Accuracy of the NEMA 17 0.9°16micro-steps & THK KR2001A
Moderators: rjlittlefield, ChrisR, Chris S., Pau
Hello Rik,
OK, firstly I have improved the mounting of the camera :-)
The pattern dependent on the 16 micro-steps looks like follows:
https://farm8.staticflickr.com/7883/470 ... d5b3_o.jpg
- 1: long
- 2-13: OK
- 14: long
- 15,16: short opposite direction
BTW, maybe everything should be shifted by 1 because I have hidden above the extremely large first step:
https://farm8.staticflickr.com/7902/321 ... c9ca_o.jpg
BR, ADi
OK, firstly I have improved the mounting of the camera :-)
The pattern dependent on the 16 micro-steps looks like follows:
https://farm8.staticflickr.com/7883/470 ... d5b3_o.jpg
- 1: long
- 2-13: OK
- 14: long
- 15,16: short opposite direction
BTW, maybe everything should be shifted by 1 because I have hidden above the extremely large first step:
https://farm8.staticflickr.com/7902/321 ... c9ca_o.jpg
BR, ADi
- rjlittlefield
- Site Admin
- Posts: 23625
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Interesting!
Now you have an unmistakable repeating pattern of 16 microsteps, with a little random noise superimposed on it. This now looks to me like you're getting high quality data about device behavior.
I would characterize this pattern by saying that in each group of 16, the first and last microsteps move a hair backward instead of forward, and those errors are roughly compensated by an unusually large forward movement next to each of the slight backward moves.
So, if you were to step by multiples of 2 microsteps, with the sequence aligned so as to split up the pair of backward microsteps, it looks to me like you'd get a pretty uniform sequence of ~0.0003 mm movements.
--Rik
Edit: fix units, "micron" --> "mm"
Now you have an unmistakable repeating pattern of 16 microsteps, with a little random noise superimposed on it. This now looks to me like you're getting high quality data about device behavior.
I would characterize this pattern by saying that in each group of 16, the first and last microsteps move a hair backward instead of forward, and those errors are roughly compensated by an unusually large forward movement next to each of the slight backward moves.
So, if you were to step by multiples of 2 microsteps, with the sequence aligned so as to split up the pair of backward microsteps, it looks to me like you'd get a pretty uniform sequence of ~0.0003 mm movements.
--Rik
Edit: fix units, "micron" --> "mm"
Last edited by rjlittlefield on Thu Feb 14, 2019 1:18 pm, edited 1 time in total.
- rjlittlefield
- Site Admin
- Posts: 23625
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Hello Chris,
The average values look like follows:
https://farm8.staticflickr.com/7899/401 ... ec7a_o.jpg
BR, ADi
The average values look like follows:
https://farm8.staticflickr.com/7899/401 ... ec7a_o.jpg
BR, ADi
Last edited by Adalbert on Thu Feb 14, 2019 1:15 pm, edited 1 time in total.
-
- Posts: 3439
- Joined: Sat Nov 20, 2010 10:40 am
- Location: Santa Clara, CA, USA
- Contact:
- rjlittlefield
- Site Admin
- Posts: 23625
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
Hello Ray,
BTW, I haven’t measured the backlash of this rail because I’ll never start the stack in opposite direction.
And in order to avoid the problem of the first steps I’m moving the rail ( two dummy-steps ) in the direction of the stack at the beginning.
http://www.photomacrography.net/forum/v ... c&start=15
BR, ADi
Yes, this should be the reason.Is that large negative first step due to a backlash compensation?
BTW, I haven’t measured the backlash of this rail because I’ll never start the stack in opposite direction.
And in order to avoid the problem of the first steps I’m moving the rail ( two dummy-steps ) in the direction of the stack at the beginning.
http://www.photomacrography.net/forum/v ... c&start=15
BR, ADi
Hello everybody,
This time one micron with TMC5130.
https://www.photomacrography.net/forum/ ... c&start=45
But I haven’t found the optimal settings for the TMC up to now.
Probably the result can be improved.
BR, ADi
This time one micron with TMC5130.
https://www.photomacrography.net/forum/ ... c&start=45
But I haven’t found the optimal settings for the TMC up to now.
Probably the result can be improved.
BR, ADi
How do you get these measurements from Zerene? Is there some output from a stack you can export with the estimated offset between each photo?Adalbert wrote:Hello everybody,
This time one micron with TMC5130.
https://www.photomacrography.net/forum/ ... c&start=45
But I haven’t found the optimal settings for the TMC up to now.
Probably the result can be improved.
BR, ADi
- Cam
Hi Kaleun96,
Yes, console log.
Please take a look at this thread:
https://www.photomacrography.net/forum/ ... highlight=
BR, ADi
Yes, console log.
Please take a look at this thread:
https://www.photomacrography.net/forum/ ... highlight=
BR, ADi
- rjlittlefield
- Site Admin
- Posts: 23625
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
For users with Prosumer or Professional licenses, File > Save Other > Save Registration Parameters will write a tab-separated-values file that will load directly into Excel.kaleun96 wrote:Is there some output from a stack you can export with the estimated offset between each photo?
The file includes the XOffset, YOffset, Scale, and Rotate values for each frame, along with the file name and some other values.
With Student and Personal Edition, the same information can be extracted less conveniently from either the console log or a saved project file.
No matter how you get them, the meaning of these numbers is explained at https://www.photomacrography.net/forum/ ... hp?t=21508 .
For high precision measurements in the micron range, the best method is to mount a patterned target on the rail and track its movement by looking from the side using a high magnification macro setup, as described at https://www.photomacrography.net/forum/ ... hp?t=11519 ("Tiny focus steps: how to make them, how to measure them"), with other applications shown at https://www.photomacrography.net/forum/ ... hp?t=16323 and https://www.photomacrography.net/forum/ ... hp?t=27549 .
Useful information can also extracted from an ordinary stack, in which case XOffset and YOffset indicate lateral stability such as nutation of the carriage. You can also use frame-to-frame changes in the Scale parameter as an indicator of step size. Using scale as an indicator for step size is much less precise than the look-from-the-side approach, particularly with ordinary photographic subjects rather than specially designed test targets. However, it is almost always available (doesn't work with telecentric optics) and it can yield useful information about system stability and accuracy with no extra setup.
--Rik