A story of learning in the complex domain
Even to get to this I had been through many sprints of failed experiments. Learning so much along the way about what works and what really really doesn’t. Creating incremental value towards my vision.
My first telescope
What was I thinking! Biting off more that I could chew for my first sprint. Long running experiments and big mistakes are costly at this stage of learning. Time for smaller easier experiments. The mount was actually faulty. I didn’t know what I was doing so I had know idea that it would only move in Declination and limited Right Ascension.
Solving complex problems can feel almost impossibly frustrating sometimes. Its the passion to achieve your vision one small step at a time that gives you the energy to keep pursuing it. Mastering new technical skills that emerge through solving complex problems creates strong foundations and growth.
Vision: Enable people to Explore the Universe from the comfort of their own home
Product Goal: Just get something working
I love learning through experimentation. There’s nothing like failing fast and setting up your next experiment. When it works its pure glee. Every night was a new experiment.
I worked in one week sprints. Even if the weather is poor there always something to be done.
I swapped to a simpler faster lower focal length refractor telescope to use as a learning platform. Namely the William Optics FLT132 f7 giving 920mm focal length. Its basically a big zoom lens. The lens the latest iPhone for instance is f1.8 and 26mm focal length so very fast and wide angle.
For starters there setting up the mount and telescope. Level, balanced, polar aligned, aligned to multiple stars to build a navigation model for the mount.
Every telescope needs polar-aligning to align to the rotation of the earth. In my case the north celestial pole. Once aligned it makes it easier to find stuff in the night sky. This means learning a solid understanding of the earths rotation. Right ascension and declination act as coordinates for finding stuff in the night sky. The sky at night is constantly changing due to the Earths tilt of 23.5 degrees.
Product Goal: Photographing the night sky
At night our eye’s adjust to the dark using peripheral vision. Everything looks black and white and most of the detail in the night sky can be hidden from our poor human vision. I need to collect more photons! That means new experiments. Each image is constructed from at least 50 sub-frame images. This way the signal will out way the noise and provides lovely images. Lots more experiments to understand how this works.
Now that I had understood how to set up and operate the telescope and mount from a computer it was time to start experimenting with taking pictures.
Understanding how stuff works
Many sprints discovering astrophotography
- Discover what a flattener is. Fit camera and flattener to telescope at the correct distance. 1mm out in any direction results in image distortion.
- Understanding focal length and back focus distance
- Creating a flat image. The lens is curved but a cameras sensor is flat! That means correcting with additional lenses
- Exposure time and gain
- Dew, Dew, Dew – experimenting with heating. Space is cold so staring at it chills the lens. When that gets to below dew point you get dew. Dew is evil. Dew is bad.
- Guiding to make corrections to telescope position during long exposures. Basically using a second camera to track a star and correct the mounts position. The better the alignment and model the better the images
- Then there’s also understanding, creating and using flat, dark and bias images. As it happens camera sensors create noise. Not all the pixels work some pixels are hot and white and some are cold and black. Because the lens is curved light does not fall on the sensor evenly and is darker towards the edges. In a daytime image with lots of light you would not notice these issues. But when the signal from targets is very low then every photon counts.
Product Goal: Eliminate Annoying Issues
As it happens cameras sensors get hot during long exposures of 5 – 20 minutes. This creates noise. Noise means bad pictures especially in summer. High gain mean faster images but lost detail.
A cooled astro camera can cool the sensor to 35 degrees below the ambient temperature. There’s a learning curve here too. Experimenting with full well depth (yes, literally electron buckets) , gain, temperature, monochrome or colour, pixel size, dynamic range low noise, seeing conditions, arcsecs/pixel, pixel size enabled me to fully understand the capabilities and limitations of these complex cameras.
The scourge of light pollution
It’s a sad truth, but according to a study done by Italian and American scientists, one-third of the world’s population and 80% of Americans cannot see the Milky Way. The Milky Way is an awesome cosmic wonder to behold and to ponder. Even more experiments are need to remove gradients created by streetlighting in the UK. Most of the lighting is not required. With modern technology, LED lighting could be switched on only when somebody approaches. LED’s are way worse than the old streetlights. LED’s light produce light across the whole visible spectrum so difficult to filter out.
Next experiment using various filters to try and tackle the UK’s terrible light pollution and learn about narrowband filters.
The scourge of satellites
Say no more. Thanks to projects like StarLink and zero regulation of night skies, Earth based astronomy science is at risk forever. The only way to remove this is to use image processing to remove the noise.
Product Goal: Stay warm using Automation
I want to be warm. Winter is cold. At this point I need to be with my equipment to move targets and take sequences images of a target. Experimenting with automation using an industrial computer strapped to the telescope via Wifi or ethernet. No laptop in the garden all night. Full Remote Control 🙂
Product Goal: More Advanced Telescopes
Refractors have great resolution but if you want more zoom then a reflector is better. Basically this is a mini-Hubble. There are difficult to use and require more advanced collimation using a laser to line up two mirrors. This required going back to basics with experimenting. To get more advanced to quickly can feel like a backward step. Its a risk that delays value but can be necessary to achieve the vision. It took three months to master the Ritchey Chretien.
Product Goal: Professional Image processing with PixInsight
Using a professional software package for processing data is a daunting prospect. The results are great however. After a few hundred experiments I’m just kind of becoming a little more proficient and consistent. The image here was constructed from 240 monochrome Hydrogen Alpha and Oxygen sub frames
Product Goal: Think Bigger
During all my experimentation and getting creative I had become very proficient in setting up and consistently taking some great images. But everything was limited by light pollution, wrong kind air flow, wrong type of weather and humidity, seeing conditions which limits image resolution.
After much research moving equipment and setting up in a dark area in Spain or Chile at site used by astro professionals and amateurs alike was possibly the answer. Fregenal de la Sierra in Spain has excellent Bortle scale and everything I need. Its also closer than Chile if I need to visit. Chile is still on the list though.
It felt great setting up in Spain knowing exactly what I’m doing. Using best practices discovered through much exploration to my advantage. Roll off roof, dual telescope setup, with industrial computer control and remote access. 250 cloudless nights a year and very low light pollution and good seeing conditions.
Product Goal: Diagnosing issues remotely
Ah yes, something always goes wrong 🙂 When you can be there its easier to diagnose and confirm issues. Experience but more likely intuition comes into play in solving problem remotely. Autofocus slipping, filter wheel broken, camera moisture levels too high to name a few.
Creating my own software to analyze image quality in real time
With all that I had learnt I was able to start to solve a few automation problems that professional packages lack. I have had some great fun analysing the quality of images in real time including curvature and sensor tilt and correction during in sprinting towards this goal.
Experimenting creates waste! Not experimenting creates nothing good at all
Best advice. Keeping experiments small, goal focused and lightweight creates less waste and optimisation of value vs cost.
Product Goal: Automation for customers targets
Bringing it all together. These sprints were about enabling public access requests to professionally photograph targets of their choice delivering raw data and automated processed images. The images enclosed in the blog are much smaller than the real thing. A real single processed image may contain a gigabyte of HDR data. To reproduce these results starting from scratch could cost 10’s of thousands. With this service results are guaranteed for a fraction of the cost.
This is why I use Scrum Framework. Small iterations, small increments, maximising value, Learning as quickly as possible while optimising investment. Scrum enables accelerated learning. This Vision took a 18 months to bring to life!
I now have a viable business model where people can request a target and receive a fully automated dataset and a processed image of their target.
Next Vision: Different focal length telescopes and Southern Hemisphere
How will you bring your curiosity and creativity to play through experimentation to solve problems in the complex domain and have fun doing it? To learn more check out https://www.bagile.co.uk/our-courses/ or contact us to discuss agile and team coaching.