Loading

MakerBot Print is our newest print-prepration software, which supports native CAD files and STL assemblies,
allows you to interact with all your printers via the Cloud, and many other exciting new features.

Download Now

LuckyPants

Dimensional Calibration Tool v9

by LuckyPants Dec 20, 2016
Download All Files

Thing Apps Enabled

Please Login to Comment

One of the best general calibration tools around. Proves that you don't need to have something fancy and new every year.
Great job!

Thanks for sharing this. I had printed several parts that should have fit together and I had been relying on sanding to get them to mate.
Until seeing your project, I wasn't aware of the importance of the horizontal compensation.
I did have a problem measuring the 2 mm web using your design. My calipers were too fat to fit into the slots, so I had to modify your design and widen the slot.
I also found that I achieved better results when I calculated separate X and Y scaling factors. Mine differed by 0.72%. I assume that differing belt tensions caused the discrepancy.
Additionally, my printer rounded the small flat section of the diamond where the calipers are to measure. As a result, these values seemed wrong and I discarded them. Perhaps if I widened these areas, I could have included them but since I was already happy with the result, I didn't bother.
It might be obvious to experienced Simplify3D users, but I didn't know where to enter the scaling factors so that it would automatically be added. It turns out that it is hidden in Tools>Options>Models>Import Actions.

No problem...

  1. Good point on the 2mm web - I didn't anticipate thicker calipers...I should open that up a couple mm.
  2. I'll rework the spreadsheet to output the scale factors individually for X and Y. Good idea.
  3. I'll open up the flat spots on the diamond - I didn't experience rounding, but its a valid problem and I should adjust the model.

Thanks much for the feedback - all good points.
Ian.

Hmmmm... this calibration tool inverted my problem. Before the calibration the outer dimensions were very accurate (150.04mm, 35.05mm, 79.94mm..) but the inner dimension where slightly too small (diamond: 24.70mm, 2.20mm web). Now, after the calibration I am getting these values with my PLA filament:

Outer values: 151.15mm / 35.20mm / 80.82mm
Inner values: 25.10mm (diamonds) / 2.12mm (web)

What am I doing wrong?

Hmmm that is weird...

If the 150mm dimension is accurate, that's a pretty good sign you've got the part scaled correctly to account for the material shrinkage. What is your scale factor for your PLA on that test? Just curious.

Assuming the part is properly scaled, then the small 2mm web error should be due mostly to the extrusion multiplier. Easiest way I think to check this print a standard 40mm calibration cube with 0% infill and no top and bottom layers, 1 shell. Prints fast and very easy to get direct feedback for layer width. It could be possible you've compensated for extrusion width via part scaling.

Could be some kind of compounding problem....I'd start from square one. Pull out all horizontal compensation and scaling from your slicer. Print the 40mm cal cube as 1 shell and tweak extrusion multiplier alone until width is dialed in. Try and run this again, with no scaling for the first part.

Are you measuring with a .001"/.020mm dial caliper?

Okay... my shell width is between 0.50 and 0.54mm :(

What is your extrusion width in your slicer? .4mm nozzle with .48mm extrusion width?

0.4mm nozzle and 0.4mm extrusion width. O.o

Lol well thats not right then!
Whats your slicer and what kind of printer do you have?

I am using an Anet AM8 (Anet A8 with 2040 aluminium profile frame). My Slicer is Simplify3D.

I'm not familiar with that printer..looks like a Prusa clone, but it sounds like overall your getting too much extrusion...I'd have a look at:

Verify extruder motor steps are correct...disconnect the bowden tube from the hotend and manually jog filament until its flush with the end of the tube. Then command something like E150. Measure the actual amount the extruder pushed the filament and verify it's 150mm. If not adjust the extruder motor steps in the firmware.

Verify the extruder nozzle size...extrude some plastic through the hotend in air with the cooling fan on. Measure the width of the extruded plastic. This isn't perfect but it can give you an idea of what diameter is coming out of the nozzle.

Verify the axes are moving the amount you are asking...Fix the caliper to the bed or frame and bring the head all the way over, then zero set the caliper to the head (make sure it's square to the machine so your not measuring on the diagonal). Command the axes say 150mm, and then extend the caliper and see what you've got....should be within about .1mm (depending on the motors you've got). Repeat a few times to see if the reading repeats.

If all that's good - extruded distance is perfect, extruded diameter is good, and axes are scaled right, then all that's left is the S3D extrusion multiplier (for PLA I'm around .92).

Hi!

I calibrated my extruder setting a few times. But the resulting E steps where very varied and depending on the extrusion speed. With 100mm/min I am getting values around 98 and with higher speeds I am getting E steps up to 115. O.o So I lowered the E steps down to 89. just to test the imfluence on the width. This changed the width down to 0.46mm. But I can't get below that value. And with 89 as E step value I am extruding far below the desired 100mm. :(

By the way. The diameter of the extruded PLA into the air depends on the set mm/minute value. With my optimized E steps (99mm ... 101mm) it is between 0.44mm and 0.48mm.

Great, now my comment is flagged for moderation......?

Just to clarify:

You command a G1 E100. F100. and the filament moves out of the bowden tube 98mm.
You command something faster (lets say G1 E100. F200.) and the filament moves out of the bowden tube, say 115mm?

It sounds like you maybe have an issue with the extruder...possibly one or multiples of the following:

Extruder is skipping steps...as a temporary check, half the acc/dec/jerk values in your firmware for the extruder...see if it has any effect. If not put it back.

Extruder feed mechanism could have an issue, such as:

  1. Hobbed bolt/gear/pinch point could be slipping from not enough pressure or may need to be cleaned.
  2. Hobbed bolt/gear/pinch point could be digging in too deeply into the filament. Have a look at the filament that leaved the extruder and
    see if the teeth pattern is pressing into the filament too much....this can mess with the reading you take for extruding 100mm of filament.
  3. Bowden tube is bad....some that say they are PTFE are not. Slide the filament in and out of the tube by had while flexing the tube....is it
    smooth with little resistance? I've also seen where the ID of the tube was slightly smaller than advertised, causing to much drag.

If those things are ok, and you still can't get a consistent extrusion (ie: G1 E100. F100. and G1 E100. F200. both move the filament out 100mm) maybe consider replacing the extruder motor....those motors are not much more than $15-$20 and it would be a good sanity check.

As for the width, I'd consider changing the nozzle size in S3D from Ø.40mm to Ø.44mm. Your nozzle could be oversized or worn and actually too large for what is in S3D. Try a vase mode cal cube with the nozzle size set to Ø.44mm, extrusion width set to Ø.44mm, and 0.92 extrusion multiplier.

My other reply is "flagged for moderation"! O.O

Yes, it's very weird. :(

My scale factors are:

Simplify3D scale factor: 100.79%
Horizontal compensation: -0.10

I will print this 40mm calibration cube in vase mode and give you a feedback. But that's one of my oldest problem. Never managed it to get an extrusion width lower than 0.44mm with my 0.4mm nozzle. During my test I lowered the E-steps and the multiplier to values I am getting massive under extrusion. But the width still remains at >0.42mm :(

What do you mean with ".020mm dial caliper?" I am using a digital caliper with an accuracy of 0,01 mm.

im getting some measurements that are more than the value they are supposed to be. is that normal?

The 7 mm thickness are without the chamfers I guess?

Correct - chamfers are only to minimize any measuring issues.

This is awesome. Do you have to increase the scale of the part in addition to changing the horizontal offset correction in the slicer? If the part needs to be scaled, should it be scaled uniformly or just on X and Y?

Thanks - I do both.

I'd measure the largest dimension I can and chock it all up to shrinkage, then scale the part uniformly (XYZ) to correct.
After that I'd measure a very small dimension (that I know doesn't have any error due to the slicer not generating the feature correctly) and correct for that with the horizontal offset, usually a pretty small value.

It's of course a tricky balancing act between shrinkage, horizontal offset, and extrusion multiplier - but I find this method gets me pretty close.

OK, that's kinda what I did. I scaled my part based on the percentage that your calc spit out and adjusted the horizontal compensation as well and my prints have never been this accurate. Kinda tricky to remember to scale my prints every time but it seems like the results are worth it.