Digital Setting Circle Computer

[mbirrell]

Hi everyone,

I am interested in making my own Digital Setting Circle Computer and have looked at a few threads on the subject. Has anyone checked out this site:
http://eksfiles.net/digital-setting-circles/

I have the know how and the tools (yes I have a Pic programmer) to make this unit.
I am after opinions.
Also if anyone needs me to program a PIC I can do that... just pm me.

[mswhin63]

PIC are quite easy to do now using ICSP programming method. It is cheap and can be done almost anyone. The main issue as seen with other post is the encoders. I personally working on an encoder installation that will be a one size fit most. But it is possible to get encoder setup from Wildcard used for the Argo Navis.

EK Box is a simple kit. I already have a USB interface and am working on a Bluetooth interface which does away with have the circuitry for the current EK box.

[bojan]

Yes, see here:
http://www.iceinspace.com.au/forum/showthread.php?t=51290

I am using Ek box as part of my Bartelised dobson, see here:
http://www.bbastrodesigns.com/BBAstroDesigns.html

[mswhin63]

I agree with Bojan, that thread contain a lot of detail over time. Usually find something there that is helpful.

[mbirrell]

I must be missing something but why are you using these expensive encoders for on a PIC when there are much cheaper alternatives?
I definatley would not use friction drives or timing pulleys either - they are way too costly and ugly... the most I would spend on an encoder would be $5 tops and that would be an expensive one!

[bojan]

That's an attitude ! :thumbsup:
But I am, actually, my encoders cost me $0.00 ;)
(they are from old mice, of course), see here: http://www.iceinspace.com.au/forum/attachment_browse.php?a=65266
Timing belts are also $0.00 if pulled out from old printers..
Currently I am working on something more expensive (sometimes a cost is un-avoudable ;)), and it is from US digital.. not belts, no pulleys, no gears, just contactless optical incremental sensor and some laser printed strips on paper tape. I think it will be according to your taste, despite $7 price tag each :)
BTW, welcome to the forum :welcome:

[mbirrell]

Thanks bojan,

The idea that I am working on works with a photo interrupter (Jatycar part no. ZD1901) and a strip printed on a overhead transparency. I was thinking of using a very wide arc to give me greater resolution.
The question is though how many arc seconds do these encoders have to be accurate to? If I am using a push to I wouldn't be able to adjust my scope to down to 1 arc sec.

[bojan]

Only one arcsec is a lot of resolution, you will not be able to avoid up-gearing if you want to achieve this.. And the accuracy will always be questionable..
My scope (with friction wheels ratio 1:20 or close to that) has around 7000 ticks per rot on both axes, which is around 180" or 3arcmin, enough for visual.

[bojan]

This is not going to be easy to use for encoder, you will need more than just transparent pattern printout (additional mask). And, you will need 2 of interrupter modules for each axis (to be able to determine the direction of movement).

Try this one: http://www.vishay.com/docs/84756/tcut1300.pdf
(they are used in optical mice)

Also, Steve (Kinetic) has his go on this: http://www.iceinspace.com.au/forum/showthread.php?t=42151

My idea is around this one: http://usdigital.com/assets/general/168_aedr_datasheet.pdf

Very simple arithmetic shows that for 10,000 t/r I need a wheel (or ~100° sector for elevation) 30cm in dia, quite reasonable for any dobson mount.
And no moving parts whatsoever (apart from telescope itself)
Alternative is this one, but more complicated to use (it needs ADC's):
http://sensors-transducers.globalspec.com/datasheets/18/AvagoTechnologies/8BA8AA78-2EA7-4682-BC78-D2E6D77A7724

[mswhin63]

It is a matter of preference, My design is quick and easy to configure requires very little effort as I don't have a lot of time ( I mean very very little time). Although as light polution is beginning to bug me in my area, I may have more time available in the near future.

Also mine is designed not to alter the structure of the DOB in any way. I am able to remove all parts and the DOB is completely original.

Some people don't mind drilling holes in there equipment. I do! What I do might sound like a lot but I am able to do lots in small bits of time. I love to experiment it was what I was born to do.

Not sure why you think friction drive is expensive. Mine was quite cheap actually, the whole friction setup only cost me about $30.00 plus the encoder. I haven't finished to whole project yet as I keep running out of time. I am patient though.

It is all a matter of preference. :thumbsup:

[mbirrell]

I recon both of your systems sound fantastic.
I think I will probably end up buying an encoder if my ideas don't provide me with enough resolution.
I was really wanting to try to design a very low cost system so to help others that can't afford the big bucks.

[mbirrell]

I have just been reading about hacking old CD drives and there may to use one as an extremely high resolution encoder. Here is the link:
http://www.repairfaq.org/sam/laserlia.htm#liarfdm

I might be able to get somewhere with this.

[Bassnut]

Slightly off the subject, but I chuckle when I see ppl needing PIC programers and bothering with C and compilers etc. PICAXE (http://www.microzed.com.au/) PIC basic is free, their PICs absolutely fly at up to 32mhz, one-word hardware interface keywords and literally plug straight into RS232 via the PIC pins for one click download and run, no programmer required. One "count" command is all thats required for encoders and a "serout" to display position on an LCD, no driver code required. And precoded commands to control steppers.

[bobson]

Hi Bojan,

I just read data sheet for your new encoder and have to tell you that one has great potential! That would solve so many problems for dob owners. No pulleys, no belts, no friction drives. Keep us posted how is it going mate. As far as I can see this could work with EK box without any modifications, correct me if I am wrong.

bob

[bojan]

Exactly right, mate :thumbsup:
That is why I ordered samples and next week or so there will be some prototyping going on here.
I will certainly posted the results and conclusions here.
The only possibly better solution (from power consumption point of view) would be magnetic sensor (the one used in callipers, with 0.01mm resolution. But I could not source them anywhere.
My Chinese connections are not good enough :sadeyes:

[mbirrell]

Sounds like you will have a good system Bojan. That sensor that you gave me the specs on is definitely a good option to use with the encoder wheel that I intend to use. to use it without mechanical parts I could print out a large encoder wheel with the required resolution of 4000 tics. If the sensors are easy to source they would be a gold mine.

Ok what I have come up with so far is:

I have pulled a spindle out of an old CD drive - it will be used for the axle to drive my encoder wheel.

I have found to necessary software to make an encoder wheel that everyone can download for free:

encoderwheelv1.3.ps - Encoder Wheel Generator by NickAmes: http://www.thingiverse.com/thing:1527
The instructions are on the web site.

GSView and Ghostscript - Postscript file viewer. you can find a version by just Googleing it. You need these programs to view the wheel encoder and print it if you are not a computer nerd like me.

A few old CDs' - Everyone has them!

As far as the optical transmitter and receiver goes I think the cheapest solution is buy savaging an old mouse. That is covered by other posts so I won't mention the process here. the trick with the wheel is that you need to print your wheel on an over head transparency.

The turn ratio will be around 4:1 which is a reasonable ratio as the chance of slippage will be reduced significantly. To make the main planetary friction wheel I was thinking of gluing about 6~8 old CDs' together. using the old CD spindle and CD planetary friction wheel will give me around 5,400 ticks which will do the job nicely.

I will follow up with pictures and more details as the project progresses.

[mbirrell]

If anyone wants me to send them the Dave Ek project encoder interface parts or build it as a working device just pm me.

[mbirrell]

I have made an encoder wheel from an old CD. I printed the code wheel on an over head transparency then laminated it to protect it. I cut the printout in the shape of a wheel then glued it to the CD with spray on contact adhesive. I found an old CD drive I have and removed the rotor part where the CD sits on to spin and what do you know the shaft from the gearbox fitted it like a dream. On the other end of the gearbox I put a remote control car wheel which press fitted nicely. I have also raided a few old mice that I have for the sensors so we will see how they go working with the encoder wheel. oh yea the encoder wheel has 1,000 ticks per rev and the gearbox ratio is around 1:4 (hope I got that around the right way).
I was hoping not to make a wheel with a gearbox but the resolution would have suffered but still it is far cheaper than buying a Encoder so far.

(http://cid-ec9599340af4987c.skydrive.live.com/self.aspx/Public/Wheel.JPG)

[bobson]

That looks good mate. Could you tell us where exactly are you planing to put that one? What sort of telescope mount do you have?

thanks
bob

[mbirrell]

Hi Bob,

I have actually progressed a bit from what I have posted. I am not particularly happy using a gearbox and having to "hack" a telescope that I will be putting it on (Meade 12" Dob). What my idea now is if I can make a wheel with enough spaces it won't need a gearbox etc which means no slippage. This might mean getting a stainless steel disk laser cut (if the price is right). For the sensors I would be extracting them from and old mouse.
I have been also thinking of connecting a Bluetooth to RS-232 converter from Jaycar (XC4130 is the catalog number) to the Dave EK box so I can use it on my new laptop.

Attached is what I am getting a quote on. I will let you all know the price when I get it.

Also I have got a quote on the MAX232CPE RS-232 level converter:

Sub total: $19.80 - MAX232CPE RS-232 level converter x 5 of them
Freight: $7.50
Total: $27.30

if anyone wants one just pm me...

[jeroenhmg]

isn't it easier and cheaper to just buy BETI from Astro Devices if you're planning to use Bluetooth? Here's a link: http://www.astrodevices.com/products/products.html

Best Regards,
J.

[mbirrell]

Hi Jeroen,

Yea it looks as if that would be a cheaper way to do things if you want bluetooth straight up. Not a bad price for what you get I recon.

[mbirrell]

My main plan is to mod a Meade Dobsonian to take a Encoder wheel without drilling any extra holes in the telescope's mount. Also I would like to make an encoder wheel that is cheap and accurate. My final mission is to enable just about anyone who has a Dobsonian to implement a push to system with minimal skill and cost.

Please be aware I have not tested most of the stuff I have written about yet but in the up and coming weeks I hope to provide photos and diagrams explaining everything.

Here is where I am up to so far...
I think I have gone through the difficult part of designing a cheap encoder wheel that will suffice the average amateur astronomer - all it is really is a 2,000 step encoder wheel printed out on an overhead transparency and the use of old mouse wheel sensors to send the ticks to Dave EK's box - no big deal.

Now for the trick of mounting the encoder wheels and sensors on the scope without hacking it up...

At the base of a Meade it has a 2 1/4" x 5/16" Hex Pivot Bolt - If I replaced that with a Tee (A nut designed to be driven into wood to create a threaded hole), a 3 1/2 inch long hex bolt and 3 nuts then I would have a way of securing the Encoder wheel in a fixed and central position.
The alt encoder wheel should be just a matter of getting double sided tape (Jacar cat no. NM2821) and placing it on the alt axle making sure that it is centered correctly.
The sensors for both will have a bracket that will be fixed with double sided tape. Wiring will be cable tied and held in place by the use of adhesive tie mounts (Jaycar cat no. HP1195). The Dave EK box will be held in place by (you guessed it!) double sided tape.
All in all I am hoping it will look as neat as I imagine (I am an Electrical fitter mechanic by trade and like doing a good job).

Lets see how I go...

[mbirrell]

Gee I thought SMD soldering would be a piece of cake... till I tried :eyepop:

[bojan]

It is.. But it requires somewhat different approach.
Instead of soldering iron and solder wire, you should be using solder paste (Jaycar) and hot plate (cooking appliance).
Microscope (or at least large lens) and precise tweezers are very helpful for placement of SMD components

[mbirrell]

yea after watching YouTube videos I discovered that is they way to go.
I however have only got a set of head mounted magnifiers, a lot of flux with a 1mm soldering tip and a tiny bit of solder...

The rest of the project is going quite well but it is a very slow journey.

[bojan]

This looks like a reflective module to me :thumbsup:
I was supposed to get my PCB yesterday..

[mbirrell]

Yea it is - one of the types that you suggested. I am going to try an encoder wheel that has been printed on a laser printer. I got the PostScript from: http://www.thingiverse.com/thing:1527 to generate it.
All in all the most work I have put in is with the circuit board - I recreated it in FreePCB. To print it out I used ViewMate (not very user friendly). Then I used this paper called Etchatron to transfer it to the PCB. I think I could have used HP Glossy Laser paper though (a little bit cheaper). All the electronic components I got from Farnell.

[mbirrell]

Well last night I got my Dave EK DSC going at last!
I have got a bit side-tracked lately as I have been working on making a CNC machine. I figure this will help me make certain parts for this project and others to come.
Back to the Dave EK board - I was a bit worried for a while that the hard to get parts that I subsituted on the board wouldn't work - but they did! Also when I programed the PIC and put it in the board it didn't work either until I selected the correct clock to use (the crystal one and not the internal one!). I think that this part of the project would have been much easier if I had have taken up the offers of people making the board for me for a few bucks extra but I just have to play with fire :rolleyes: