Possible accel/decel bug while sending a G1 in MDI

[bmk5752]

I am assisting some students with setting up a cnc Plasma that they have designed. When we send a G1 command in MDI, the axis accel/decels very quickly. It does not accel/decel smoothly. The higher the accel time in PID the faster it moves or jumps. A G0 works just fine. Has anyone else experience this behavior?

Thanks,
Barrett

[ShawnM]

Please post a current report from the machine so we can see how it's set up. What drives are you using? What wiring diagram did you follow? We need much more info to be able to help.

[cnckeith]

i moved this post to the Plasma Forum.

like shawn said, please post a fresh report.zip so we can see your setup so we can help you out!

and photos and details of the machine and control components and configuration will go a long way in providing us with the information we need to provide technical support.

more info on how you can provide info we need to help is here
https://centroidcncforum.com/viewtopic.php?p=5010#p5010

[Joey]

Also, any changes should be done in the wizard for acorn plasma.

I would just do a fresh install and complete the setup in the wizard.

[bmk5752]

Good Morning,

I wanted to start off and apologize for not providing more information in my original post. It was done in haste, trying to get a possible answer before the students left for the weekend. I have attached the current report.

A little background; I have installed and programmed several Centroid controls ranging from CNC7 to CNC12 for use on custom 3 axis hydraulic cutting presses. It has been awhile however, and the first time using the plasma version. I am currently working in the Engineering Department at Penn State - Altoona and advising two students on their senior capstone project. They have redesigned a previous CNC Plasma cutter project. The CNC is being driven using rack and pinion on the x and y axis and a ball screw on the Z, with Gecko G251 drives and Nema 24 motors. There is only one motor per axis. I helped them wire the Acorn with not really using any one wiring diagram. We initially used the wizard to get one axis running, but have since just made changes using the Config screens.

The problem first happened when we were troubleshooting a rigidity issue with the gantry, which actually was mostly being caused by the fast jerky accel and decel of a G1 X10 F100 move in MDI. I put a ridiculously slow accel of 10 and ran another G1. The axis took further to get up to 100 ipm as expected, but ramped at the beginning and end of the move at a higher feedrate than the one set with G1. A G0 moves smoothly with correct accel/decel rates. As does the program in PID Config.

I hope this explains it a little better. Please let me know if you need anymore information.

Barrett

[Joey]

Thanks for the Report. This is not looking like a bug.

I'm Chalking this up to the Extremely Low Resolution with Smoothing.

Turn smoothing off in the Profile manager and try your tests again.

With the current V5.08 you can reduce Parameter 223 from 1000 down to around 100.
This will Prevent Smoothing from starving for vectors due to the Low Pulse rate.

The Slower the feed rate the Lower the pulse rate. This causes Smoothing to act up at slow feed rates ON extremely Low Resolution systems.

You may find that currently for feed rates about say, 250ipm smoothing works great but much below 200ipm smoothing starts creating extra vectors hence the extra acceleration.

The Latest 5.1 RC has Improved smoothing for Low Resolution Systems.

Let me know what yinz find out.

[bmk5752]

Good Morning,

Just an update. They are able to get stable Linear moves with Smoothing turned off.

Thanks for the help,
Barrett