Arduino Ascom focuser Mark2

[cthorpey]

First Light for the focuser......and it all works!

APT telling me its 0.5deg C outside, slewed to Betelgeuse and tried the autofocus. 2 mins later it was spot on. I LOVE this.

Currently imaging M42 before it vanishes behind the house next door.

Thanks again to everyone involved, this thing is awesome.

Chris

[redtail]

I had a look at your eeprom code but believe you have some serious issues.

EEPROM in the controller is somewhat limited to 10,000 writes.

The way the code is written, this could be very quickly reached?

As example, put a little counter in that increments each time the eeprom is written, then do an imaging session and at the end of the night just print out/dump how many times it was written (might need another serial command) - then divide 10,000 by that number and what do you get? The number of sessions before the eeprom will likely fail?

Cheers

Robert

Does this mean that the focused will stop working?

[brown_rb]

It means that once the eeprom limit is reached, the eeprom will not be able to store the data correctly, this may result in incorrect readings of focuser position on power up. Its more than likely once the limit approaches that some bits within the storage location fail first, so that storing something like a position of 1024 might then be read back as 513. In other words, unreliable. What effect this might have on the focuser operation is best answered by the author.

There are many ways to overcome this issue. For example, only write to eeprom after a move operation, and after a time delayed (so that it is not triggered by autofocusing), and to cycle locations within the eeprom. These approaches I have taken in my own code, such that if one imaged every night of the year, yes, every night of the year, it would take over 9 continous years for the eeprom to reach the limit.  

Cheers

Robert

[tekkydave]

It means that once the eeprom limit is reached, the eeprom will not be able to store the data correctly, this may result in incorrect readings of focuser position on power up. Its more than likely once the limit approaches that some bits within the storage location fail first, so that storing something like a position of 1024 might then be read back as 513. In other words, unreliable. What effect this might have on the focuser operation is best answered by the author.

There are many ways to overcome this issue. For example, only write to eeprom after a move operation, and after a time delayed (so that it is not triggered by autofocusing), and to cycle locations within the eeprom. These approaches I have taken in my own code, such that if one imaged every night of the year, yes, every night of the year, it would take over 9 continous years for the eeprom to reach the limit.  

Cheers

Robert

Im working on a version that will store to EEPROM only when a new position is reached rather than every step, which it does at the moment. I'll try and get it out soon. If you have any problems with EEPROM you can change the location used in the sketch to start using fresh memory locations if necessary. There is no way of avoiding this effect just minimising the chances and delaying the inevitable, even though it might take a long time (years). It wont stop the focuser working - it just wont remember where it was positioned the last time you used it.

[redtail]

Thanks guys and happy new year

[Gina]

Yes, the problem of EEPROM failure can be delayed for several years be judicious use of the EEPROM addressing but if you are really worried the data could be stored on a separate device such as a memory card (like I am doing with my weather station).  But in this application I think an SD card would be overkill.

Edited January 2, 2015 by Gina

[tekkydave]

Yes, the problem of EEPROM failure can be delayed for several years be judicious use of the EEPROM addressing but if you are really worried the data could be stored on a separate device such as a memory card (like I am doing with my weather station). But in this application I think an SD card would be overkill.

I thought about storing the position in the windows registery but a nano only costs a few quid and has lots of spare eeprom locations so decided it wasnt worth the effort.

[richyrich_one]

Hi

You've inspired me to give this a go. Thanks for sharing this project, an excellent DIY!

I'm going to give the direct drive a go on my 10:1 crayford.

Parts ordered, and they are so cheap.

Cheers

[tekkydave]

Hi

You've inspired me to give this a go. Thanks for sharing this project, an excellent DIY!

I'm going to give the direct drive a go on my 10:1 crayford.

Parts ordered, and they are so cheap.

Cheers

Thanks, Rich. Good luck with your build. Post some pics when you get it going.

Dave

[richyrich_one]

Ok, got all the bits connected and working on the bench.

For simplicity power wise I'm going to eventually go with a 12v stepper, 2 on their way on a very slow boat. Really loving how cheap these bits are.

When I get it boxed and coupled up I'll post some pics.

One thing, I noticed the sketch controls the motor in high and low speeds. Is there a reason for this? What am I missing?

[tekkydave]

Ok, got all the bits connected and working on the bench.

For simplicity power wise I'm going to eventually go with a 12v stepper, 2 on their way on a very slow boat. Really loving how cheap these bits are.

When I get it boxed and coupled up I'll post some pics.

One thing, I noticed the sketch controls the motor in high and low speeds. Is there a reason for this? What am I missing?

The high and low speeds was just a feature I added to speed things up if you were moving the motor a long distance. It should be easy enough to change the speeds or the threshold value where it changes from hi to lo to suit your situation. In normal use doing small focus movements it will usually move at the lo speed. Looking forward to some pics in due course.

[richyrich_one]

The high and low speeds was just a feature I added to speed things up if you were moving the motor a long distance. It should be easy enough to change the speeds or the threshold value where it changes from hi to lo to suit your situation. In normal use doing small focus movements it will usually move at the lo speed. Looking forward to some pics in due course.

Thanks Dave.

I have already been tinkering.

[richyrich_one]

If anyone else is looking to do a direct drive connection to a 10:1 crayford these look perfect.

http://www.electromorph.co.uk/Pages/Metric-Shaft-Couplers.aspx?lu=m&ld=2.5&ru=m&rd=5

[brown_rb]

If anyone else is looking to do a direct drive connection to a 10:1 crayford these look perfect.

http://www.electromorph.co.uk/Pages/Metric-Shaft-Couplers.aspx?lu=m&ld=2.5&ru=m&rd=5

You should really use flexible couplers - search ebay for "flexible coupler", as these reduce the stress load caused by any inertia jumps of the stepper and misalignment of the two shafts, as well as reduce vibration.

http://www.ebay.com/sch/i.html?_from=R40&_trksid=p2050601.m570.l1313.TR0.TRC0.H0.Xflexible+coupler&_nkw=flexible+coupler&_sacat=0

[richyrich_one]

You should really use flexible couplers - search ebay for "flexible coupler", as these reduce the stress load caused by any inertia jumps of the stepper and misalignment of the two shafts, as well as reduce vibration.

I did think about the alignment issue, and yes a flexible coupler would have been best but they don't appear to be available in the correct size as well as being quite large from what I've seen. As for the stress load from inertia, on a 10:1, IMHO I really don't think it will be much of an issue. I stand to be corrected of course. :-)

I will let you know how they work out when they arrive.

[AstroShomy]

I have modified my stepper motor 28BYJ-48-5V to get a bipolar version since I need more torque.

It works perfectly with the driver L293D, I tested it with a test app.

But, I am unable to use this Arduino code without modifications. Someone suggested me I should use AccelStepper library, since I am no longer using a unipolar stepper motor.

Is there anyone willing to point out what should be changed in the code? I am pretty new with Arduino.

Thanks,

M

[brown_rb]

I did think about the alignment issue, and yes a flexible coupler would have been best but they don't appear to be available in the correct size as well as being quite large from what I've seen. As for the stress load from inertia, on a 10:1, IMHO I really don't think it will be much of an issue. I stand to be corrected of course. :-)

I will let you know how they work out when they arrive.

There is a misconception that driving via the 10 speed reduction is safe. This is not such a good idea, as the mechanics and manufacture of the mechanisms employed mean that they are not robust enough to have that level of force (from stepper motors) consistently applied to them - over time this results in increased wear and tear with more backlash, or if the focuser limits are exceeded, the complete breakdown of the fine focus mechansism - in other words, not a good idea. If you need to drive the fine focus mechanism (FFM) then it would be better using a belt drive to prevent damage in the event of exceeding the focuser limits. 

Regards

[richyrich_one]

There is a misconception that driving via the 10 speed reduction is safe. This is not such a good idea, as the mechanics and manufacture of the mechanisms employed mean that they are not robust enough to have that level of force (from stepper motors) consistently applied to them - over time this results in increased wear and tear with more backlash, or if the focuser limits are exceeded, the complete breakdown of the fine focus mechansism - in other words, not a good idea. If you need to drive the fine focus mechanism (FFM) then it would be better using a belt drive to prevent damage in the event of exceeding the focuser limits. 

Regards

Does anyone else have an opinion on this? At the moment it looks like I might as well scrap the idea.

[SnakeyJ]

Does anyone else have an opinion on this? At the moment it looks like I might as well scrap the idea.

For what it's worth, I have found it very hard to differentiate true point of focus over about 6-7 steps driving my single speed OC1. Doubtless in really good seeing it will be easier to refine, but then how often are we blessed with that in UK? For my rig driving a 10:1 microfocuser would be overkill.

[Gina]

I use the 11:1 reduction drive on my Esprit - driven with a timing belt straight onto the knob.  With the MN190 and the ST80s I drive the main shaft because they don't have reduction drives The MN190 uses timing pulleys with about an 8:1 reduction and the ST80s use printed spur gears with 13:1 reduction ratio.  I use the 28BYJ-48 stepper motors driven from Arduino via ULN2003A drivers.

[richyrich_one]

Hi Dave

Not to sound like a backwards step but I would like to integerate your original manual control (Mk1) so that the focuser could be used without a laptop for quick observing sessions.

It would be useful to have both options available. It should be doable no?

If you can supply the Mk1 arduino sketch I could probably integrate it, I usually do VB6 but I could probably get my head round it. I thought I saw the MK1 stuff on your sourceforge page but I can't see it now.

Pretty please

[tekkydave]

Hi Dave

Not to sound like a backwards step but I would like to integerate your original manual control (Mk1) so that the focuser could be used without a laptop for quick observing sessions.

It would be useful to have both options available. It should be doable no?

If you can supply the Mk1 arduino sketch I could probably integrate it, I usually do VB6 but I could probably get my head round it. I thought I saw the MK1 stuff on your sourceforge page but I can't see it now.

Pretty please

I'll upload it to the SF site as soon as I can find the original code :-)

[richyrich_one]

I'll upload it to the SF site as soon as I can find the original code :-)

Cheers, much appreciated.

[tekkydave]

I have uploaded the Mark1 code and a schematic to SF https://sourceforge.net/projects/arduinoascomfocuser/files/Mark1

The sketch contains a basic ascom serial functionality. This can be ignored or removed if you dont want to connect the usb cable.

The 5-way switch I used is commonly used on electric guitars.

[richyrich_one]

I have uploaded the Mark1 code and a schematic to SF https://sourceforge.net/projects/arduinoascomfocuser/files/Mark1

The sketch contains a basic ascom serial functionality. This can be ignored or removed if you dont want to connect the usb cable.

The 5-way switch I used is commonly used on electric guitars.

Thanks, I'll see whatI can do.