Precision DIY Calipers? That’s A Moiré!

Moiré patterns are a thing of art, physics, and now tool design! [Julldozer] from Mojoptix creatively uses a moiré pattern to achieve a 0.05 mm precision goal for his custom designed 3D printed calipers. His calipers are designed to validate a 3D print against the original 3D model. When choosing which calipers are best for a job, he points out two critical features to measure them up against, accuracy and precision which he explains the definition of in his informative video. The accuracy and precision values he sets as constraints for his own design are 0.5 mm and 0.05 mm respectively.

By experimenting with different parameters of a moiré pattern: the scale of one pattern in relation to the other, the distance of the black lines on both images, and the thickness of black and white lines. [Julldozer] discovers that the latter is the best way to amplify and translate a small linear movement to a standout visual for measurement. Using a Python script which he makes available, he generates images for the moiré pattern by increasing line thickness ratios 50:50 to 95:5, black to white creating triangular moiré fringes that point to 1/100th of a millimeter. The centimeter and millimeter measurements are indicated by a traditional ruler layout.

Looking for more tool hacks and builds? Check out how to prolong the battery life of a pair of digital calipers and how to build a tiny hot wire foam cutter.



23 thoughts on “Precision DIY Calipers? That’s A Moiré!

  1. The measurement method is really clever, but there is no way in hell he can actually get that accuracy from the mechanics of the measurement device. A caliper is a somewhat unsuitable measurement device for truly accurate measurements as the offset between the measurement point and the readout causes inaccuracy on the order of hundredths of millimeters in itself, making them very susceptible to variations in temperature and measurement force. There is a reason micrometers with sleepwheels where invented. Making calipers out of plastic makes them even less accurate.

    1. He does discuss this in the video though, aims for 0.5 mm accuracy noting the problem with using plastics, the precision on direct remeasures is about 0.05mm.

    2. He will get 0.5mm accuracy without much trouble. If his target was more demanding than that, no real chance. Precision of 0.05mm is achievable. His explanation of accuracy, resolution, and precision is quite good. I think he is over-optimistic about the accuracy and precision (0.1mm and 0.05mm) of many cheap calipers, though.

      He leveraged the best features of the appropriate tools, and compensated for the weaknesses. If the laser printer is a good unit and care is taken, dimensionally good results can be had. The biggest issue is that many of the transparency films, and the 3-d printed material it is is attached to, are not temperature or humidity stable.

      Moire pattern indication is not new, but is not common. A variety of older instruments used the concept. I don’t know if I have seen the arrow concept before in the wild, but it has been a number of year since I regularly used any gear with this type of optical scale.

    3. Okay, what’s a sleepwheel? A quick search for that term turns up nothing. I assume it’s the torque-limited thumbwheel that’s used to achieve a consistent closing force?

      1. Not all calipers with that shape are called “verniers”.

        Vernier calipers are different than dial and digital calipers.

        Julldozer’s scale is a lower precision, lower accuracy version of a cheap set of verniers. It’s neither easier or faster to read for a given precision.

        It looks beautiful though.

  2. The powder scale in my reloading kit works kind of like that. Pretty neat and you can go down to 1/10 of a grain very easily and can probably get pretty close to 1/20 if you have a steady hand and are careful.

    1. Yes. Last year I made a servo from a brushed DC motor and a moiré incremental encoder. I used a laser printer for the pattern. Didn’t work at high speed, because I was lazy to find a good phototransistor. But otherwise 1000 PPR (or even higher) is achievable, I believe.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.