frugal

1 hour ago, jimboozle said:

did you get it working with SGP? I'm now highly tempted to use the arduino I was given a while ago to add motor focus to my WO66.  I've got a moonlite on the 120ED, but not willing to shell out for another 'real' focuser...

Unfortunately my mounting technique caused some problems. It made the scope far too camera heavy, and the mounting box also clipped the altitude adjustment bolt. So I have had to take it off again until I can come up with a new mounting approach

1 hour ago, Danny83uk said:

Would be also be acceptable to stick some stupid bunny ears and a dainty nose on them? 

That would also work.

In all seriousness there are a couple of places you can put your images to share:

 - Instagram

 - Flickr.com - They have an Astrometry group and if you add your images to the group they are plate solved and annontated by Astrometry.net (takes a couple of days)

 - Astrobin.com - Where I initially put my images up. You can add details of the number of light and dark frames, and it will calculate the total integration time as well as plate solve and annotate the images; you can also have multiple versions of the same image if you canture more data or improve the processing. However it go so popular that the site maintainer needed to move to a paid model in order to keep up with the bills. Currently $18/year for 12 images a year, or $36 for unlimited.

- dso-browser.com - I had not really appreciated the image abilities of this site, but like Astrobin you can upload an image along with all of the details of the equipment used and the number of integration frames; automatic plate solving. It also adds your image to the community images for a target, so that if you look at the details of a target you get to see everyones images. The only thing you can not do is upload a revised image if you improve the processing. The only problem is that there is no easy way to share the images to the forums (sharing to Facebook, Twitter and Google+ is fully supported).

I believe you are only allowed to post astro images on Instagram if you first apply a faux 1970's Polaroid filter, or show it next to a plate of food ?

7 minutes ago, dyfiastro said:

Glad you got there in the end. You are using pretty much the same setup as I am in regards to the skywatcher focuser.
I found the same thing when I changed my motor but also came across a small issue when using APT.

For some reason I could not go past 10,000 steps in APT which using the standard code was not enough to be able to rack the focuser all the way in or out.
I ended up changing the ascom focuser steps section in the code to 19 instead of 8, this brought it down to just below 10,000 and solved the problem.
If you are only going to use the bundled software then it will not be an issue but if you end up with problem then this may be the solution and is still fine enough for critical focusing.

I am going to be using Sequence Generator Pro which assumes you have it in focus to begin with. You tell it how many steps to take and how big each step should be and then it takes a photo at each step and works out which one is most in focus. So if I say take 7 images at 250 steps apart, it will rack out by 750 steps and then take a photo and then rack back in again 250 steps at a time until it is at 750 steps the other side of the starting position, then it works out a curse and moves to the lowest point of the curve. So so long as I declare an initial position high enough it should be fine.

I decided that as I still had all of the bits for the focuser I must have just not tighten something up in the correct way. So today I went out and disassembled the focuser and put it back together again. Fine focuser now works. The coarse focus knob is quite stiff to turn which I do not think was helping the poor little motor last night. I looked at various high end expensive electric focusers, and even the Baader Steeltrack focuser uses a belt from the motor pulley to the fine focuser knob; so I have decided to ignore the American's advice and just do what most other people have been doing and drive the whole thing from the fine focus knob.

The electronics box is mounted to a surplus piece of 1.5mm steel that was part of a router fence and had the advantage of being bent at 90 degrees so it would not sag. The steel support is then bolted to the end of the guide bar. The steel support had two slots cut in it so that I can adjust the tension by sliding the whole lot up and down the guide bar by about an inch if needed. The belt was made up from a length of GT2 belt that was cut to size and superglued together on a diagonal. and yes that really is a strip of Elastoplast you can see providing some extra strength to the joint.

The red electrical tap is covering up a couple of mounting holes for the motor that I cut in the wrong place. The black tape is attaching the thermometer to the side of the unit (I figured it should be outside the box to get the best reading).

Using Dave's focus tester I need to do movements of about 250 to be able to detect any movement in the drawtube, so the accuracy should be super fine. The next step is to wait until there is a clear night and have a go on some stars with SGPro.

Some days I shouldn't be allowed string, let alone expensive mechanical devices like telescopes...

Got everything neatly (well sort of neatly) packaged into the little box I got for it. Connected it up to the PC and the power, all still works. Time to mount it on the scope. Decided as the scope sits on it's dovetail when it is not in use that I would attach the box to the top of the tube rings using the camera mounting point. All good so far.

I decided that the least invasive method to connect to the focus knob was a belt drive. I have read in the instructions with Rob Browns focuser that you should never use the fine focus knob for the belt drive as the constant tension on the knob causes it to fail. So I thought "4096 steps per revolution I shall drive from the coarse focuser". First thought was to put some of the belt on the knob to make a large gear out of it. Of course when you put a 2mm pitch belt around the focus knob you get a 2 and a bit mm pitch gear, so the belt kept slipping over it.

Next idea was to attach one of the 16 tooth pulleys that came with the belt directly to the centre rod of the focuser. So following the instructions on a You Tube video of how to fit an autofocuser to an ED80 I disassemble the focuser and try. Of course the pulley has a 5mm bore and the rod is only 4mm in diameter, plus if I do that then the rod is not fully supported at both ends so it wont work anyway.

Third idea. Drill out the end of the focus knob and put in a bit of 5mm rod, then attach the pulley to that. Ok, that seems to work. However when I put it all back together and fit the belt, it won't turn the focus knob. The commands are getting to the motor, and I can see one side of the belt getting tighter and one side getting slacker, but the knob is not turning round.

Plus to cap it all off, something I have done has disconnected the fine focus knob. It now goes around the same speed as the coarse knob if I turn the coarse knob; and it just turns freely with no effect on the focus if I turn the fine one.

So now I not only don't have electric focusing, I now don't even have fine focusing ;(

Attempt 2 seems to have gone a lot better. I have a very basic board soldered together (no Bluetooth, and everything apart from the motor powered from the USB). Dave's Focus test software works flawlessly and it picks up both the temperature and the focuser position.

Now I just need to figure out how to attach it all to the box and how to mount it to the scope.

Fail the first: I was experimenting to check that I could at least get an the Nano to turn the motor. All connected up with Dupont connectors. Turns out that if you connect a 9V battery to the motor controller board the wrong way round nothing happens; if you connect a 12v power supply the wrong way round there is a pop, a fizz and then the magic smoke escapes from the motor controller... This is why I bought a 5 pack

Fail the second: I figured I should practice soldering components onto the vero board as I have not held a soldering iron since I was 13. Turns out that the soldering iron I inherited from my dad doesn't heat all the way to the tip properly, so I was trying to use the side of the tip about 1/4" back from the tip. Not really going to work. Amazon Prime to the rescue, new soldering iron and a heat proof mat should be here tomorrow (so I can put the chopping board back in the kitchen)

Fail the third: Figured that I should have a look at Dave's schematic and figure out where the voltage regulator fits in so that I can power the 12v stepper motor without blowing up the cheap Nano clone. Turns out that I had forgotten to add the capacitors to the shopping list in order to smooth out the DC voltage on either side of the regulator. Turns out that you can't buy two capacitors on Amazon Prime

Did I mention it has been 30 years since I last did any kind of electronics...

The last bit turned up today. So I am having a play with the various bits before trying to remember how to use a soldering iron (I think I last picked one up 30 years ago doing DT at school). Just connecting the Nano to the stepper motor board and then the 12v power supply to the stepper motor board I can successfully get the motor to turn. Now to figure out how to wire it all up onto the veroboard.

Interesting. I have all of the parts for mine, except for the Nano itself which is on a very slow boat from china

Thank you.

Are you mounting the stepper motor onto the tube rings somehow?

12 minutes ago, rwg said:

@frugal

Are you using the 2.10 beta? That has improved star detection and more tweaks to it too - it seems to have fixed issues in 2.9 with faint stars not working well for a number of people, particularly with lodestar cameras which seem to have no gain control

Set a digital gain of 2x if you have no camera gain control and see if that helps. With a camera gain control (QHY, ZWO, Altair, etc) I'd usually advise exposures in the 1-4s range and a high gain.

I am using the 2.10 beta with the correct licence key.

I was using a x4 (I think) digital gain and an exposure time of 15 seconds. 8 seconds did not give the required 15 stars.

I think the problem boils down to the field of view just being too narrow for this technique to work. With the ST80 my field of view is only 0.9 x 0.6 degrees

As an experiment I left the alt/az untouched and watched the plate solving for a while. Due to the field of view and the fact that it kept changing which stars were selected, the solved location would change. For one solve it declared that the NCP was at the top right of the screen, and the next time it declared that it was off the screen to the bottom left.

Also as a feature request: There does not seem to be any indication of whether the software is currently plate solving or not. It shows the "Plate solve failed", "Plate solve successful" message, but I do not know which frame that applies to. How can I tell if the software is currently trying to plate solve, or if it has finished?

Also, does it update the frame view with the solved information straight away, or only when the next frame is received from the camera? I.e. if I start taking 15 second frames at t=0, at t=15 the first frame is ready, the software shows the frame with the higlighted stars; at t=19 the software finishes plate solving frame 1. Does it update the view straight away? or does it wait until t=30 and the second frame arrives?

Robin, 

I managed to complete the polar alignment process this evening, but it was not a simple process. I think that most of my problems come from the fact that the field of view is only 0.9 x 0.6 degrees. Plus I need to increase the exposure time to 15 seconds in order to recognise enough stars. I think that this led to a problem where the Solved solution moved around.

It looks like a really nice solution, but I just don't think that my system is setup with a wide enough field of view for this version of Polar Alignment to be successful.

Certainly not at 10:30 at night as I was desperately trying to set everything up and SharpCap was insisting on a reinstall and re-boot, I was clicking OK buttons as fast as they were coming up 

I think what you have done is sufficient. If people don't want issues, then they shouldn't be using beta software. I am certainly not going to complain.

One point regarding the software: What is your view on some kind of splash screen? It often takes about 15-20 seconds from the time I double click the icon to the time I get a window. A number of times I have thought I had miss clicked, so I try again and I end up with 2 copies opened.

1 hour ago, rwg said:

If you look on the beta download page there is a license that you can use with the beta for now. This will continue to be available and updated during the beta period. So yes, you can still use the features for the beta period, but we've got to the point that the license entry/checking code needed to be turned on to make sure it didn't have bugs in it!

cheers,

Robin

Thank you for the quick response. I did have to download a new version of the Beta as when I started it last night, as the previous version had expired. I had not realised that there was a beta licence file to go along with it, probably because the software automatically downloaded the new version for me, so I did not think to go looking for a licence file to go with the download.

I managed to fiddle with the settings to get SharpCap to recognise about 26 stars last night by tweaking exposure time, digital gain, noise reduction and some others. After a few failed attempts (I know I am trying this at twice the recommended focal length, so I was chancing my arm), I got the image to Plate Solve. Then when instructed, I moved the RA axis 90 degrees and it Plate Solved again. Woot!. Then I clicked Next as told... "Guided Alignment is only available in SharpCap Pro"... [removed word]

This is with the latest beta version. Having a look at the website it says that the beta versions will have all of the licenced features until the end of the beta period. Has this changed?

Does the Polar alignment feature work well in 2.9? Is it worth uninstalling 2.10 beta and installing 2.9?

Hmmm. I was hoping to go down the Bluetooth route in order to save on precious USB ports, and only have the focuser connected to the 12V input supply. I shall have to consult with Herself as the resident Electrical Engineer to figure out how to wire a voltage regulator in safely

ebay, here I come...

PS - Thanks Dave.

A question from the hardware ignorant: Given that the Nano spec say that pin 30 can take 6-20V, and the ULN2003 wil take upto 12V. If I use a 12V stepper motor as Pauls72 did; is there any reason why I can not run this off of a 12V DC power supply rather than a 9V battery?

I also note that 4203 - 4207 all have a number of additional files connected but after 4208 to 4219 there are no additional files.

4203 47 extra files

4204  47 extra files

4205  11 extra files

4206 11 extra files

4207  11 extra files

It is not so much that those indexes have additional files, but more that the index is so big that it needs to be broken up into multiple parts to be usable.

I have not altered any index files so I assume all of them. Not really sure what to do or how to tell which ones I don't need??

If you have all of the index files then it will be slowing tour matches right down. You should just need to install a few indexes based on the camera and scope combination you have. Thee is a section in the original post

Julian,

Which index files are you using?

Hmm... if your max is 2.5 how can your min be 2.4 deg?

As I know that my images are exactly 2.44 degrees wide, I wanted to explicitly clamp the scale to as close to that value as I could. After all why search scales from 1.22 to 4.88 when I know that the scale will be exactly 2.44. It solves faster with 2.4 to 2.5 than with 1.22 to 4.88, but I got fed up with resetting the values.

The bottleneck on plate solving is now waiting for Backyard EOS, and converting the 22mp jpeg to a ppm and extracting the stars. So I am not too worried.

Hi

I think I read somewhere that your scale max value should be 50% greater than your calculated fov and that the min value should be 10 or 20% of that value. You'd therefore need the index files that cover that range. Also, scale refinement set to 0.1.

Interestingly I have noticed that after I solve an image, AstroTortilla changes the Max Scale to twice the width of the image in degrees and the Min Scale to half the width in degrees. So even though I tell it that the scales should be 2.5 and 2.4 respectively (my setup produces about 2.45 degrees across the frame), after solving AT updates the settings from 2.4->2.5 to 1.22 -> 4.88. So AT itself is using 50% and 200% rather than 20% and 150% as it's own documentation suggests...