I'm just about done setting everything up with my new focuser but one thing
that I can't find an answer for, or maybe don't understand the answer, is in
the Windows software myFP2W v2.5.1.3 on the "motor" tab there is the Step
Mode setting FULL, HALF etc. and beside that is the Step Size. I'm using a
28BYJ-48 gear stepper modified to be bipolar (2 separate coils) and from
what I can deduce it natively has 2048 steps per rev, but I am half stepping
so do I set the "Step Size" to the FULL STEP or HALF STEP value of microns?
My CFZ is 79u and with the 2048 steps step size is 8.54u and half step is
4.27u.
Thanks!
PS, for anyone else using this great little motor, here is a dissected description and analysis.
Hi Paul
Thanks for the Q. I will probably go overboard with this reply, because it presents me with an opportunity to write down my thoughts and opinions, as usual, take everything with a dose of salt.
My answer about which to use when using half steps is at the end of this message.
Stepmode relates to the number of steps the motor will move over a defined measurement, like 1 revolution. This is relatively unimportant. What is more important is what the distance of the focuser will move when when 1 motor step occurs. We look at a mechinal setup where rotational motion gets translated to a horizontal motion (in and out). When step mode changes it will also change stepsize (they are related, and we use it in focuser basics to do some basic math)
Obviously stepsize is the "distance" a motor step will move.
Stempode is a setting you set, like 1:1 or 1:2
Stepsize is something you have to calculate and have to enter in. This is important because client apps dont want to know stepmode. That tells those apps nothing.
What is more important is stepsize. From that value client apps can calculate a lot of useful data.
Those apps like NINA also ask you about the telescope, Focal Length and Focal Ratio. They can then calculate steps in CFZ from stepsize and Focal ratio, then compare this with your cameras pizel size, resolution (number of pixels H and W), and determine things like oversampling and best settings for camera. Then of course PHD etc use it to determine whether to bin the pixels of the guider cam for better sampling.
This also shows up in ASCOM - A focuser driver must support stepsize or generate an exception in the stepsize property is not supported by the focuser. But stepmode is not mentioned in the ASCOM standard for focusers.
For the focuser basics - one calculates the CFZ and then does the math to determine what is required to get approx 10 motor steps within the CFZ. Looking at that we see that if we only have say 5 steps in the CFZ using stepmode FULL. We then half step the motor and so the number of steps within the CFZ doubles.
Lets say the stepmode is FULL and step size works out to be 24 microns, with a CFZ of 240 microns. This means 240/24 = 10 full steps in CFZ.
Now lets change the stepmode to HALF. Each motor step now has a stepsize of 24/stepmode = 12 microns. The CFZ does not change because it relates to Focal Ratio. We now have 240/12 = 20 steps within the CFZ.
You can see that changing one affects the other. To increase steps in CFZ we can use stepmode to give us more steps. As stepmode increases, the distance amount travelled (stepsize) for each motor step decreases by half each time. So
1.1 Stepmode = x (the distanc taken by 1 step)
1/2 Stepmode = Stepsize / 2
1/4 stepmode = stepsize / 4
giving Stepsize = x / 2
What we do in focuser basics is to do the math the will result in enough steps with CFZ using a motor that is running in full step mode. Why? Because full step mode gives the greatest torque. If we said 1/4 or 1/8 was fine that would be an error because at 1/8 stepmode there is little chance of pulling an imaging train weighting 4kg up with the telescope pointing at the Zenith.
With focusing I have found more is best. No point going to all that trouble and time ending up with something that disappoints. Is this overkill? I would counter that by saying do you want to be disappointed or do you want to be happy. I dont do much visual, so I have an imaging train that meets the requirements and doesnt drop my $2000+ camera and other gear on the floor.
Spending $50US on a geared stepper to make things work well and reliably I think is money well spent.
Back to basics though, many users do visual or do not the requirements I do. That is okay. I give the basics and you decide how to do things.
One caveat. When we work with a gear train using grooved belts things are a little different. geared belt systems are a method of creating more torque and reducing stepsize.
Okay back to your question now.
but I am half stepping
so do I set the "Step Size" to the FULL STEP or HALF STEP value of microns?
My CFZ is 79u and with the 2048 steps step size is 8.54u and half step is
4.27u.
Remember that apps need to know what stepsize actually is. Its also a reason why I say over and over - set stepmode once and DO NOT CHANGE IT. If I decided to change step mode (without changing step size setting) then I have stuffed up any app that connects (because it is still using stepsize as before), AND I have stuffed up both Position and MaxStep at the same time.
The answer is
79/8.54 = 9.2 steps
79/4.27 = 18 steps
So the fulls steps is better. The half step is basically over sampling. If you did some focus curves in NINA for each you will find full is better, whilst half will give you a softened flat bell curve showing multiple samples in the bottom of the curve, which leads to poor focusing, because NINA cant get a sample that gets the best FWHM (or which settings you use) because at the bottom of the curve it has multiple steps that give the same value.
Regards
Robert
I attach a few focu curses that illustrate a good setup. Many thanks to Francisco Vitale for sharing his Focus Curves
Wow Robert! Thank you for the detailed reply, I really appreciate it and I hope many others will see it and benefit. You made it very easy to understand the whole concept.
Thank you.
BTW 79/8.54 = 9.25 but I assume that is still a good number of steps.
my maths gets worse as my age increases. a bit like the inverse square rule that applies to satellites and signal strength vs distance. Or I can just blame it on the MS calculator that I used.
😄
1
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm just about done setting everything up with my new focuser but one thing
that I can't find an answer for, or maybe don't understand the answer, is in
the Windows software myFP2W v2.5.1.3 on the "motor" tab there is the Step
Mode setting FULL, HALF etc. and beside that is the Step Size. I'm using a
28BYJ-48 gear stepper modified to be bipolar (2 separate coils) and from
what I can deduce it natively has 2048 steps per rev, but I am half stepping
so do I set the "Step Size" to the FULL STEP or HALF STEP value of microns?
My CFZ is 79u and with the 2048 steps step size is 8.54u and half step is
4.27u.
Thanks!
PS, for anyone else using this great little motor, here is a dissected description and analysis.
https://cookierobotics.com/042/
Last edit: PaulW 2024-09-26
Hi Paul
Thanks for the Q. I will probably go overboard with this reply, because it presents me with an opportunity to write down my thoughts and opinions, as usual, take everything with a dose of salt.
My answer about which to use when using half steps is at the end of this message.
Stepmode relates to the number of steps the motor will move over a defined measurement, like 1 revolution. This is relatively unimportant. What is more important is what the distance of the focuser will move when when 1 motor step occurs. We look at a mechinal setup where rotational motion gets translated to a horizontal motion (in and out). When step mode changes it will also change stepsize (they are related, and we use it in focuser basics to do some basic math)
Obviously stepsize is the "distance" a motor step will move.
Stempode is a setting you set, like 1:1 or 1:2
Stepsize is something you have to calculate and have to enter in. This is important because client apps dont want to know stepmode. That tells those apps nothing.
What is more important is stepsize. From that value client apps can calculate a lot of useful data.
Those apps like NINA also ask you about the telescope, Focal Length and Focal Ratio. They can then calculate steps in CFZ from stepsize and Focal ratio, then compare this with your cameras pizel size, resolution (number of pixels H and W), and determine things like oversampling and best settings for camera. Then of course PHD etc use it to determine whether to bin the pixels of the guider cam for better sampling.
This also shows up in ASCOM - A focuser driver must support stepsize or generate an exception in the stepsize property is not supported by the focuser. But stepmode is not mentioned in the ASCOM standard for focusers.
For the focuser basics - one calculates the CFZ and then does the math to determine what is required to get approx 10 motor steps within the CFZ. Looking at that we see that if we only have say 5 steps in the CFZ using stepmode FULL. We then half step the motor and so the number of steps within the CFZ doubles.
Lets say the stepmode is FULL and step size works out to be 24 microns, with a CFZ of 240 microns. This means 240/24 = 10 full steps in CFZ.
Now lets change the stepmode to HALF. Each motor step now has a stepsize of 24/stepmode = 12 microns. The CFZ does not change because it relates to Focal Ratio. We now have 240/12 = 20 steps within the CFZ.
You can see that changing one affects the other. To increase steps in CFZ we can use stepmode to give us more steps. As stepmode increases, the distance amount travelled (stepsize) for each motor step decreases by half each time. So
1.1 Stepmode = x (the distanc taken by 1 step)
1/2 Stepmode = Stepsize / 2
1/4 stepmode = stepsize / 4
giving Stepsize = x / 2
What we do in focuser basics is to do the math the will result in enough steps with CFZ using a motor that is running in full step mode. Why? Because full step mode gives the greatest torque. If we said 1/4 or 1/8 was fine that would be an error because at 1/8 stepmode there is little chance of pulling an imaging train weighting 4kg up with the telescope pointing at the Zenith.
With focusing I have found more is best. No point going to all that trouble and time ending up with something that disappoints. Is this overkill? I would counter that by saying do you want to be disappointed or do you want to be happy. I dont do much visual, so I have an imaging train that meets the requirements and doesnt drop my $2000+ camera and other gear on the floor.
Spending $50US on a geared stepper to make things work well and reliably I think is money well spent.
Back to basics though, many users do visual or do not the requirements I do. That is okay. I give the basics and you decide how to do things.
One caveat. When we work with a gear train using grooved belts things are a little different. geared belt systems are a method of creating more torque and reducing stepsize.
Okay back to your question now.
but I am half stepping
so do I set the "Step Size" to the FULL STEP or HALF STEP value of microns?
My CFZ is 79u and with the 2048 steps step size is 8.54u and half step is
4.27u.
Remember that apps need to know what stepsize actually is. Its also a reason why I say over and over - set stepmode once and DO NOT CHANGE IT. If I decided to change step mode (without changing step size setting) then I have stuffed up any app that connects (because it is still using stepsize as before), AND I have stuffed up both Position and MaxStep at the same time.
The answer is
79/8.54 = 9.2 steps
79/4.27 = 18 steps
So the fulls steps is better. The half step is basically over sampling. If you did some focus curves in NINA for each you will find full is better, whilst half will give you a softened flat bell curve showing multiple samples in the bottom of the curve, which leads to poor focusing, because NINA cant get a sample that gets the best FWHM (or which settings you use) because at the bottom of the curve it has multiple steps that give the same value.
Regards
Robert
I attach a few focu curses that illustrate a good setup. Many thanks to Francisco Vitale for sharing his Focus Curves
Last edit: brownrb 2024-09-26
Wow Robert! Thank you for the detailed reply, I really appreciate it and I hope many others will see it and benefit. You made it very easy to understand the whole concept.
Thank you.
BTW 79/8.54 = 9.25 but I assume that is still a good number of steps.
Last edit: PaulW 2024-09-26
my maths gets worse as my age increases. a bit like the inverse square rule that applies to satellites and signal strength vs distance. Or I can just blame it on the MS calculator that I used.