tinyG Settings

[aldenhart]

Thanks for the great response.

I've been wondering if it makes sense to change the input and display of this value to a number that's divided by a million. On some machines I've set jerk as high as 100 billion. It's a lot of zeros to manage. It might be easier to set it to 100000 for 100 billion. Also, the newer version of the code allow numbers to have embedded commas, so it can be entered as $xjm=100,000,000,000 Obviously this doesn't work for JSON as the commoa is a field separator, but in text mode it's fine.

Thoughts on dividing the number by 1 million?

Alden

[cvoinescu]

What if, instead of mm/min^3, jerk would be expressed in mm/s^3, or even m/s^3? That would scale it by 216,000 or 216,000,000, respectively. The m/s^3 units are more intuitively understood, too: 1 m/s^3 jerk varies acceleration by about a tenth of a gee per second.

[JuKu]

Divide by million or even by billion - or change the units as cvoinescu suggest. But do make the numbers smaller, please.

[aldenhart]

FYI - I have posted example Shapeoko settings for TinyG here. These are a good starting point for a TinyG Shapeoko setup. Please let me know if you have any comments or corrections.

https://github.com/synthetos/TinyG/wiki ... eoko-Setup

[chambrot]

Hello, i am trying to get my shapeoko calibrated.

Not sure how do go about doing this...I want to figure out how much my travel is per motor revolution. This is whats happening:

1. I set a ruler down with a marker at 0mm.

2. I send a command for a 100mm move on the X-axis...its very close.

3. I send a command for another 100mm move on the X-axis (should not be at 200mm)...it overshoots by very little.

4. I send a command for a bigger distance, move to 450mm...it over shoots by even more.

Seems that the further I travel the worse the accuracy gets. I know there is an formula for figuring this out somewhere, but can't seem to find it.

Also is there a command to send to the TinyG to tell the motors to move 1 revolution or 200 steps? this way I can use a caliper to see how much it actually travels when the motor turns 1 revolution.

Thanks in advance!

[JuKu]

You need to know your pulleys and belts. When you do, just do the math: For example GT2 belt has 2mm pitch. Assuming a 20 teeth pulley, that works out as 40mm per revolution. Ask for 400mm travel. That is 10 revolutions. If you got 400.5mm, tell the TinyG that one revolution is 40.05mm.

[WillAdams]

The basics of this are on the wiki: http://www.shapeoko.com/wiki/index.php/Calibration (see also http://www.shapeoko.com/wiki/index.php/ ... figuration for the under-lying math and http://www.shapeoko.com/wiki/index.php/Tuning)

twforeman did a excellent blog post on this linked from the first page: http://timf.anansi-web.com/wp/shapeoko-accuracy-really/

[chambrot]

Got it!! It was 35.956mm per rev.

Had to fine tune that number a bit, but its pretty close now...with a ruler at least. Gotta try with something more precise.

Thanks!!!

[aldenhart]

One further note: The travel-per-revolution ($1tr, $2tr...)for Shapeoko is 40.0. The pulleys are different than the Shapeoko1's that were discussed when the previous entries were written.

The GT2 pulley has 20 teeth at 2mm spacings. Hence 40 mm per revolution.