Forum: Virtual Classroom - ThumbScope

Log In via ThumbSat
via Facebook
Not Registered? Register
Log In
8 + 3 =


I've been very inspired by various "travelscope" telescopes such as this one , and I've been thinking about them for a while.

In ThumbPointer, we already have a system that points fairly accurately into space, and can be manually or remotely controlled. A telescope co-aligned with the antenna could be very useful, for example:

1. For spotting ThumbSat reflector experiments, for tracking.

2. Potentially, experiments with Wi-fi using light ("Li-fi") from ThumbSats, for downloading data.

3. For aligning the ThumbPointer towards known stars etc.

I have in mind a Newtonian reflector, but with a small camera in place of the usual flat secondary mirror. We have lots of tiny cameras left over from the ThumbSat builds. So it's basically a primary mirror holder, a carbon fibre composite tube, and a camera holder, with some adjustment. Weather protection could be a localised "eyelid" over the primary mirror.

I'm not looking for amazing performance, so I'd like to have fun with lightweight and cheap optics etc. I suppose that I'd be looking at a field of view that just about takes in the whole moon.

I guess this comes under our ThumbWorks "skunkworks" program. If anyone is interested in helping out (or taking over!), we'd like to hear from you!
Shaun 26th Apr 2016
Hi Shaun

I have an old Meade auto-tracking Newtonian telescope that I have been planing to use for LEO sats antennas, since they are lightweight. But I have not had the time to do anything with it. It tracks fairly well for a five inch mirror, but lacks in accuracy for modern scopes. It will track the moon for an hour or more without losing sight of the moon.. so should be ok for low orbit sats.

I have not look into getting the tracking elements for the LEOs into the tracking unit.. but I think it can be done. That is the most problematic issue. The standard way of calibrating it is to locate two known stars and center the scope on each and set the coordinates of each, which the electronics calculate the scopes real location on the earth.... (told you it was old) The newer scopes use GPS satellite downloads to calculate its position on tera-firma.

So my scope is not ideal, as it would have to be calibrated at night... cloudless night... I dont know any way to add a GPS unit to this old thing...

Bobbie 26th Apr 2016
Thanks Bobbie, it's useful to know about your experience with the Meade.

I'm not really thinking about a separate telescope, I am thinking about a co-aligned, simple and cheap telescope that can be added onto the other side of the Thumbpointer axle, opposite the antenna. Most of the hard work is already done with ThumbPointer, so it's just simply (?!) a case of adding a cheap, lightweight convex mirror, a couple of holders, our camera and the carbon fibre composite rod to hold the mirror and camera in the correct position with respect to each other.

Someone could have a lot of fun and satisfaction developing this. Right now the core ThumbSat team is just too busy getting the flight ThumbSats ready, so if any volunteers want to step forward, we'd appreciate it :-)

Shaun 26th Apr 2016
Hi Shaun

I think that the support tube for the scope along with the entire mounting system of the tracker will be too loosely coupled to keep the scope steady, even when the rotor is locked.. My experience with high powered scopes is that the foundation and support system is never quite steady enough... but with an electronic feedback system, such as with most new scopes and cameras, which keep the image steady even when my hands and arms are shaking probably would take care of this problem.. Bobbie 27th Apr 2016
Thanks Bobbie

Funnily enough, my space career actually began with a huge carbon-fibre composite tube that we developed with Russian organisations. The tube formed the "optical bench" between massive (150kg) telescope mirrors and the CCD detectors at the focal point. The problems were slightly different (we were more interested in distortions between 1g and microgravity, and thermoelastic effects) but it does make me fairly confident that I can understand this part of the telescope structure, i.e. I feel that we can make a rigid, low distortion "optical bench" to join our lightweight camera to the mirror.

I'm also fairly confident that the stepper motor will hold very steady.

However, I agree with you about possible movement caused by backlash in the gears and general wobbliness of the mounting, caused by wind etc. I feel that a lot of this can be designed out, but I also agree that it makes sense to use the computer in every ThumbPointer to remove shake using software.

I'm sure there are far smarter people than me out there who could take this little project and run with it. As this is outside the normal ThumbSat and ThumbNet work I'd be happy to personally provide some funds for materials etc if anyone fancies playing.

Shaun 27th Apr 2016
Hi Shaun.

Sounds like you have an excellent background with building cutting edge scopes.

Just a thought... would a stabilizer board such as ones used in the quad copters be useful in keeping the scope on target in windy conditions, etc..??? The board and servos are very light weight and quite efficient .

What timeframe are you interested??

BTW: I have an 8 inch Dob that I no longer use.. it is rather old and I have forgotten the brand name.. lol It is heavy and awkward and I have been planing to build an open frame for it so that I can handle it better... your composite rod system has peaked my interest...

I may be having surgery in a few days, and may be out of work for a couple of weeks...

Bobbie 29th Apr 2016
Hi Bobbie

Playing *my part* in building advanced space telescopes :-) I'm a great believer in KISS, so the active stabiliser sounds a little complex to me - for example, I'd rather fill the base of the ThumbPointer with water ballast ;-)

There's no huge rush with this project, so if you fancy giving it a go with the primary mirror from your Dob then we'd be happy to support you by supplying funds for materials, camera, and having things printed up.

Most importantly, if you do need to have surgery, we wish you a speedy recovery!

Shaun 29th Apr 2016
Kiss is good.. lol I am very lazy and kiss works well for me..

I was thinking of using four stepper motors with planetary gears to collimate the mirror controlled by an Ardunino nano, but I have not decided on the type of feedback to use.. thought about infrared diode attached to the spider and use the camera to control the nano to keep the infrared at peak intensity, but I also was thinking of using an infrared camera, so that wont work... lol I think it would be easy to do it manually, but I would want it to be automatic... thinking remote locations.

Also, could use a nano to auto focus the scope, but that will take some more thought... I need to look at some sensors that are cheap enough to make it worth while.. Could focus it manually with a nano with a pulse rotor and servo motors. But, seems like I already stated that I was lazy..

Bobbie 30th Apr 2016
OK Bobbie, just have a play and see what comes out of your experimentation. If you need any components, let us know.

Just to be clear, I expect ThumbScopes to be mounted to ThumbPointers, so they will already be pretty much autonomous from a pointing perspective. They can also take advantage of the ThumbPointer computer. I'm sure that we can make the whole camera-mirror mechanics have very low distortion. So I only expect to change focus if this is necessary because of the position of the distant object, not to compensate for distortion.

Shaun 30th Apr 2016
Do you have four stepper motors with planetary drives that could be controlled via a nano and relays? I can purchase them online, but they are much more expensive than the standard stepper.. Columnating the mirror may not be a problem with the Thumbpointer, but my limited experience with small scopes on tripods i have bumped them and wind flipping the tripod etc and collimation was required. And I expect the Thumbpointer will get slapped around at times in high wind and ice buildup..

I agree that the scope will probably not need to be focused after installation. Unless one wanted to see a couple of planets when they are at perigee.. or the moon... but that will depend on the camera... the digital cameras that I play with do not need to be focused .... just point and the camera sees everything near and far the same...

Also, I am looking for some carbon composite rods to make the support for the camera, as I plan to remove the entire mirror assembly from the Dob, so the support tube and spider will be discarded... If my memory still works, the scope tube is five feet long... so the camera will need to be a few inches beyond 5 feet as the reflector mirror will also be discarded.. the camera will be at the focal point with a straight line of sight to the mirror. The five inch Meade scope would be easier to work with as it is about 3 feet long, but I had rather not remove it from the computerized tripod mount as it still works great....

just more random thoughts.. I will probably disassemble the 10 inch Dob next week unless i get tied up at the hospital...

thanks for the conversation... I should know Monday or Tuesday if and when surgery will be scheduled... not looking forward to that..

Bobbie 1st May 2016

When you get yourself all patched up and in top health, why don't you email me at wade @ thumbsat dot com, and we'll continue the details of this discussion offline?

When we come up with fantastic solutions, we'll come back and post them here as well as put the plans and details in the Classroom.



Wade 2nd May 2016
If it's possible, it may be quite an interesting idea. Applications are numerous, including easier calibration, more entertaining to make popular shows for the public, and so on. But I have several obvious and fundamental questions in mind:

1) Accuracy of the pointing device is expected in range of "0.25 to 0.5 degrees", as discussed in . 0.5 degrees is the size of whole moon/sun. So, if field of view "takes in the whole moon", pointing accuracy will be about the same as field of view. To be useful, it should be at least x10 better.

2) Does those cameras you have in mind are sensitive enough for any astronomical applications? Stars are not very bright, you know ) Does the sensor of the camera are big enough? Is it possible to remove the optics, and exposure the sensor itself? Is it possible to provide camera's datasheet?

3) How big exposure time you are thinking about, taking sensitivity into account? 1 sec? 2 sec? 0.1 sec? Small exposition needs more sensitive sensor, bigger ones needs guiding (compensation of the earth rotation at least), and rotator is just not accurate enough for any guiding, as far as I understand.

4) Again, do you have any idea for align tuning procedure to reach a co-aligning of antenna and telescope? Yuriy 7th May 2016
As always, great points and great questions Yuriy.

All of those questions will need to be looked at and we'll have to decide if it's truly possible to do what we have in mind, or if it is just not an option. In which case, maybe we'll put lasers on the off antenna side and create a sky show on the clouds. :-D

As you know by now, we love to think around the corner and try to create and develop new ideas that, at first glance, don't seem feasible. At first glance, this might not seem feasible either, but that's just one more reason for us to look at it! :-)


Wade 7th May 2016
The easier way to test the idea is to try to build simple refractor with your camera, just to understand if sensitivity is good enough.
Mind: if it's NOT good enough, camera cannot be used for star tracking / orientation detection on the satellites as well.

All you need is to buy one lens from the eyeglasses shop. They normally uses 50 or sometimes 60 mm lenses to carve a lenses of needed shape for your specific eyeglasses type. So they have a range of lenses of different strength and materials. You need simple glass lens without coating. If glass lenses unavailable, plastic one may be used as well. In Russia it's quite easy and cheap to get one.

Then you need to buy a plastic sewer pipe of appropriate diameter. Plastic bung for this pipe can be used to fix the lens on one end of the tube - it works quite well. If your lens is too big for your tube, eyeglasses shop can curve it to needed diameter for the reasonable price.

The only hard part is to make something to attach the camera to. There should be something to move the camera to and fro to catch the focal plane. I used a piece of water pipe foam insulation pies with the screw and nut glued to the foam, but it was before 3D printers becomes accessible - now you may print something even better.

This system is not very good, because chromatic aberrations may be quite big. It may be fixed by replacing the objective lens, but at the first stage it just not worth it. It should be more than enough to understand if your camera is sensitive enough. Yuriy 10th May 2016
If your camera have build-in lens, it should be removed, of course, and sensor should be exposed directly.
Actually, we effectively replacing the build-in lens to the much bigger one :)

Again, what about datasheet of the camera?
Is it available or maybe it is covered by some NDA for some strange reason? Yuriy 10th May 2016
Hi Yuriy

Thanks for all of this information, I really appreciate your taking the time to respond. I'm afraid that I gave the wrong impression.

1. I actually have a pretty good amount of experience with optics and digital cameras.

2. I'm not looking to build a particularly good performance telescope in the first instance.

3. I'm not really looking for ideas about how to do this, I'm looking for someone with the time and energy to actually do it.

At least that was my original intention. I think that Bobbie is going to experiment with the more advanced things that he has mentioned above, we will provide him with any materials that he needs.

I am going to give myself the luxury of a weekend in summer to simply connect a mirror and a camera on a stick and just play. It can be very liberating not to over-analyse something, and just get down to building :-)

Thanks again!
Shaun 10th May 2016
Hi guys.. I have had good results with two of my scopes with placing my cell phone camera on the eyepiece and focus the eyepiece until the image on the phone looks ok.. I have a new 8 Megapixel camera for my raspberry Pi 3 and a simple angle mount.. my first experiment is to place the RPi camera on a low power eyepiece and take some photos of the moon. I think that the actual mounting of the camera at/near the focal point of the mirror without the secondary mirror is a rather simple matter... I can operate the RPi headless and see the live images from my desktop.. the problem has already been mentioned by Yurly, in that the rotor may not be accurate enough to be accurate enough.. the computer controlled Meade scope that I use is very accurate and stable, but it was designed for that purpose. So, my main experiments will be trying to get the scope to be stable on the rotor mast. and keep the mast steady. I am sure that metal mast will be required and probably aircraft quality aluminum schedule 40 minimum.

I think that Shaun's idea of a camera on a stick is excellent... I mentioned before that a carbon composite rod might work well.

But, I have surgery scheduled for Friday morning and wont have time to test the RPi camera with the scope this week..

I will try to keep you posted.. Bobbie 11th May 2016
Bobbie, you are rapidly becoming a ThumbSat Star! This is exactly the sort of approach that I am talking about. Play a little, make improvements, play a bit more... We can get creative with stability without making it too complex, and we can compromise with co-alignment and field of view.

I look forward to the results of your experimentation, but most importantly we look forward to having you with us again, fit and healthy, following your surgery :-)

Best wishes

Shaun 11th May 2016

There is no secret with the existing camera module we are currently using aboard ThumbSat.

The spec sheet can be found here:

We have access to other cameras of course, including some we are having custom built, but these are the ones that we have in surplus from ThumbSat at the moment and are easiest for us to put into immediate use.

Thanks for asking.


Wade 12th May 2016
I am back from the hospital, surgery went well... now to stop the pain meds and heal.. hope to be involved with the telescope experiment within a few days..

Bobbie 15th May 2016
So very, very glad to hear things went well, Bobby.

I wish you the speediest and most comfortable recovery.

Don't think about this stuff until you're completely back on your feet. This should be the least of your priorities!

Wade 15th May 2016
Great news, Bobbie, welcome back!!
Shaun 15th May 2016
While lying in my hospital bed, with my computer by my side, and a goose-neck mount for the monitor, which makes placing the monitor for comfortable view... (soap box aside)

I have tried to use the camera that Raspberry sells with their excellent mini computer, I find that the drivers for the camera and the dongle conflict, The camera works but the Thumbsat dongle stops working.. I have to disable the camera to get the dongle to work... I will look again when I feel better, and try different drivers for the dongle. I have three different drivers for the dongle and I have only tried one. So, more experimenting to be done.

But, looks like using a camera with the dongle is problematic.

I am using Direwolf digital decoding software package with GQRX and it works fantastic with the Raspberry Pi 3b. I will upload a screenshot of it running along with gPredict and fldigi, so look for it in the media section...

I am having problems getting data from GQRX to fldigi. Since I have Direwolf working, I think it is a matter of getting fldigi to connect with UDP to GQRX.. any help with this will be greatly appreciated.

There is a possibility that Direwolf will decode the Thumbsat signals with a little tweaking. I am not sure that I can accomplish that... more help needed..

Bobbie 28th May 2016
My word, that is dedication, Bobbie! I hope that all is going well in the hospital!

Wade has more experience with the Pi than I do, so I'll leave him or Fraser to answer, but I wanted you to know that we appreciate your dedication!

Shaun 28th May 2016
Hi Bobbie,

My best guess is that the drivers for the dongle and camera both require single thread apartment use of something (I don't know what without spending some more time). I've had this sort of problem before and that was the issue, but this could be something completely different!

Thanks for your dedication and get well soon!

Cheers, Fraser James Robinson 28th May 2016

very interesting Bobbie 21st Jun 2016
The link to StarPi is very interesting indeed, and coincidental!

If you look at Chris' log, you'll see that he was at the UK Maker Faire this year. I too was there (and got to see the Copenhagen Suborbitals rocket and test capsule), and chatted to Chris and the PiKon guys. While neither are quite what we are looking for, they are certainly inspiring and they promote thoughts.

Shaun 22nd Jun 2016