that you got it right. In my case I already had it right and was at 98% the first time I tired it after conventional collimation.  It was also useful when the mount "lost it's marbles" one night and cracked a truss tube. It allowed me to check remotely to see if the collimation had been messed up (remarkably it had not) as well as rechecking after replacing the broken part. Very useful
Helpful Engaging Supportive
Brian Valente avatar
Brian Valente avatar
Agreed that it is pretty much the gold standard for collimation, but a GSO RC8 is a pretty inexpensive bit of gear. iirc the skywave license would be close to half the price of it
Jérémie avatar
Jérémie avatar
Brian Valente:
Agreed that it is pretty much the gold standard for collimation, but a GSO RC8 is a pretty inexpensive bit of gear. iirc the skywave license would be close to half the price of it

I understand from their website that the « pay per use » model for a reflector of aperture up to 13 inches is priced at ~ 180 $, which is reasonable compared to the price of lasers like Howie glatter etc.. What’s unclear is if you need to add a fixed price for the software (they don’t seem to sell SKW alone, but in bundle with SkyGuard and Skyxxx, which would add ~ 120 $). All in all, to get th ebest performance of your telescope, that’s ok.
Now, I don’t know how many trials and errors you need to adjust everything (are the 64 image analysis sufficient to align the axis, then tweak the secondary ?). And I haven’t seen the price of the « refill ».
They really should put a calculator to make realistic simulations of how much it costs you, but I guess that high-end telescopes owners don’t really pay attention to that compared to the initial investment 😂.
Alex avatar
Alex avatar
SKW collimination is awesome software and best advantage overall is that it assists you collimation under real conditions.

many folks doing the collimation with scope flat on a table and then wondering why it’s not performing well under sky giving the fact that the different orientation (scope looking up) may have an impact especially on critical or fast systems. 

used it some time ago for an C11 edge hd and will soon buy a new licence for my 14“ Newtonian.

pay per use is absolutely ok. If I remember well used 5-6 credits, if even for a spot on collimation of the SCT and not expecting much more for the Newtonian.

so the included 64 credits will last quite a while especially with a permanent setup.

I hope SkyWave Wavefront Analyse tool will soon be available as well, as for years in beta stage. contacted Gaston 2-3 times about but unfortunately without response.
Helpful
Jérémie avatar
Jérémie avatar
John Hayes:
I worked with Gaston at IF! to write a paper about SKW that we gave at SPIE a couple of years ago.

I just read about it, very interesting use of neural networks (I happen to do research in computer vision).
It made me wonder why we haven’t seen yet (or at least I didn’t…) solutions of collimations using small step motors controlling the screws, to adjust the tilt of primary and secondary mirrors, using the realtime feedback of a camera (and adjusting the mount to follow the stars automatically and accordingly) ? 
One could train an agent, through deep reinforcement learning, to send commands of adjustment to both mirrors, using a proper reward function based on these pre trained models (defocused stars images to wavefront).
I am not a mecanical engineer, but it doesn’t seem that complex to motorize, nor adding too much weight (small step motors are lights), at least on mid size telescopes. And training such an agent wouldn’t take long given the low dimension of the action space.
Reading on forums the distress that collimation put people in, they probably would be willing to pay for such a device and software (of course, where would be the pleasure then ? But that’s another story…)
Helpful Insightful Engaging
John Hayes avatar
John Hayes avatar
Jérémie:
John Hayes:
I worked with Gaston at IF! to write a paper about SKW that we gave at SPIE a couple of years ago.

I just read about it, very interesting use of neural networks (I happen to do research in computer vision).
It made me wonder why we haven’t seen yet (or at least I didn’t…) solutions of collimations using small step motors controlling the screws, to adjust the tilt of primary and secondary mirrors, using the realtime feedback of a camera (and adjusting the mount to follow the stars automatically and accordingly) ? 
One could train an agent, through deep reinforcement learning, to send commands of adjustment to both mirrors, using a proper reward function based on these pre trained models (defocused stars images to wavefront).
I am not a mecanical engineer, but it doesn’t seem that complex to motorize, nor adding too much weight (small step motors are lights), at least on mid size telescopes. And training such an agent wouldn’t take long given the low dimension of the action space.
Reading on forums the distress that collimation put people in, they probably would be willing to pay for such a device and software (of course, where would be the pleasure then ? But that’s another story…)

ASA makes some very fast systems that include remotely controlled tilt adjustment on the secondary mirror.   These scopes could incorporated the kind of fully automated optical alignment that you are describing.  The complexity of such a system would be a down side and the benefit might not be that much more than what you can get from using SKW to do alignment remotely.
John
JohnAdastra avatar
JohnAdastra avatar
Offhand does anyone know what is happening at Innovations Foresight? I have two friends who have tried to place orders for parts and software and claim the order system on their website has been down for many weeks. Are they on vacation or something?

Thanks,

John
Well Written
Daniel Carter avatar
Daniel Carter avatar
Offhand does anyone know what is happening at Innovations Foresight? I have two friends who have tried to place orders for parts and software and claim the order system on their website has been down for many weeks. Are they on vacation or something?

Thanks,

John


I tried to order software from their site yesterday and had an error and wasn't able to proceed. I downloaded their software and emailed foe a trial license and received a response fairly quickly. Ill send an email to him today and ask about the payment processor.

*UPDATE*

I sent an email to Gaston with the error I was getting during checkout. Ill pass along word when any update is available.
Daniel Carter avatar
Daniel Carter avatar
I heard back from IF and they are having an issue with their payment processor. They are currently getting the problem fixed. As of 8-28, they mentioned that it could be several days before the problem is fully resolved.

Hopefully the payment processor will get the issue resolved quickly.
Well Written Helpful Respectful Concise
Gaston Baudat avatar
Gaston Baudat avatar
PayPal performed system updates that caused some issues, but the IF's store payment is now back online. Apologies for the delay and inconvenience.

Some information:
  • SkyWave CT (collimator) is available as a stand-alone version (named SkyWave or SKW) and is also integrated into SkyGuide and SkyGuard (SKG), but you do not need SKG to be able to use SKW. The stand-alone SKW CT is free, and you can self-register the license. It comes with a generic model available automatically after setting up an instrument for your telescope in SKW. This evaluation model is limited to coarse collimation (scores below 6) and helps users learn to use SKW, label the collimation screws, and practice with the collimation tool. For full collimation, users can purchase a model specific to their telescope.
  • Active optics using SKW is under development, with ongoing discussions with telescope manufacturers. The goal is to use actuators on the mirror(s) for remote, automated collimation. With SKW we could also automatically control the tilt of the camera using actuators too (see below).
  • The latest SKW version introduces two new features in the collimator tool. Both use a star field (defocused stars) to compute the wavefront and tilt, which is based on defocus aberration rather than star size. It also supports balanced astigmatism correction for RCT and similar telescope collimation. All of that using a single image (of a defocused star field).

    • More to come...
Helpful
Jérémie avatar
Jérémie avatar
@Gaston Baudat Merci beaucoup pour ces précisions !
Dark Matters Astrophotography avatar
Dark Matters Astrophotography avatar
Gaston Baudat:
PayPal performed system updates that caused some issues, but the IF's store payment is now back online. Apologies for the delay and inconvenience.

Some information:
  • SkyWave CT (collimator) is available as a stand-alone version (named SkyWave or SKW) and is also integrated into SkyGuide and SkyGuard (SKG), but you do not need SKG to be able to use SKW. The stand-alone SKW CT is free, and you can self-register the license. It comes with a generic model available automatically after setting up an instrument for your telescope in SKW. This evaluation model is limited to coarse collimation (scores below 6) and helps users learn to use SKW, label the collimation screws, and practice with the collimation tool. For full collimation, users can purchase a model specific to their telescope.
  • Active optics using SKW is under development, with ongoing discussions with telescope manufacturers. The goal is to use actuators on the mirror(s) for remote, automated collimation. With SKW we could also automatically control the tilt of the camera using actuators too (see below).
  • The latest SKW version introduces two new features in the collimator tool. Both use a star field (defocused stars) to compute the wavefront and tilt, which is based on defocus aberration rather than star size. It also supports balanced astigmatism correction for RCT and similar telescope collimation. All of that using a single image (of a defocused star field).

    • More to come...



Hey Gaston,

Just to clarify for my own understanding, SkyWave Collimator comes with the new tilt analysis feature now? If so that's huge and I'll need to upgrade my client!
Brian Valente avatar
Brian Valente avatar
For those of you using SkyWave with an osc camera, which wavelength do you use?

I'm thinking 520nm ish green

I could also insert a red filter in there, but I'm not sure how well that will perform vs. stock osc

Brian
Monty Chandler avatar
Monty Chandler avatar
I bought the license but found it difficult to collimate my scope in the dark.  Had much better luck using lasers in the garage.
Gaston Baudat avatar
Gaston Baudat avatar
Monty Chandler:
I bought the license but found it difficult to collimate my scope in the dark.  Had much better luck using lasers in the garage.

Could you tell me more about your scope and your collimation procedure with SKW CT?

Some recommendation for using SKW CT
  1. Defocus Star: Avoid any saturation. Keep the signal level at half to two-thirds of the full ADU range. Do not process the image in any way (raw monochrome image only), not even dark/calibration, SKW takes care of pre-processing including hot pixels.
  2. Star Selection: Ensure the star is not a double star and is not near other stars that could interfere. Use a red filter when doable, if so set the wavelength accordingly, 650nm for red.
  3. Exposure Time: Use an exposure time of at least 10 to 20 seconds to average out seeing conditions. For large apertures or poor seeing, consider longer exposures towards the end of collimation for fine adjustments. If tracking is an issue, shorter exposures are acceptable. Use the scatter plot option in the collimator tool to check the resulting cluster of scores. The center of the cluster will indicate the long exposure score, averaging out seeing-induced wavefront fluctuations.
  4. Defocus Error: Ensure the defocus star is within the SKW mathematical model tolerance. Check the vertical gauge (left of the view port) for defocus error relative to the model target. Aim for values close to zero or at least within the green region.
  5. Collimation Considerations: For most scopes (e.g., CDK, iDK, SCT), focus on coma aberration at the center of the frame. For RCTs, additional work is required as no coma on and off-axis does not guarantee collimation. Off-axis astigmatism must also be considered.

    6. Should you have any issues, questions, or need any support feel free to contact me at any time at gaston@innovatiosnforesight.com
Helpful
Gaston Baudat avatar
Gaston Baudat avatar
Brian Valente:
For those of you using SkyWave with an osc camera, which wavelength do you use?

I'm thinking 520nm ish green

I could also insert a red filter in there, but I'm not sure how well that will perform vs. stock osc

Brian

With an OSC, you should extract the red channel and set the wavelength to 650 nm. Using a red filter in front of the camera is also acceptable; just ensure the wavelength is set to 650 nm in SKW too, which is a suitable value for red filters. Seeing is reduced at longer wavelengths.
Well Written Helpful Insightful Concise
Gaston Baudat avatar
Gaston Baudat avatar
SKW now features OSC color channel extraction and also supports monochromatic TIFF (uncompressed), BMP, and PNG files.

SKW also supports image tilt measurement (even in the presence of field curvature) and balanced astigmatism calculation for RCT collimation, both using a star field.
Well Written Helpful Concise