Main mirror focusing with three motors via Ascom

[Oliver Schneider]

Hello
my name is oliver and i run an observatory together with a friend.
I have already implemented three of the Ascom focusers (version 1) for myself privately and I am very impressed with the system.
Thanks a lot for this !!!

Now my friend and I have a problem that we hope to solve with this great tool.

Description of the situation:
We operate a remote observatory in France.
A large telescope is now to be set up there.
The telescope does not have a focus drive in the primary focus.
It is focused on the main mirror cell by three micro stepper motors. You can also use the three motors to adjust the telescope using the primary mirror cell.

As of today:
The whole thing is currently running manually via a 4-axis motor control for a CNC milling machine with GRBL software (GCode sender).
Motorshield A4988 with Nema 17PG27 motors and 1:27 gearboxes.
As far as we know, the 4-axis motor control used is not supported by the Arduiono Ascom software.
See the Files

Problem:
To focus, it is necessary that all three motors are addressed at the same time. At the moment this is done manually using a macro in the GCode sender software.

Explanation:
As far as we know, the existing Arduino Ascom software currently only supports the addressing of a motor in an Ascom instance.
If only one motor is addressed, the adjustment of the main mirror is changed.
Even the existing system, which enables two Ascom instances to run, is of no use in this case, because two motors would then be addressed, but can be controlled by different software and thus not at the same time.

Solution:
In order to get automatic focusing via Voyager, APT, Nina etc., it would be very good if the Ascom interface could address not just one motor but all three motors (or more ...) at the same time.

Our question:
Is such a thing possible?
So that the Arduino Ascom software outputs its commands to more than one motor, which then all proceed in the same way.

Thinking into the future ... you could expand such a system so much that e.g. with APT, Nina, Voyager when focusing via an image analysis, a misalignment of the telescope is determined and then automatically adjusted and focused using "auto adjustment", if the motors could be controlled individually.

Best Regards
Oliver

[brownrb]

Anything is possible, given time money and resources. I am just going to write down my thoughts about this. And yes there is no support in myfocuserpro projects to a cnc 4-axis board.

But there is always another way to look at things.

Solution:
In order to get automatic focusing via Voyager, APT, Nina etc., it would be very good if the Ascom interface could address not just one motor but all three motors (or more ...) at the same time.

rather than having a focuser control three separate motors that need to move in unison

why not do this

connect the 3 motors in series so the controller sends a pulse which goes to all 3 - we do this with printers now, for example the dual z axis drives on a lot of 3d printers wire the two z motors in series but it works. All motors would move at the same time,

You would need to wire it correctly, so that each motor moves in the same direction for each pulse. But they would move in unison.

One might need to up the voltage to the motors a bit, but it should work. So you don't need 3 controllers, only 1.

You also mentioned
*Thinking into the future ... you could expand such a system so much that e.g. with APT, Nina, Voyager when focusing via an image analysis, a misalignment of the telescope is determined and then automatically adjusted and focused using "auto adjustment", if the motors could be controlled individually. *

There has been a lot of work done in many areas, for example

using stepper motors to control alt-az and thus being able to polar align automatically

moving the secondary cell on a celestron SCT to focus when in hyperstar mode.

I think the most impressive thing I ever saw (and owned for that matter) was the Orion steady star adaptive system. That used micro steppers to control the image in real-time (around 40 times per second) That would have to be my all time favourite and has absolutely out of the ball park performance.

That is not a digression, but simply pointing out that where things may not have been possible 20 years ago, many of them are now in reach.

I think what you were saying about controlling the 3 motors individually one could automate misalignment by adjustment. I would think that was more about collimation and making the image central to the focal plane. That is different to focusing - which is about moving the entire image plane forward/backward to achieve focus.

So in that instance you would need to control each motor separately. there would always be some lag between the motor steps, not all would move at eactly the same time, but then you use a "delay after move" feature to allow things to settle before an image took place. So one could make such a controller with three drivers chips but use a single protocol that the move to each stepper is in a single command, move in 300, would be sent to each driver chip etc

As something gets more complex, so does the design and code.
To drive a stepper using A4998 requires step, dir, enable (if one hard wires the ms0-ms1 etc lines that control stepping modes), so 3 motors * 3 pins = 9 pins in total (minimum) requied for moving the stepper motors. So yes, it would be possible to implement an arduino + 3 driver chips controlling 3 stepper motors and using 1 ascom interface to handle the focusing. Any collimation would need to be done using a windows app talking to the controller.

R

[Oliver Schneider]

Hello Mr. Brown
Thank you for your fast reply.

I was able to solve the problem with the control thanks to your tip. All three motors now run syncronously forwards and backwards.

We will adjust the telescope manually using the GRBL software.
it's been going well for years and it's not complicated either

Greetings and have a nice fourth Advent
Oliver