I think it is high time for an update in case anyone is wondering whatever happened to this project.
Well, serial number 001 has been capturing photons in Tom's automated observatory for over a year now and I've been getting a lot of feedback that has enabled me to finalize the design.
The thermal control worked well in all seasons and no observing time was lost due to mirrors getting fogged up, but one of the fans failed after about a year and I replaced all four at that stage. The chrome plated fan grilles started to show signs of rusting after a year and SN002 will get 3D printed ABS grilles.
The secondary focuser has been generating lots of nice V-curves every clear night and caused no trouble, but Tom said that it took a bit of time to get the autofocus parameters optimised because existing software was written to work best with very small step sizes. My secondary focuser has a step size of about 12 microns and that means only 2 or 3 steps will fall in the critical focus zone. In theory that should be enough and it did turn out to be so once the software parameters were right.
SN002 will use a conventional rack and pinion focuser but not because of the issues with the step size. The main reason is that I want to free myself from software development. Simply not enough hours in a day. When I wrote the Arduino software for SN001 I used the same pins for the motor drive and the ambient temperature sensor as Rod Brown's open source project, so that his focuser driver could talk to my shield and make it ASCOM compatible. This was going to be just a temporary fix until I got round to writing my own driver. Since then I decided that I can't afford the time needed. Software also needs updating from time to time, to maintain compatibility with the evolving ASCOM platform. By using an off the shelf controller, such as the AEF, my software problems should be solved. However I will have to design a new smaller PCB, based on Arduino Nano, for the thermal control. That should be no problem as it is a stand alone unit that does not need to interact with other software.
Last but not least is what I learned about the optics. While SN000 had a Pyrex primary and Borofloat secondary, SN001 had both mirrors made of Schott Borofloat glass. I had a lot of trouble making this optical set. It was very difficult to get consistent readings during the figuring process and in the end the Roddier test came in at about 0.9 Strehl. That is a bit better that what is called diffraction limited, but in practice, for a high resolution instrument, that means somewhat larger stars than theoretically possible in a given seeing. So I made another Borofloat set and I got the same result. In the end I figured out what the problem is with what I now call Gremlin Glass. When examined between crossed polarizers, there is no sign of internal stress, as long as one only looks trough the disk in an axial direction. When examined from the side, there are very strong parallel stress layers that cause problems once one cuts a deep curve into then.
So in the end I made a mirror set using old Pyrex stock that came in at Strehl ratio of about 0.97 and installed it in SN001 about a month ago. The weather has been bad but Tom is reporting good performance and he is getting closer to producing a first image with the new mirror set.
Meanwhile I'm working hard at making an exorcising kiln for Gremlin Glass.
The attached photo shows the new focuser attached to SN000 and the new fan grilles.
Sounds you're making great progress gathering loads of data. This stuff is extremely time consuming and involves a lot of man hours to make incremental changes for testing again and again. Kudos for sticking at it.
Thanks Marc,
Yes, it's been a long journey, but I think I am getting close to having a competitive product. Borofloat glass should make very good mirror blanks after going through my annealing kiln, so the supply of mirror blanks will be resolved soon. I also managed to secure a good supply of Schott BK7 glass for the corrective lenses.
Thanks Marc,
Yes, it's been a long journey, but I think I am getting close to having a competitive product. Borofloat glass should make very good mirror blanks after going through my annealing kiln, so the supply of mirror blanks will be resolved soon. I also managed to secure a good supply of Schott BK7 glass for the corrective lenses.
It's a shame you couldn't source a reliable supply of glass. You end up doing things you could have done without. Although you'll have more control over the quality of the final product. No surprises.
Thanks for the update, Stefan.
How do you go with maintaining secondary to primary mirror spacing on the CDK, when they are suppose to have a Fixed exact distance? I know there is about 1mm tolerance, is this distance what you use to focus with?
It's a shame you couldn't source a reliable supply of glass. You end up doing things you could have done without. Although you'll have more control over the quality of the final product. No surprises.
Schott Borofloat is technical glass that is not intended to be used for telescope mirrors even though it is a high quality version of Pyrex. The internal stress comes from the manufacturing process - the glass is cooled a bit too fast on the production line, as it is a continuous casting process. People reported making good Newtonian mirrors out of it and it is available at a good price.
Maybe the shallower curve of an f/4 mirror doesn't free up as much of internal stress as my f/2.7 primary.
Quote:
Originally Posted by Joshua Bunn
Thanks for the update, Stefan.
How do you go with maintaining secondary to primary mirror spacing on the CDK, when they are suppose to have a Fixed exact distance? I know there is about 1mm tolerance, is this distance what you use to focus with?
When I first set up the optics, I place a Ronchi screen at the focal plane, which is at a fixed distance from the backplate, and point the scope at a bright star. Then it is just a matter of adjusting the secondary collimation screws until the focus is on the screen. One of the collimation screws has a bit of Loctite on it and requires a flat screwdriver to be adjusted. Once the spacing is close, then the collimation is adjusted with the other two remaining screws that don't require any tools. This procedure is done with the secondary focuser set at the midway position. The focuser has a range of plus minus 1mm, which equates to about plus or minus 5mm at the focal plane. Raytracing analysis shows that the focal plane position tolerance is about plus minus 1mm.
So your colimation screws have a grip on the secondary backplate rather than just pushing on it, allowingboth spacing adjustment and push/pull collimation adjustment?
Regards Joshua
So your colimation screws have a grip on the secondary backplate rather than just pushing on it, allowingboth spacing adjustment and push/pull collimation adjustment?
Regards Joshua
There are two backplates with a nominal 1mm gap between them. One plate caries the secondary mirror and the secondary baffle, and the other one is attached to the linear bearings of the focusing mechanism. The second plate carries the adjusting push screws, the secondary heater and the temperature sensor. There are three spring loaded pull screws that go through the second plate and engage the first one. The 1mm gap can be changed with the collimation screws for tweaking mirror separation.
In the new version, with RP focuser, the second plate will be bolted to the spider hub with 2mm spacers to simulate the midrange of the no longer used secondary focusing mechanism.
Quote:
Originally Posted by TrevorW
Such as it is why not forego making a focuser thereby people have the option to attach their own preferred focuser
I will be able to offer a cheaper version OTA with no focuser, for those who want to use their own, as well as the full plug and play version.
Hi Stefan,
I just found and devoured this thread- what an amazing piece of work.
Please add me to your list of potential buyers. A Beamtech 250 CDK would replace my Celestron Ultima 9.25 almost as a drop-in solution!
What camera would you recommend for this astrograph?
Again, amazing work.
Neville
Lonely Spoon Observatory
Hi Neville,
Thanks for the positive feedback. The potential list of buyers will have to be short as I intend to produce only a couple per year at this stage. The ideal pixel size for this f-ratio is about 5 microns. I personally would recommend an APSc size sensor for deep sky imaging (galaxies and planetary nebulae) and a full size one for larger objects.
Quote:
Originally Posted by strongmanmike
Really nice looking bit of kit Stefan,...yeeees I can see that at 1450m AMSL
Do you have a web site yet?
Mike
Thanks Mike for the "thumbs up". 1450mm FL at 1450m AMSL- interesting coincidence.
I started building a web site but it is not fully functional yet. I want to add a few more products before I enable the commercial side of it. If you google beamtech precision astro gear, you should be able to find it. I don't want to publish the link here, as I may breach IIS commercial rules.
Hi Stefan. Good to be aboard the Beamtech world with my SN0003 CDK250 that arrived yesterday. It looks great at first sight when unpacked.
Of course the couriers decided to have it arrive right at the start of a weekend when I had to go away from home, and with a forecast for the next week of all cloudy nights here at my location in New Zealand, as usual with any new astro gear , but I'm really looking forward to setting it up on my 10Micron mount on Sunday when I get home, and connecting my EAF, Camera, etc ready for when I do get a clear night to check collimation after the delivery journey, build and save a sky model for it in the mount, and try some first light imaging.
Thanks for your effort in packing it so well for transport.
I can see that at 1450m AMSL 
