3D Printer Input Shaper Calculator
Enter the ripple measurements from a ringing tower to calculate X and Y resonance frequencies, optional fine-tuning values, and firmware-ready input shaper settings.
Recommended maximum acceleration (mm/s2)3,000
- Measured X resonance frequency (Hz)
- 49.42
- Measured Y resonance frequency (Hz)
- 46.15
- X fine-tuning START and FACTOR
- START=31.08, FACTOR=0.75
- Y fine-tuning START and FACTOR
- START=29.02, FACTOR=0.7
- Fine-tuned X frequency (Hz)
- 47.93
- Fine-tuned Y frequency (Hz)
- 44.76
- Ringing-limited acceleration (mm/s2)
- 3,000
- Smoothing-limited acceleration (mm/s2)
- 3,000
- Firmware settings
- [input_shaper] shaper_type_x: mzv; shaper_freq_x: 47.93; shaper_type_y: mzv; shaper_freq_y: 44.76
Axis frequency summary
| Axis | Measured frequency | Sweep START | Sweep FACTOR | Fine-tuned frequency |
|---|---|---|---|---|
| X | 49.42 | 31.08 | 0.75 | 47.93 |
| Y | 46.15 | 29.02 | 0.7 | 44.76 |
Fine-tuned frequencies use the optional sweep band entries; if you did not run that sweep, use the measured frequencies instead.
If ripple spacing was inconsistent, repeat the test because the marks may not represent one clean resonance.
How to use this calculator
- Choose Klipper or Marlin so the output uses the right setting syntax.
- Enter the constant outer-wall speed used for the ringing tower.
- Count complete ripple periods and measure the matching span for each axis.
- Enter the acceleration tower bands that still had acceptable ringing and smoothing.
- Use the generated frequencies and maximum acceleration as a starting point, then confirm with a test print.
How the input shaper values are calculated
A ringing tower turns visible ripple spacing into a resonance frequency. For each axis, the calculator multiplies the constant wall speed by the number of complete oscillations, then divides by the measured ripple span:
F = speed * oscillations / measured span
The result is in hertz when speed is in mm/s and the span is in mm. The same calculation is done separately for X and Y because a bedslinger, CoreXY, or toolhead can have different moving masses and stiffness on each axis.
Optional fine tuning
Klipper documents an optional frequency sweep for refining the first ringing measurement. The sweep command uses:
START = F * 83 / 132
FACTOR = F / 66
After printing that sweep, enter the one-based band with the least ringing. The refined frequency is:
Frefined = F * (39 + 5 * band) / 66
If you did not run the optional sweep, use the measured X and Y frequencies instead of the fine-tuned rows. Klipper notes that the extra sweep is unnecessary for most printers.
Acceleration limit from the tower
The acceleration tower bands are also one-based. Band 1 is the starting acceleration, and each later band adds one step:
Aband = Astart + (band - 1) * Astep
The calculator finds the acceleration from the last acceptable ringing band and from the last band before the small smoothing gap became enlarged. The recommended maximum acceleration is the lower of those two numbers, so a visibly smoothed corner can lower the result even when ringing still looks acceptable.
What changes the result most
The measured span drives the frequency. Counting too many or too few ripple periods, measuring from mismatched points on the waves, or including extrusion artifacts will move the answer quickly. Low frequencies, especially below about 20 to 25 Hz, usually point to a mechanical issue or a heavy moving assembly before they point to a firmware problem.
What this calculator leaves out
This tool cannot tell whether the marks are true ringing, multiple resonance peaks, belt artifacts, loose hardware, pressure advance error, or inconsistent extrusion. It also does not replace accelerometer-based resonance measurement when a printer has more than one strong peak on an axis. Maximum acceleration still depends on motor torque, belt grip, moving mass, extrusion flow, and the printer's own mechanical limits.
Worked example
Suppose a ringing tower was printed at 100 mm/s. On X, 6 oscillations span 12.14 mm, so the measured frequency is 100 * 6 / 12.14 = 49.42 Hz. On Y, 6 oscillations span 13 mm, so the measured frequency is 46.15 Hz.
The optional Klipper sweep values are START=31.09 and FACTOR=0.75 for X, and START=29.04 and FACTOR=0.70 for Y. If band 5 looks best on both axes, the refined frequencies are 47.92 Hz for X and 44.76 Hz for Y.
With an acceleration tower starting at 1,500 mm/s2 and adding 500 mm/s2 per band, band 4 maps to 3,000 mm/s2. If both ringing and smoothing remain acceptable through band 4, the recommended maximum acceleration is 3,000 mm/s2.
Common questions
How do I count oscillations on a ringing tower?
Count complete repeating ripple periods, not every light and dark mark. Measure from the same point on the first ripple to the same point on the last ripple, such as peak to peak, across the counted periods.
Which side of the tower is X and which side is Y?
Use the labeling or motion directions from the test model and firmware instructions you printed. If you are unsure, command a small X move and a small Y move before measuring so the physical faces match the axes you will configure.
Should I use ZV, MZV, EI, or a hump EI shaper?
Use the shaper type selected from the test procedure for your printer. MZV is often a practical starting point in Klipper, while EI and hump EI shapers can reduce sensitivity at the cost of more smoothing. Marlin M593 integrated input shaping uses ZV syntax, so non-ZV Klipper choices do not change the Marlin commands.
Why is a low resonance frequency a warning?
A low resonance frequency often means the moving system is heavy or not stiff enough. Klipper recommends checking belt tension, frame rigidity, toolhead mounting, and moving-bed mass when measured frequencies are low.
Do I need the optional fine-tuning sweep?
Usually no. Klipper states that the initial ringing measurement is enough for most printers. Run the sweep when you want to refine a manual measurement or compare nearby frequencies after the first result.
Should I retune after changing printer parts?
Yes, retune after changes that affect stiffness or moving mass. Belt tension, toolhead weight, hotend mounts, frame braces, bed plates, and major speed profile changes can all move the resonance frequency.