Home made ONAG (On-Axis Guider)

[Sevo]

I`m starting this topic, as a DIY On-Axis Guider, inspired from Poor Man's On-Axis Guider.
But I have decided to step a little bit further and change the essence of the usage of a Cold Mirror substrat with a Hot Mirror one. The difference is that the cold mirror reflects the visible light, and passes the infrared, but Cold Mirror is totally opposite. It reflects the IR and passes visible. I simply didn't like the idea, the main imaging camera to use the reflected light and to be located 90 degrees from the main beam. I have searched the internet and found one excellent Hot mirror substat, from a UK based producer. Quickly made a purchase for a 50mmx50mm sample and started the design phase. 50x50, allows full-frame cameras to be used and the light-path (from adapter to adapter is minimal (56,8mm)
I have a nice 3D printer and some “Autodesk” skills, so I have ended up with the design on the attached pictures. I also decided to use Starlight Xpress standard adapters for the telescope and camera side and a standard T-2 thread for the guider. I also tried to keep the plates as thin as it can, to manage the lightpath short. Printed the 3 parts, assembled the set and...here what I have smile.gif
Now I have to wait for some clear skies, to give it a real live test.
I count on your comments behind the fact that I`m just a “small” (around 100kg) astro-amateur from Bulgaria and I was inspired only from pure DIY without any commercial intentions 

 

N.B 

The blue color of the mirror is because there is a removable protective coating, that will be removed before initial usage.

The glass is completely transparent as seen on the diagram bellow. 

Edited March 8, 2020 by Sevo

[michael8554]

Very impressive work Sevo. 

It may be that if you image through glass at 45 degrees you will have distortion, which may be why they decided to image the reflection instead?

I hope it's not so. 

Michael 

[Sevo]

As the weather in Ireland is awful these days (or month(s) ), for astronomy observations, I have decided to upgrade the body a little bit more and produce the second version of the guider.
The idea was to be able to move the guider camera across the whole field of view more precisely and ideally with just one hand. Additionally, I have used a precise low-profile focuser, with around 5mm travel, enough for good focus on the guider camera side.
The result is in the pictures.
The single drawback is that the back-focus is increased to 119.3mm in total. The astigmatism between the guider and imaging side is 60.7mm
But the result is quite good (I think) . Share me your thoughts, I’m appreciating your ideas.

 

P.S.
As I won’t use the version 1 of the body, I can send the set (at the cost of the materials) to someone who wants to give it a test. Contact me in the private messages section.

 

 

 

 

[billdan]

I just came across this thread. You have done a professional looking and well thought out project with that ONAG (thank God for 3D printers).

Is it necessary to have a movable guide cam? Just fixed in the centre would take care 90% of good guide-stars. Simply fine tune the RA/DEC controls and adjust the main imaging camera framing to ensure a decent star is near the centre of the FOV.

Please keep us informed with your testing, as I wouldn't mind having a go, if it works OK -- ( I have a 3D printer).

 

 

 

Edited May 6, 2020 by billdan

[Sevo]

The guide-cam movement and adjustment is the most essential part of this guider. Usually the guide cam has smaller chip than the main imaging cam and it wont be able to see the same FOV as the main cam. This leads that less stars will be available to the FOV of the guide cam. That`s why you need to move it across the FOV of the imaging cam and find the brightest star. This approach is especially suitable for Adaptive Optics (Starlight Xpress ...) which OAG not always have a suitable guide star outside of the FOV of the main imaging camera.

 

[andrew s]

Have you tried it yet? With my ONAG the transmitted beam definitely suffers from astigmatism does your?

Regards Andrew 

[Sevo]

No I didn`t.

Sold my scope  and now can`t. The new Esprit 150 will be here in early June. Hope that COVID-19 lock-down to be over till then.

I`ll gladly send the first or second version to anyone that want to give it a try. I`m already developing V.3, that will be hybrid (manual or motorized )

All that you need are the Starlight Xpress adapters and the Hot Mirror.

Contact me in the private section if you want it  shipped.

 

Sevo

[eric_astro]

I'm looking for a cold mirror. they all seem to have 2-6 lambda surface flatness unless you spend $3000 for 1/10 lambda.

I've tried Edmund Optics and Thor Labs.

Anywhere that would have something for a reasonable price? 

[Sevo]

Mine Hot Mirror is from Edmund Optics and is very well build.

BTW I have tested the Hot Mirror and it has astigmatism. Not so dramatic, but visible. That`s why the use of Cold Mirror is in the production.

Now I`m changing the design of my ONAG with Cold Mirror plate and will post the result when available.

[eric_astro]

Looking forward to your test results. I have a Meade flip-mirror that I will modify. It's odd that they spec the transmitted beam wavefront, but not the reflected beam.

The dichroic from Edmund Optics (#69-903) is 1/4 wave for the transmitted beam, and no spec for the reflected beam.

They also don't provide the reflection/transmission curves past 535 nm. So even the tech didn't know how far into the blue it works.

In the end, the EO tech suggested a pellicle - great transmitted beam, terrible reflected beam. But have to give up light to the guider.

[WatchObservatory]

I built one this summer (2021) and it works fine. You have to use a cold mirror because there will be internal reflections through transmission causing one or more ghost images, unless you pay lots of money for special anti reflection coatings. As for wavefront distortions, I got a science grade 80x50mm from Alluxa for 550$ USD delivered (2020 Oct) to Canada with 1/12 wave accuracy. Only down side is that it is 1mm thick, would have preferred more for strength. Here's my contraption (link below) - its going through trials, but main camera is flat across the entire field, guide camera doesn't show ghosts (but it is there) due to my DIY thresholding image star centroid app. Wit a 0.5x, the guide camera has the same field as the main, and very close to the same arcsec per pixel, no need to have an X-Y adjustment. 

I would not have used 3D printing because imaging train is heavy and I can't trust such on an 8K$ camera over time, at -25C, etc.

http://watchobs.com/astronomy9.html

Cheers

Edited September 26, 2021 by WatchObservatory