New taz 6 need help please with dimensional accuracy

[shawnkoh]

Hi guys,

Just got my Taz 6 (first printer ever) so forgive me if this is a stupid question.

I printed a few hollow 15mm³ calibration cubes that are 1.5mm thick
I used a vernier callipers to measure the cubes and found that they are generally between about 0.2mm thicker than modelled (i.e. 15.2mm and 1.7mm thick instead)

I also tried printing several circular holes, and found that their diameters were off as well

Diameters in mm
Model Actual

10mm 9.72mm
8mm 7.8mm
6mm 5.74mm
4mm 3.8mm
2mm 1.76mm
1.8mm 1.4mm
1.6mm 1.22mm
1.4mm 1mm

May I know how do I rectify this?
I have already checked that my taz 6 is square as per lulzbot’s service bulletin.

Thanks!

[kmanley57]

I tend to use a 20-40 MM cube to check the X/Y/Z step settings, but it is thus. Bigger calibration cube = better results in calculation.

/* =

Example:

40 MM cube model measures 40.2MM on X axis but 40MM on Y. Assumed default firmware settings for steps on axis and flow rate.

40/40.2*100.5 = 100 <- New X axis steps value

So you would update the X axis setting in the configuration menu with the new steps you have just calculated. Y would not need changed as it has no error in size. But if Y did have a size difference you would calculate and update its steps also.

The X/Y steps are located in the same advanced settings as the extruder steps. See step 6 here
https://ohai.lulzbot.com/project/extruder_calibration/calibration/

This could be off since I am pulling this off the top of my head right now, but is the general process.

[shawnkoh]

@kmanley57

I tend to use a 20-40 MM cube to check the X/Y/Z step settings, but it is thus. Bigger calibration cube = better results in calculation.

/* =

Example:

40 MM cube model measures 40.2MM on X axis but 40MM on Y. Assumed default firmware settings for steps on axis and flow rate.

40/40.2*100.5 = 100 ← New X axis steps value

So you would update the X axis setting in the configuration menu with the new steps you have just calculated. Y would not need changed as it has no error in size. But if Y did have a size difference you would calculate and update its steps also.

The X/Y steps are located in the same advanced settings as the extruder steps. See step 6 here
https://ohai.lulzbot.com/project/extruder_calibration/calibration/

This could be off since I am pulling this off the top of my head right now, but is the general process.

Cheers for the explanation

Hope you don’t mind if I clarify further.

1. When exporting the 40mm cube to stl (I’m using rhino3d). There are some settings for tolerances as shown here.
   

   
   
   Will this impact the accuracy of the print?

2. Do different filaments affect this calibration setting on the number of steps?

3. Are there any other settings I should calibrate? Like eg flow rate

[kmanley57]

@shawnkoh

Cheers for the explanation

Hope you don’t mind if I clarify further.

1. You do, but not really enough that you will notice that small of a tolerance.

2. It is a edge to edge you are adjusting, so expansion of material does affect this in some types of plastics. Calibrate for the filament type you plan to print most and adjust other settings to compensate for the differences.

3. Your settings like the width settings you use in the slicing adjust these factors.

I would recommend you do the calibration using only default settings as that will give a fixed starting point time after time if you have to do this again for any reason. Flow rate would tend to adjust across the board for material property changes.

[partsoven]

I was having some problems with dimensional accuracy (very bad) on my Taz 6 and finally figured it out. On the y-axis motor the timing pulley gear had slid down a bit and was slightly rubbing on the y-axis motor, causing the y-axis moves to “bind” just a bit. This was causing my prints to look terrible, and overall poor machine performance. I loosened the timing gear on the motor and spaced it according to the OHAI docs, retightened, and sure enough I’m back to getting perfect prints. Hope this helps.