EEVblog 1573 - TEARDOWN: How a Rotary (Angle) Pulse Encoder Works

How a Rotary (Angle) Pulse Encoder works. This is entirely different to the normal quadrature phase encoder you have used before.
A teardown from the Haefely Trench PESD 1600 ESD Tester.

00:00 - This is NOT a regular quadrature rotary encoder!
02:08 - Tool Tip, Pin Vice
05:41 - EDC tool shootout, Victorinox MiniChamp vs Gerber Dime vs Olight Otacle
06:34 - We're in like Flynn!
07:13 - New Lab Jack for the Tagarno microscope
08:09 - Of course that's how it works
09:59 - Why is this not used more often?
12:29 - How it works
14:49 - Is there another example of this type of encoder? if not, why not?

Or with crypto:
BTC: 33BsprBQNBtHuVzVwDmqWkpDjYnCouwASM
ETH: 0x68114e40ff4dcdd384750500501e20acf3875f8c
BCH: 35n9KBPw9T7M3NGzpS3t4nUYEf9HbRmkm4
USDC: 0x68114e40ff4dcdd384750500501e20acf3875f8c
LTC: MJfK57ujxy55su4XicVGQc9wcEJf6mAoXF

Other channels:

#ElectronicsCreators #Teardown

[SeanBZA]

Long spring goes opposite to the indented one, you can see the place for it. The reason it stopped working after contact cleaner is you removed the lubricant that provides drag between the shaft and the bump mechanism, so it would not move the switch. Probably would have worked with a drop of silicone oil down the shaft, to make a thin viscous film. The reason it failed was probably the low viscosity oil evaporating out of the switch.

[coriscotupi]

I have been using this kind of encoder (manufactured by Alps) in flight simulator hardware. It is used for selecting heading, speed, altitude, radio frequencies, etc. Extremely reliable, not a hiccup in 20+ years of everyday use.

[elitezararus]

im so glad you actually made a video for the teardown. very educational. thanks a lot

[tlhIngan]

The quadrature encoders are trivially easy to decode in hardware - you only need a D flip-flop and you will get out a clock and direction. One of the encoder outputs is connected to the clock of the D FF, the other output is connected to the D input on the flip-flop. Now if you turn the rotary encoder, the Q output of the D FF will give you the direction, and the clock used to clock the flip-flop is a clock for pulse counting. It's not an efficient way as you cut your encoder pulse count down, but it's a cheap and easy way to get direction and clock decoding.

[geoffmorrison]

At 12:29 rotate by about 75 degrees anticlockwise the part you are holding so that the remains of the plastic rivets line up in top and bottom halves. It is then obvious that you have the the spring is in the wrong location. Possibly the other long piece of metal then goes between that spring and the shell wall.

[Citizen]

Long sprint pushes the white part against the cogwheel, the other shaped spring pushes against the cogwheel from the other side so the clickety ratchet action is balanced and smooth and probably to ensure a full step is made and the whilte part stops centered.

[danpatterson]

If your reading glasses are marked 1.25, that's the diopter value, which is the reciprocal of the focal length in meters.

[victoriamarotosilva]

11:39 94€ per unit 🤨 It must be shipped in some really beautiful handmade wooden box decorated by japanese masters with a golden painture mixed with the tears of a unicorn.

[quadmods]

Many years ago, I wanted to convert an up-and-down buttons for a synthesizer to a shaft encoder. I wasn’t smart enough to design it so I put it in front of two other engineers. One used a PIC micro controller, and the other use some 74 LS something logic and both gentlemen able to do it. The same thing as your pulse encoder does in one little package. Very advantageous. I went with the logic and not the PIC…

[gs]

The spring was FLAPPING IN THE BREEZE dave

[sdp]

I was trying to figure out how it would work and was expecting some kind of directional clutch but the actual mechanism is elegantly simple. I too wonder why it is not more popular.

[rickyrico]

Nowadays obviously this would be done in software but the mechanical solutions always make me smile.

[BRUXXUS]

What an outside the box way to think about how rotary encoders work! Performs the same task in a really cool way.

[ahhuhtal]

I have encoders from ALPS that operate in a similar fashion, but which only have three terminals. The way they operate is when you turn the shaft clockwise, the common pin will first connect to the CW pin, and the CCW pin will be left open. Then for each step of the shaft rotation, the connection between common and CW will be broken briefly. Same goes to the counter-clockwise direction. The common will first connect to the CCW pin (and CW will be left floating), and for each step, the CCW pin will open for a brief moment. I bought these long ago from an electronics surplus store thinking they were quadrature encoders. Came with no model info, so can't point to a datasheet. Over the years, I've used them in a handful of projects, but they are a bit of a pain to decode reliably. They tend to occasionally give either 0 or 2 pulses for a single step of the shaft.

[Kirillissimus]

The problem with the design is that it is way too german to survive average users and our price competitive manufacturing. It just does not forgive any mistakes! It must be manufactured precisely to very tight specifications and it must be used in near perfect conditions or otherwise it just will not work. The approach is just not fit for the modern era of disposable everything.

[bobweiram]

YouTube serendipity! Earlier today, I was trying to locate a replacement Tektronix encoder made by Bourns. Its quadrature sectors are etched directly on the PCB board and mounted using plastic pins.

[brainndamage]

I saw your previous video, I think they only used the NO contacts, so it's like if you had two buttons, left and right. It would be very easy to hack in two regular tactile switches. This may be actually why they designed it this way, so that it could be retrofitted into existing designs that used up/down buttons without changing the code.

[IanScottJohnston]

I guess it's not as popular especially in industrial use as quadrature encoding as there is less ability to detect a fault. With quadrature if one of the 2 outputs dies then you know the encoder is knackered.

[PaulHawke]

I saw some like that in the USSR stuff before. Every one I saw was broken for what it's worth.

[mikefochtman]

Around 12:36, looking at the holes you had to 'drill out', you have the cover upside down. That detent spring would be on the side opposite the white piece, so it probably just presses against the black wheel's 'teeth', not the white 'pendulum' piece. That long thin piece of metal, I wonder if it fit into a spot in case to 'push' the white piece against the wheel?