Fill vs. line cutting. Results in different dimensions!

FabCreator · 2 March 2021 10:59

LightBurnSoftware/Documentation/blob/master/ScanningOffsetAdjustment.md

[Return to main page](README.md)

----

# Scanning Offset Adjustment

Modern lasers are capable of moving very fast, and with remarkable precision, however firing the beam still takes time.  Some power supplies and tubes may respond in less than a millisecond, but many take longer.

At 100 mm/second, 254 dots per inch means your dots are 0.1mm in length, fitting 1000 of them in 100mm. At 100 mm/sec, if your power supply and tube take 1 millisecond to fire, your engraving will be offset by a full dot width.

At 500 mm/second, that 1 millisecond delay means you'll be off by 5 dots, or 1/2 a millimeter. Still not very much, but visible. Many power supplies and tubes will take even longer to fire.

The result often looks like ghosted edges.  The image below is a 20mm square at 1000 mm/sec, with a 1ms delay, resulting in a full mm of skew between scans:

![](./img/ScanningOffset-Illustration.png)



LightBurn has a setting to counter this, called Scanning Offset Adjustment, in the [Device Settings](DeviceSettings.md):

This file has been truncated. show original

With raster engraving it is necessary to find these values as it need to compensate for the laser turning on. we found the value for 500mm/s and tend to just engrave at this speed.
The values should be pretty much identical for every machine so @cmachtel values should work with yours.

We are looking into an issue of what would appear to be lost steps. somthing which is not possible with closed looped motors found in the FabKit/CorePlus.
We opted for these motor on recommendation from a friend who uses them in a metal 3D printer. They discovered a slippage on the pulley could occur when accelerating/changing direction quickly. You cannot feel it when testing for blacklash as you need the sudden torque of the motor to create the slip.
We currently dont have a FabKit/CorePlus, more to assemble next week. Once i confirm thats whats happening it should be an easy fix with a bit of loctite but it does explain why it only seems to lose precision when doing lots of small repetitive vector engraving
An option for the moment can be to reduce the acceleration.

The PWM value should be correctly set as its not a skew you are seeing but a ghosted image which is to do with either backlash problems or the delay DC CO2 lasers have.