Dobsonian Mirror cooling

I got involved in a thread on CloudyNights about cooling mirrors. It started when someone had put their mirror in a water bath (half submerged!) as he had mistakenly let it get very warm during the day.

So there was a disucssion about whether water or air was better, the effect of blowing air, and finally the idea of blowing cooled air.

I jumped at the opportunity to do some real experiments. It turns out that blowing air on both sides of the mirror is about as effective as you are likely to get. The limiting factor then being the internal heat conductivity of the glass material. Here are some results.


I hadnt really bothered much with fans on my 18”, but I’m installing them now!

Then the discussion got serious about active cooling. In some climates this might well be necessary. I've located a low cost Peltier cooling module and will be repeating the experiments when it arrives. Watch this space!


Am I going to get involved in EEA? (Electronically Assisted Astronomy) I think I am already so might as well see where it goes.

The trigger was the amazing images I was getting from my eFinder with just a 1 second exposure. Here’s M13.

3 - 1 (3)

The eFinder will only be used when I have the 18” set up, and so I could usefully use it almost any time.

I had also realised that the top of my conservatory roof was a good mounting point. My All SKy Camera is up there and It can see most of the sky, especially to the south.

Enter my new toy. A Skywatcher Az-GTi. I knew they can be upgraded to work in EQ mode and it would certainly be able to carry the eFinder scope and camera. The first one delivered had a dodgy RA drive, the second had lots of Dec backlash but that was fixable. In taking it apart to adjust the backlash I was impressed with the internal design and build quality. Good value for money I think.

IMG 4274 (1)

Here it is in my workshop getting ready to go on the roof. It works seemlessly with Kstars, including plate-solving and goto refinement. Its connected to the Pi via wifi, but I’ve also got an EQDIR cable which works nicely.

I’ve discovered PiSynScan which is a fully functional Python module that allows direct control of the mount.

eFinder Update

Just returned from a brilliant week at Kelling Heath. 7 good nights out of 7 - amazing for the UK!

The eFinder got a good shake down and is now working extremely well. Achieved pointing accuracies with respect to actual sky (J2021.3) is 1-2 arc minutes.

Bentley Ousley in Kansas has contacted me as he was trying to make a similar finder but was stuck on the code. We worked together with me producng a new version for his ServoCat driven 20” New Moon scope, and he made me a 3D printed 72mm f2.7 finder scope.

By now I had also abandoned the Raspberry PI HQ ccd (pixels too small) in favour of an ASI120mm-S mono ccd. This ccd with the f2.7 guide scope produces excellent images with just 1 second exposures.

3 - 1 (1)

All Sky Camera

I’m pretty satisfied with how the camera set up is working now. It will run all night OK and I’ve sorted out some reflections I was getting off the dome.


eFinder update

The eFinder project progresses.

Had some clear skies for a couple of hours so tested the William Optics scope, cameras and plate solving routine. Success!

The Raspberry Pi HQ camera needs to be binned 2x2 or 4x4 to get good sensitivity but then it manages to capture about 20-30 stars in a 5 second exposure. The ZWOASI 120MM-S manages  the same in 2 seconds.

Here’s a shot of the WO scope, RPi HQ camera and RPI

IMG 4060

and here’s the new hand controller. It has the ScopeDog functions (on the right) plus the new eFinder on centre and left. Its currently displaying the calculated delta in Azimuth and Altitude from solved position to target position, in arcmin.

IMG 4061


electronic Finder project

Browsing on Cloudynights (‘CN’) I came across someone who had made an electronic viewfinder, using a CCD, Raspberry PI and small guide scope. The clever bit was to use plate-solving to workout where the scope was pointing.

Now I had just about all these parts going spare, what I didnt have was enough built in knowledge of Python and the inner workings of the Raspberry Pi (RPi) operating system. Definitely a Covid ‘lockdown’ project!

I’ll start a full description on a fresh page in my ‘equipment’ section, but here’s a brief synopsis of progress so far.

Built protoype with…...

SVBony 50mm guidescope (now replaced with William Optics)
Raspberry Pi HQ ccd (ZWO ASI 120MM-S on hand if more sensitivity needed)
Raspberry Pi4 running Astroberry/Kstars/Ekos
Small 16x2 line red character display with buttons.

About 400 lines of Python writen so far to read the Nexus DSC, capture images, plate-solve them, and convert the result into the necessary scope adjustment. Yet to do is send the scope adjustment to ScopeDog (telescope controller)

Its all tested in the workshop (using library star field images) and now I need a clear night to test the capture and plate solve on real star images.

Finally had to buy a Windows PC!

I’m a committed Mac & iOS user. I like the seamless integration and generally they look after themselves by way of updates etc.

Getting astronomy software for the Mac can be difficult. Up until now I have managed OK but I really like the AllSkEye  software for driving my All Sky Camera. It has lots of features and even includes a meteor detection option. Its Windows only though.

Looking around I was attracted to the mini-PCs now available, especially on Amazon who frequently have promotions on certain models. They’re Chinese which I didnt want to buy, but then a lot of my Apple products are made there too. I took the plunge and bought an AWOW mini-PC for £179 inc. a 128GB SSD & Windows 10 home. Delivered next day and set up within an hour. Its amazing! I access it via VNC Viewer from my Mac and so it doesnt need a display/mouse/keyboard. It runs off 12V so can easily go on field trips.

Screenshot 2020-12-11 at 16.07.58

All Sky Camera continues….

Had a few nights to test the camera. All is Ok except…

Condensation on the inside of the acrylic dome.

Investigation showed rain and dew was getting in past some of my ‘seals’. Easily fixed with more attention and some silicone sealant. I noticed that the low power dew heater setting doesnt seem to be effective. Its only about 0.6W which is much less than the camera and RPi, so not worth it. Instead I installed a small fan and used the now redundant relay to switch it. Also Installed a desiccant capsule.

The Indigo Sky server stops running on the Raspberry Pi after about 4 hours.

No idea yet while this is so, Much more investigation needed. The Indigo_deadlock_detector.txt file is empty, so no help there. I suspect its down somehow to the QHYCCD camera with its known dodgy firmware.

All Sky Camera

I’ve had an old QHY5L-ii-M CCD lurking in a box for years. Having spent too many evenings at Kelling Heath this year waiting for clouds to disperse (they didnt!), I thought I should make an ‘all sky camera’ so I could view the sky from the warmth, and I’d like to capture meteors with it too.

On digging out the camera, I was amazed to find I also had a suitable lens. So all I needed was a box, acylic dome, dew heater and a long USB cable. Or so I thought.

Some quick testing of the camera reminded me that it is a temperamental beast with poor drivers. The most reliable drivers appear to be the ASCOM variants, but even these weren't stable enough with the applications I wanted to run.

Mainly I wanted to use AllSkEye which is purpose written for whole sky imaging, capturing and processing images all night if needed, and even featuring a meteor detector!

This wouldnt work with the QHY5L for more than a few minutes before crashing. The AllSkEye creator (Mike Poelzl) responded to emails and suggested I use a Raspberry PI (RPi) as an Indigo server out in the camera box.

Indigo is like Ascom in some respects, but clearly aimed at controlling the many parts of an astrophotography set up. Running on the RPi I could connect to it via wifi, thereby eliminating the long USB cable requirement.

I had a spare RPi and it didnt take long to set it up with the Indigo Sky server software. A bit more learning and I was able to access the camera very nicely from AllSkEye and other applications running on my desktop Mac. Very impressed with the Indigo suite.

I went a bit further and added a relay board so that I can control the dew heater (off - 5V - 12V) as it does heat the camera box up, and control a homebrew motorised focus.

Here’s the finished unit

IMG 3978

Here’s the inside

IMG 3977

and here’s a image from my first night, complete with meteor!


2020 Autumn Kelling

What a bad week, from a weather point of view. Really only one good night (Tuesday). Weather forecasts were all over the place and often gave false hope!

Good to see a large turn out though with perhaps 75% of pitches taken.

The rebuilt scope worked very well - much stiffer in winds. Pointing and tracking accuracy was excellent too. It was much easier to pack into the car, and I had room this time for my wife to come along. The new observing steps/chair was brilliant. Being comfortable while observing is so important - allows longer and more relaxed views of objects, which leads to better acuity and sensitivity.

I've taken to hiring a caravan recently. Makes the event so much more pleasurable, especially when cold and wet. I have it delivered and collected from the site. (Mustangs cant tow!)

Lets hope Winterfest in November is better. (edit - it was cancelled due to Covid lockdown, and it was cloudy all week as well)

IMG 3912 (1)

© AstroKeith 2021