No, it wasn’t covered here, sorry about that Shawn. Since these scales are not produced/sold anymore, I assumed anyone still following from before was familiar with this project. I’ll try to explain a bit
These scales are encoded using a pure binary numbering system referred to in Bits. A two-bit number has 4 possibilities, 00, 01, 10, & 11 (0 to 3, the zero counts). For each added bit, the possible amount of individual numbers doubles, i.e. a three-bit number has eight possibilities, 000, 001, 010, 011, 100, 101, 110, 111.
The scale tape on the Absolute Origin scales is coded using a 3 track pure binary numbering system. What we dubbed the Fine track is a 9 bit number with 512 possible numbers (0-511). What we found was this track repeats itself over the length of the scale approximately once every 5.04 mm. The Mid track is an 8 bit number (0-255). This sequence is repeated along the scale length over the distance of 8 Fine track cycles. Last is the Coarse track, another 8 bit number that can only cycle once over the length of the scale encompassing 256 Fine track cycles. This is what sets the maximum possible length the scale can be, based on the track period, and is what makes it an Absolute position system.
So by combining the numbers of each of the 3 tracks, you get a unique number combination (or address) for each and every position along the length of the scale. A simpler way to envision this is think of a combination of three rulers set side by side. The Coarse track is a single ruler numbered from 0 to 9. Next to that is the ten Mid track rulers, each 1/10th the length of the Coarse track ruler and each one also numbered 0 to 9. Next to those are the Fine track rulers with each single ruler being 1/10th the length of a single Mid track ruler and are also numbered 0 to 9. So the final lay-out would be one Coarse track ruler, ten Mid track rules, and 100 Fine track rulers with the total length of each track being equal.
Now, think of the readhead like a straight edge laid perpendicular to the rulers. When you read along the straight edge you see a combination of 3 numbers, say 4, 8, & 7. If you look, this number combination exists only once along the entire length.
So why aren’t all scales made this way ? Easy, cost and functionality.
Most non-absolute scales use a simple count logic. When you set it to zero and start moving, it just counts up or down dependent on direction of travel. If it loses power or is moved too fast, it has no idea of where it’s at and displays zero again when power is restored or displays a false position because it lost count. On the plus side, these are cheap and easy to manufacture and can be any length you want.
On the absolute scales, they know exactly where they are at since each position has its own unique address, but are more expensive to produce and are limited in length based on the number of tracks used and the resolution of each track. To make an Absolute scale longer you have three choices.
1. Add more tracks
2. Decrease the resolution
3. Combination of 1 & 2 above.
I hope this helps to explain why this system won’t work with non-absolute scales. It even won’t work with other absolute scales unless they used the same algorithm to encode the scales since part of the algorithm is each track is used to correct the others for noise. I won’t go into details, but you’ve probably seen scales where the position isn’t very stable, it bounces around. These use the different tracks to correct the noise in the others making them very stable. You should be able to see that in the video where he position indications are rock solid, with the possible exception of the least significant digit if it’s sitting on the edge between one position address and the next like you will be able to see in the beginning of video (once Google finishes processing it so it can be viewed) where the Z-axis indication is bouncing between 123.22 and 123.23.
