Testing for Max Speed

General Discussion
1

Basic Information:

Printer Model: Sovol SV06+
MCU / Printerboard: Manta E3EZ
Host / SBC: CB2
klippy.log

klippy.log (91.4 KB)

Issue:
…

I am using the above to determine max acceleration and speed of X and Y. I have slowed the printer all the way down to 50 for max speed and 100 for max acceleration and am still losing steps!

I imagine I have something setup wrong in my cfg. Could someone please look it over and see if I have something setup wonky?

2

First off, could you be more descriptive in what your problem is? What do you mean when you say you are “losing steps” and how do you know?

Along with the TEST_SPEED macro from Elli’s tuning page, you have a lot of macros running in your system.

The first thing I recommend to somebody having problems who has a lot of macros is to take them all out and see if the problem goes away and, if it does go away, then start restoring the macros until you find the one that is causing the problem.

3 Likes
3

After running Elli’s Test_Speed the starting MCU position and the ending MCU position are at minimum 4-6 microsteps apart. From what I understand from Elli’s instructions is that the beginning and ending MCU coordinates should be less than 1 microstep apart.

I can provide a picture tomorrow if needed.

I’ll disable macros and retest tomorrow as well.

1 Like
4

Hi @JimmieNeu ,

It looks like you’re using 64 microsteps on X and Y. Stepper motors only lose full steps, not microsteps, so as long as your values are no more than 64 off in either direction, you haven’t lost steps.

Screenshot 2025-01-26 at 6.32.40 AM
Screenshot 2025-01-26 at 6.32.40 AM1408×588 72.4 KB

5

My understanding is a bit different:

  • We have to differentiate between “digital steps” as produced by Klipper and mechanical steps as executed by the motor.
  • On a typical bipolar stepper, it takes 4 digital steps to move the rotor one stable mechanical step.
  • A stepper can only lose mechanical steps but not digital ones.
  • Assuming no microstepping, the information in Klipper would have to indicate 4 lost digital steps before you actually have a lost step.
  • Taking microstepping into account, it means you have to lose 4 \times \text{microstepping_resolution} to lose one mechanical step.

I’d happily stand corrected if this is wrong.

3 Likes
6

20250126_123855
20250126_1238551920×2560 639 KB

With none of the macros disabled.

20250126_124327
20250126_1243271920×2560 451 KB

With all the macros disabled.

The numerical differences are nearly identical at about 256… or 4 steps.

(Sorry for the second picture being so crooked)

7

Do you know, that you can Copy&Paste from the Mainsail/Fluidd console? There is no need of making photos.

1 Like
8

Good to know!

9

Increasing Y run_current to 1.2 resolved the issue for Y, but not for X