input shaping. Feature Request

More
15 Oct 2023 04:21 - 15 Oct 2023 20:30 #283031 by pinder
Hello from Pinder.

I was looking at how much difference Klipper with input shaping make in 3d printing.
I was hoping this would be a nice feature for Routers Lasers and plasma Cutter.
Klipper uses a USB accelerometer for calculating resonance freq.
​​​​

thank you in advance

Pinder
Last edit: 15 Oct 2023 20:30 by pinder.

Please Log in or Create an account to join the conversation.

More
15 Oct 2023 12:38 #283067 by tommylight
Yes it does, it just avoids the resonant speeds by lowering feed rate.
3D printers move at very fast rate with huge accelerations and use belts that get stretchy at those rates, so for us it wont change much, if anything.
It works for X and Y axis only, we have a bit more, well plenty more when we count in the joints.
Mills and lathes have other vibration issues that input shaping can not fix.

Please Log in or Create an account to join the conversation.

More
15 Oct 2023 17:06 - 15 Oct 2023 17:07 #283080 by pinder
Hey Tom
I agree this is not much useful for millings.
I was looking for the opportunities it has in rapid motions for co2 lasers while engraving.
Even plasmas and Fiber lasers can use rapid move movements when not cutting.
Last edit: 15 Oct 2023 17:07 by pinder.

Please Log in or Create an account to join the conversation.

More
15 Oct 2023 19:33 #283087 by pinder

Yes it does, it just avoids the resonant speeds by lowering feed rate.
 

​I think it does more than just feed rate. It shapes the accelrations profile to compensate for resonance.
Huge scope for decreaseing production times.
​​​​​​

Please Log in or Create an account to join the conversation.

More
15 Oct 2023 21:10 #283092 by rodw
Replied by rodw on topic input shaping. Feature Request

​I think it does more than just feed rate. It shapes the accelrations profile to compensate for resonance.
Huge scope for decreaseing production times.
​​​​​​

Sounds you are really talking about limited jerk or S curve accelleration. That is a huge can of worms.

Please Log in or Create an account to join the conversation.

More
17 Oct 2023 11:32 #283168 by snowgoer540

Hello from Pinder.

I was looking at how much difference Klipper with input shaping make in 3d printing.
I was hoping this would be a nice feature for Routers Lasers and plasma Cutter.


I've had this same thought. I think it definitely has application here, especially where there is no real interaction between the cutting head and the material (e.g. plasma, laser, etc.). I've mentioned it before outside of the forum. I briefly looked into the math Klipper employs, and then got side tracked. It would require updating/rewriting at least portions of the trajectory planner.

There's also a whole other component necessary to measure the resonance in the system.

It would definitely be fun to test. Especially for changes in direction with my Y axis, it would be pretty easy to see if it made any difference.
The following user(s) said Thank You: tommylight, Clive S, pinder

Please Log in or Create an account to join the conversation.

More
17 Oct 2023 13:05 #283176 by pinder
Klipper and now Marlin have their ways of measuring resonance frequency. 
They have implemented input shaping and they are open-source. I dont know how much usable their code would be though.

Please Log in or Create an account to join the conversation.

More
17 Oct 2023 13:38 #283178 by snowgoer540
I looked into it briefly one day. I got side tracked because they do all of their trajectory planning in Python which I thought was interesting.

The project is very well documented. I would think between the documentation and the code to look at it could be ported over for us to at least try. The measurement portion may just basically drop in and be usable. Id have to find time to do digging.

Please Log in or Create an account to join the conversation.

More
11 Mar 2024 19:37 #295673 by egemens
I have the exact idea since i upgraded to Klipper and tuned my 3d printer to use input shaping and pressure advance.

On my cnc router, especially sudden stops cause quite a bit of vibration and flex because i think the gantry is a bit heavy. I minimize it with accel/decel. but that results in slower motion in short moves as it should be long enough to reach full speed. I would like to use the full potential of the motors. so i think there can be a potential if vibrations are cancelled by an opposing wave like in noise cancelling headphones. 

Please Log in or Create an account to join the conversation.

Time to create page: 0.088 seconds
Powered by Kunena Forum