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

Подписаться 941 тыс.

Просмотров 38 тыс.

50% 1

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.
Unboxing & Testing video of 16kV ESD tester: www.eevblog.com/2023/10/13/ex...
Pointless video about toggle switches: • A Totally Pointless Vi...
Forum: www.eevblog.com/forum/blog/ee...
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?
If you find my videos useful you may consider supporting the EEVblog on Patreon: / eevblog
Or with crypto:
BTC: 33BsprBQNBtHuVzVwDmqWkpDjYnCouwASM
ETH: 0x68114e40ff4dcdd384750500501e20acf3875f8c
BCH: 35n9KBPw9T7M3NGzpS3t4nUYEf9HbRmkm4
USDC: 0x68114e40ff4dcdd384750500501e20acf3875f8c
LTC: MJfK57ujxy55su4XicVGQc9wcEJf6mAoXF
Web Site: www.eevblog.com
Other channels:
EEVblog2: / eevblog2
EEVdiscover: / eevdiscover
T-Shirts: teespring.com/stores/eevblog
#ElectronicsCreators #Teardown

Наука

Опубликовано:

12 июл 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 193

@SeanBZA 8 месяцев назад

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.

@EEVblog 8 месяцев назад

I think the additional problem was due to me heavy handidly putting my knife down each side. The case bulged out. Most likely that's not good for the tight tolerances required. But desoldering/soldering could also be an issue.

@SeanBZA 8 месяцев назад

@@EEVblog Glass filled nylon there, so likely robust enough to survive soldering. Think it just lost lube there, plus the contacts wearing the gold finish off on the spring brass wiper, so it stopped making good contact. Also probably not helped by too low a contact current, they have a minimum contact current on switches for a reason. This probably has 10k resistors, and thus not enough current flow to keep the contact surfaces clean long term. Probably needs 1k and a 1n ceramic capacitor across the switch contacts, to provide that little bit of spark to clean them.

@theslimeylimey 8 месяцев назад

The white pawl is acetal aka Delrin which is self lubricating and will be happy to do its thing without lubrication in this type of mechanism.

@EEVblog 8 месяцев назад

@@theslimeylimey Yeah, shouldn't need lube I suspect. That's what she said.

@sirtra 8 месяцев назад

My theory is that it failed due to wear on the cog and pin - on the assumption you didn't bend the left side gold plated strip during opening it, it looks visibly bent with a kink compared to the right side. To elaborate, as teeth wear down the white piece will gradually move higher (to the centre/pivot point). Once the wear reaches a point high enough, when the white piece moves sideways to push the gold plated strip to touch the outer pin, it will push it beyond the elasticity point and gets plastically deformed. I'd imagine it might begin to fail intermittently at first, like as it's rotated and clicks over say 24 times, electrically it may only contact the outer pin sufficiently to register 13 times. Also i wonder if one potential undesirable characteristic of this design would be that the strip may connect both pins at the same time (or neither pin) - with such tight tolerances i don't see how it wouldn't. Just a spit-balling theory though, i could be completely off base and wrong.. but super interesting how simplistic something so expensive is, oversimplified it's just a fancy bidirectional switch no? 😂

@victoriamarotosilva7093 8 месяцев назад

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.

@coriscotupi 8 месяцев назад

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.

@EEVblog 8 месяцев назад

Part number?

@coriscotupi 8 месяцев назад

@@EEVblog It's been 20+ years, so I can't say for sure, but I want to think it's the *SRBM1L0800.* I'm a bit thrown off by the "6 positions" title at the top of the page, but from the electrical diagram shown at the bottom of the page (two momentary contacts, each driven by CW or CCW rotation) it could not encode 6 discrete positions. Perhaps the "6 positions" title is carried over from the SRBM switch series, most of which are in fact simple 6-position switches, as seen by their diagrams. It does seem that SRBM1L0800 is the switch that I bought so many years ago.

@geoffmorrison3648 8 месяцев назад

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.

@tlhIngan 8 месяцев назад

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.

@robstamm60 8 месяцев назад

Certainly a cheap way to get speed and direction but you get unexpected results when changing direction. In one direction it will first count in the opposite direction when the clock signal changes before the direction signal - you can even get it to constantly count up by turning it back and forth - acceptable for a cheap input device but useless for accurate position tracking.

@quadmods 8 месяцев назад

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…

@SeanBZA 8 месяцев назад

LS logic would be a 74LS74 to do edge detection on the inputs, after running through 74LS132 with some RC filtering, to get a good sharp edge and debouncing, as slow edges are a bad thing with flip flops. One half provides the up pulses, and the other the down pulses, with only 2 packages and some resistors and capacitors.

@Graham_Langley 8 месяцев назад

@@SeanBZA Classic quadrature decoder is a D-type like the CD4013. Inputs to the Clock and Data, Q or /Q will give the direction. First used it around '72 so it's nothing new.

@Citizen5000 8 месяцев назад

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.

@danpatterson8009 8 месяцев назад

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

@gutschke 8 месяцев назад

For reading glasses, you can actually specify several parameters. They can have diopters, astigmatism correction, and magnification.

@danpatterson8009 8 месяцев назад

@gutschke As far as I know prescriptions are expressed in diopters.

@elitezararus286 8 месяцев назад

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

@sdp8483 8 месяцев назад

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.

@EEVblog 8 месяцев назад

Yeah, I was surprised by the simplicity. But the instant I saw it it was obvious.

@nkronert 8 месяцев назад

If this mechanism turns out to be cheaper to produce but just as reliable as current implementations, someone will likely start producing them after becoming aware of it through this video, as any patents must have run out by now. Then again, looking at how cheap rotary encoders are nowadays, I can also imagine that no one will put in the effort.

@Kirillissimus 8 месяцев назад

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.

@omgsrsly 5 месяцев назад

Well, for clumsy fat-fingered savages this clearly is unknown technology

@gs425 8 месяцев назад

The spring was FLAPPING IN THE BREEZE dave

@flagg85 8 месяцев назад

Just looked up Rotary angle pulse encoder, and the Wikipedia page uses a screenshot from this video 😂

@Joetechlincolns 8 месяцев назад

😂😂

@mikebarushok5361 8 месяцев назад

It's a little bit misleading without the toothed shaft.

@EEVblog 8 месяцев назад

Wow, imagine that!

@I.____.....__...__ 8 месяцев назад

@@Okurka. So it was a choice to not included a reference to the video or channel? 🤔

@kick1ass20 8 месяцев назад

First time I've come across a mechanical encoder. Definitely worth the teardown!

@BHSAHFAD 8 месяцев назад

Awesome video, i love when you explore rare components

@PaulHawke1337 8 месяцев назад

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

@rickyrico80 8 месяцев назад

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

@JKnight89 8 месяцев назад

Very interesting component!

@ChongMcBong 8 месяцев назад

cheers dave, cunning bit of design :)

@BRUXXUS 8 месяцев назад

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

@andrew_koala2974 8 месяцев назад

DAS IST VERRRRY ENTERESING DANKE SCHON

@chrislemus9392 8 месяцев назад

Excellent.

@frankowalker4662 8 месяцев назад

That is so cool and useful. You don't even need software to decode it.

@bobweiram6321 8 месяцев назад

RU-vid 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.

@ahhuhtal 8 месяцев назад

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.

@gutschke 8 месяцев назад

That's exactly why you want both the NO and NC contacts. By having two contacts for each direction, you can reliably debounce the signal. This is easy to do in either software or hardware and should completely eliminate the problem that you encountered.

@ahhuhtal 8 месяцев назад

@@gutschkeYou're probably right. There's a lot more redundancy for the decoding with the extra pins. I've attributed the additional pulses to the fact that even a tiny amount of back-rotation on the shaft will disconnect the direction pin from the common, and which occurs in time scales much longer than the contact bounce time. I experimented with many different lengths of dead time for the debouncing, but the double counts persist until the dead time is so long that it's otherwise unusable. You can get it to a point, where the double counts approximately equal the zero counts, and thus you're at least on correct on average. I've been using them with the common grounded and pull-ups on the direction pins, which to me feels like the right way to do it.

@gblargg 8 месяцев назад

I just noticed the pixellation of the mechanism in the video thumbnail. That's some funny clickbait there. Well done.

@brainndamage 8 месяцев назад

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.

@EEVblog 8 месяцев назад

All 4 lines go to pullups and the micro.

@NiHaoMike64 8 месяцев назад

@@EEVblog Looks like it would be possible to bodge in a pair of SPDT microswitches and a 3D printed lever mechanism to activate one or the other when pressing it from either direction.

@AbhishekKumar-el7vo 8 месяцев назад

@@NiHaoMike64 this guy here did exactly the same. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-I41c4Bo2q7I.htmlsi=IpoM_lLYuPleZbTM

@gutschke 8 месяцев назад

@@EEVblog Reading both the NO and NC contacts is probably done for debouncing. You could do the same, if you replaced the rotary pulse generator with two push buttons.

@olegbashtenko 8 месяцев назад

@@gutschke IMHO - exactly! These SPDT contacts allow you to use simple RS flip-flops for absolutely reliable debouncing (one for each channel). So basically this encoder + 2 flip-flops give you 2 logic output lines equivalent to UP and DOWN buttons, but with "built-in" debouncing. Very clever design, never saw it before! Thank you Dave for sharing this.

@TobyRobb 8 месяцев назад

Pretty interesting mechanism.

@MrTurboturbine 8 месяцев назад

Add velocity to your encoder interface so we don't have to turn the thing 100 times to change a value from 0 to 10...

@GordieGii 8 месяцев назад

That is one of the standard way of sending control pulses to an industrial servo amplifier.

@Damien.D 8 месяцев назад

I'm pretty sure it can be replaced by two push buttons. The micro only look for a pulse. There's nothing to decode here, it's juste two switches and the magic is purely mechanical.

@ELECTROxigeno76. 8 месяцев назад

Interesting Device,

@pa4tim 8 месяцев назад

I use 1.5 for normal reading but bought a 3.0 for use as magnifier glasses. Now teardown a different version, please :-)

@andywander 7 месяцев назад

If you look at the locations of the formerly heat-welded posts that held it together, you'll see that the formed spring does not bear against the white pawl, but fits in the opposite side and will just provide detents by acting directly on the ratchet wheel teeth. I suspect that the flat strip is a spring that presses against the pawl.

@mojoblues66 8 месяцев назад

I think this is quite clever.

@IanScottJohnston 8 месяцев назад

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.

@EEVblog 8 месяцев назад

True.

@SeanBZA 8 месяцев назад

But made inputs easier with a low code space, just debounce in hardware, and 2 inputs direct to a little code to increment or decrement a counter. No need to have fast code looking for edges using an interrupt, and then decide if it is valid, and update the counter. Just poll input, and look for a high on the enable pin, and then look for edges.

@TLang-el6sk 8 месяцев назад

If I remember correctly I had a similar thing in a kit for a model railroad transfer platform controller. I think it was from ALPS. Oh, sorry - that one was a quadrature encoder, and one of the best...

@datawolk 8 месяцев назад

Very interesting. Maybe a interesting topic for another video: electronic torque measurement device, like from Amazon Aliexpress etc. Very curious of how they work and can be so acurate for that price.

@johnclawed 6 месяцев назад

The way it pulses one pin or the other, it seems like it was intended to drive 74192/193 counters directly.

@danman32 8 месяцев назад

I would imagine this SPDT per direction would be easier to handle debounce, since you would only count the NO contact after switch had returned to NC contact, since the DT appears to be break before make.

@donaldhoot7741 5 месяцев назад

Looks like the flipper on a ducks a$$. Great video!

@TomLeg 8 месяцев назад

You say it costs $6, but the Burklin page lists it for 93 Euro

@EEVblog 8 месяцев назад

@@Okurka. 6 Euro each, plus postage.

@chrisridesbicycles 8 месяцев назад

I guess the 93€ are mainly the cost for it sitting in a shelf since 1993 😂

@hexagonrecords6033 7 месяцев назад

You should try the bourns ACE encoders, they are also creative types of encoder

@michael.a.covington 8 месяцев назад

I think you'll find the +1.25 and +1.75 numbers describing your glasses are not magnification. They are the reciprocal of the focal length in metres. So if your eyes can only focus on far distances, the first pair makes them focus to 1/1.25 = 0.8 m, and the second, 1/1.75 = 0.6 m. In practice your eyes can also focus closer than that, but the glasses enable them to focus even closer.

@GordieGii 8 месяцев назад

The bent one goes opposite the white piece to produce the detent. The straight one pushes the white piece against the teeth.

@electronics.unmessed 8 месяцев назад

Nice principle, so simple and clever, thanks for sharing. BTW, I never expected that there is a german expression for an electronics component and no proper english word. Usually, it is vice versa.

@johnwettroth4060 8 месяцев назад

Interesting- one thing that this design tradeoffs is that the backlash is not controlled. It depends on friction and how fast that little shuttle finds itself pushing on the contact. Quadrature encoders have well defined backlash important in positioning apps. This is better for a manual knob since its simple to decode. You can take some resistors and diodes and make a one terminal encoder- in one direction, the signal idles low and pulses high and in the other, idles high and pulses low- pretty simple to decode with just one pin.

@tomvleeuwen 8 месяцев назад

Did you turn it both ways when testing? I think I'll go back to the original video to check that out. I checked out the original video, and yes it was broken both ways. What we now know is the normally closed contact was permanently open.

@KeritechElectronics 8 месяцев назад

Pretty cool. Hmm, coupling it with an up/down counter comes to my mind...

@EEVblog 8 месяцев назад

Debounce!

@snik2pl 8 месяцев назад

that's like ideal substitution for plus minus buttons

@rj7855 8 месяцев назад

I wouldn't say it's unreliabl:, it worked 20 years

@abrudner 8 месяцев назад

I wonder if they used this over the standard due to the application it was in. Like the static from the brushes could affect the instrument qualification with noise or something

@KACPER050599 8 месяцев назад

I've seen n something similar in climate control panel from a car, but based on a switch like ESE24MV1T from PANASONIC. protrusions end detents were built into knob and front panel. Using discrete switch could improve reliability and tolerances issue, but in this space this propably results in small number of impulses per rev

@ludviktesar6551 8 месяцев назад

This rotary pulse encoder can probably be emulated by normal quadrature pulse encoder and one dual JK type flip flop chip (e.g. 74L73): The two outputs A, B of standard encoder would have to be fed into 74LS73 like this: A-->J1+K1+CLK2, B-->CLK1+J2+K2. Then Q1 and /Q1 goes instead of the two pins on left side and Q2 and /Q2 goes instead of the two pins at right side. This is assuming TTL logic is used, otherwise 4027B can be used.

@rileyjones7231 8 месяцев назад

It seems kinda weird that the flat "contact bridge" isnt bound to one of the pins. It seems like the pressure against the "normally open" pin could pull the contact off of the "normally closed" pin

@joegee2815 8 месяцев назад

I think optical detection of rotation is easier/cheaper.

@NINEWALKING 8 месяцев назад

Principle looks awesome to me. Sure I wonder how sensitive to dust it is because of the small force applied to contacts. Looks like little bit od dust would be able to act as an isolator. Anyhow something tells me if that encoder was scaled up just to like 1,5X of it's current size, it would be easy to make it reliable as well. As it is now it is like fine watch mechanic but using plastic parts as well. Sure isolating properties are needed but plastic is not most stable thing to use at that size. Nowadays there are better and reinforced plastics and Teflon's and Delrin's and such and possibly ceramics that would do way better job I think. Anyhow better plastic and bit bigger size, maybe even just 1,3 times bigger and that could be more reliable I think.

@I.____.....__...__ 8 месяцев назад

I recently had to replace the quadrature rotary-encoder of my mouse because the wheel was getting flaky. I put in a cheap Chinesium one from eBay and it's back to being flaky (perhaps more so) after just a couple of months. I don't know if the encoder itself is the problem or the fact that it wasn't the right size like the listing said and maybe it's developed a cold-join from cracked solder from the stresses of middle-clicking. 🤔 I can't be arsed to reflow it right now and swapped mice.

@Fluxkompressor 8 месяцев назад

A long time ago I was working for the automotive industry R&D, blasting climate controls with an ESD gun very similar to yours Some batch of these climate controls had a problem with their encoders. They failed at 10kV I believe. I could not measure anything and wondered how on earth a simple switch contact could be damaged by an ESD pulse. I opened one up and found out they were optical with very complex light pipe action going on inside. Tiny little things with just about 10mm on each side. And sure enough after knowing what to measure I found out that one of the Phototransistors was open They had like a backplane with the active stuff and lightpipes leading to that backplane (probably to avoid ESD problems ... yikes) You could also take one functioning phototransistor out and replace it from another unit. I made a bunch of projects using these "refurbished" encoders

@thomasunsworth425 8 месяцев назад

Very clever. Not sure what pins are supposed to make or break as it turns.Looks a bit like a mini morse code key

@DjResR 8 месяцев назад

That operation looks like fancy way of two buttons._

@gblargg 8 месяцев назад

Two variations of a quadrature decoder: int decode( unsigned gray ) { static unsigned old = 0; unsigned binary = gray ^ (gray >> 1); int delta = (binary - old) & 3; old = binary; if ( delta == 2 ) delta = 0; return delta - ((delta & 2) > 1); int delta = binary - old; old = binary; switch ( delta ) { case -1: case +3: return -1; case +1: case -3: return +1; } return 0; }

@rocketman221projects 8 месяцев назад

I prefer the lookup table method. It does a good job software debouncing mechanical encoders. int8_t decode(uint8_t gray) { const int8_t states [] = {0,-1,1,0,1,0,0,-1,-1,0,0,1,0,1,-1,0}; static uint8_t old = 0; old

@QsTechService1 8 месяцев назад

If I would guess The contact goes to the left 12:14

@mikefochtman7164 8 месяцев назад

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?

@mikefochtman7164 8 месяцев назад

looking inside the 'cover' (not the part with the contacts) I think I see two tiny slits for the straight piece to slip into. That would make it a sort of 'leaf spring' to hold the white piece inward toward the center and allow it to slide side-to-side as you turn the 'gear piece' each direction.

@Janom66 8 месяцев назад

Yes, you have violated the relationship between the parts after disassembly. 6:53 You need to rewatch the video and it turns out that the curved spring is located on the opposite side. It is necessary to restore the sequence and there will be a place for a flat spring.

@hermannschaefer4777 8 месяцев назад

German is not that hard for British: Drehimpulsgeber -> Dreh-impuls-geber Dreh --> drehen verb --> cognate with Old English throwen, meaning to twist, still in throw a pot from clay on a wheel. --> to turn, twist impuls --> impuls from Latin impulsus, the PPP of impello, to push geber --> noun of verb geben --> cognate to Old English giefan --> English to give --> noun: a giver --> turn-impuls-giver So easy.. :D

@gblargg 8 месяцев назад

So the double-pole arrangement is to avoid having to deal with switch bounce.

@ShtopoRrr 8 месяцев назад

no, you won't need a micro to replace the encoder. just two buttons: one to go up, another to go down. It may be even an option to improve usability even now when you have encoders

@ErrorMessageNotFound 8 месяцев назад

I feel like that functionality could be replicated with a miniaturized version of the rotary switch mechanism used on multi-meters.

@drkastenbrot 8 месяцев назад

Dave briefly mentioned expecting wiper contacts in this thing, just like a quadrature encoder would have.

@WizardTim 8 месяцев назад

Very elegant design, was wondering how on earth it worked inside, or at least how on earth they got it to work in such a low profile base. I do however wonder how fragile it is during the soldering process, the fact the pins are also the contacts and that the tolerances appear to be pretty tight makes me wonder if you reworked it a few too many times or dwelled a little too long or with a bit too much force if you'd break it or reduce the life expectancy.

@Damien.D 8 месяцев назад

the datasheet is indeed recommending not to heat it too much.

@SeanBZA 8 месяцев назад

@@Damien.D That is true for almost all solder in switches, where the contacts are going to melt the case a little, so they have a max temperature profile to handle that, so the pins do not move in the housing.

@todayonthebench 8 месяцев назад

At least this is trivial to convert into two push buttons.

@tra757200 8 месяцев назад

I’m kinda partial to optical encoders. That one seems like it would wear out too soon.

@gutschke 8 месяцев назад

The data sheet lists some insanely large number of operations though. For most applications, you shouldn't see the part wear out over the lifetime of the product that it is used in.

@KoRntech 8 месяцев назад

Nice, I wish o could've found a replacement for my Yeasu 1500M encoder, poor design snapped the knob off then a couple years later they no longer had the replacement part.

@EEVblog 8 месяцев назад

Custom jobbie?

@KoRntech 8 месяцев назад

@@EEVblog I got rid of it at a hamfest several years ago. Super glue worked but I wanted it to be original. It was just a poor design, if you saw the front of the radio you'd see a brick with a plastic knob ready to be struck by anything off to the side to snap the stem. Solid receive and my first Yaesu.

@falkmachtsachen 7 месяцев назад

Won't a SLA-Printed one do the job?

@nigeljohnson9820 8 месяцев назад

A simple up /down switch and some software would have provided a more reliable solution. Surely good design practice would require at least two independent sources for a critical component.

@falkmachtsachen 7 месяцев назад

In the 90s you had Alps, EBE, Möller and Marquard.

@nigeljohnson9820 7 месяцев назад

@@falkmachtsachen there is still a less complex, more reliable solution using simple push to make switches.

@pman1902 8 месяцев назад

First thing i liked about this video was the Bitcoin logo on the mic :D

@Fifury161 7 месяцев назад

...but there are no kangaroos in Austria!

@der.Schtefan 8 месяцев назад

It's not even weird that the manufacturer does not market it internationally; they probably have their 20 German customers that you never heard of, that make stuff for another 20 German companies that you never heard of, that then make stuff for another 20 German companies that you might know if you are in a specific field, that then make stuff for companies you know very well. Typically all small/middle-sized, family run since 1887, industrial manufacturing companies, with HQ in a small town in the black forest or "somewhere near Stuttgart" or Cologne. The company would probably not even think of marketing this internationally, since they can't compete with the low cost quadrature encoders.

@falkmachtsachen 7 месяцев назад

Did a project with a customer, where we needed a modification to one of EBEs sensors. The chief dev wanted to do it in software, but we were thin on memory (those were the days) so, we asked them and they delivered five handmade prototypes in two days.

@johnsonlam 8 месяцев назад

I guess it lost the precision over time due to metal fatigue.

@EEVblog 8 месяцев назад

Possible.

@GapRecordingsNamibia 8 месяцев назад

Yes.... I too, suffer the cursed age eye correlation thing..... I have my regular multi focal's..... Then a 1.8 and a 3 for table work..... So, I tote 3 pairs of of eye glasses around at all times, the 3 I only wear when doing PCB work as that is like 30cm max focal distance...... Getting old is not for sissies........😂🤣😂🤣 My daily carry tool... Leatherman, always have. 30 years now.

@EEVblog 8 месяцев назад

The Leatherman can't fit in your fob pocket.

@coriscotupi 8 месяцев назад

Similar situation here. Multifocals for everyday use, 4 diopters for minute work, plus an "arm's length" prescription, specific for the computer screen. This last one is highly recommended, much much better than mutifocals for large PC screens.

@piotr-Garou 8 месяцев назад

Musisz to składać odwrotnie. Zobacz na kołki, które rozwierciłeś.

@firstnamegklsodascb4277 8 месяцев назад

"Everyday carry shootout" Lol

@user-tn9pq7bv5h 8 месяцев назад

doesn't a small spring with a tooth create a click on the gear?

@granthoughton769 8 месяцев назад

So just the 94 € then! I wonder why no one uses them...

@Graham_Langley 8 месяцев назад

The question is why do it like this? Only advantage I can see is there's no need for logic to decode the direction as with a quadrature encoder, but that only needs a D-type and a use can usually be found for the other half of say a CD4013 if only as an inverter or buffer - how to do this is left as an exercise for the reader.

@ronniepirtlejr2606 8 месяцев назад

I was going to recommend some small buttons, if you couldn't find the replacement part. Not sure if they would have worked? It would have been slow dialing it in but, anything's better than bricking it!👍

@stefanhauser2804 8 месяцев назад

Drehimpulsgeber: er gibt drehimpulse

@denniskubik7360 8 месяцев назад

Those are just 2 microswitches with 2 outputs each in one housing? I don't know why it shouldn't work reliably as you get older. Ultimately, the well-known micro switches also work very reliably.

@XtianApi 8 месяцев назад

Once he fixes it he's like, "how do I set the clock, ah whatever, I don't want it anymore, it's not broken." Well done sir!

@tgfcujhb7583 8 месяцев назад

TYRE TEARDOWN....🤣🤣 CAPS👍👍

@kurtu5 8 месяцев назад

Seems like a perfect encoder for a MEMS. But as we can see, at larger scales, its "dicky."

@AdityaMehendale 8 месяцев назад

"turny impulse-giver"

@Damicske 8 месяцев назад

Pfff mine glasses are 7.25, but with all the new technologies they aren't that thick.

@EEVblog 8 месяцев назад

WTF!

@Damicske 8 месяцев назад

@@EEVblog without them it's all blurry after 10 cm ;)

@Ma_X64 8 месяцев назад

Hi Dave! Could you explain all the weird little phrases you use? Apparently these are cultural references, but people from other countries might be interested to know where they come from and what they mean.

@EEVblog 8 месяцев назад

And I'd have the eep explaining them over and over and over again, year in year out...

@Ma_X64 8 месяцев назад

@@EEVblog So it's a kind of tradition) Would you give me a link?

@TheHuesSciTech 8 месяцев назад

Very cool video, I was curious how it worked! But, this is insane to me -- it feels like the market segment that this device fills must be created by incompetent and/or impossibly lazy firmware engineers. Given a choice between this clever-but-fragile mechanism, and a normal encoder plus a few lines of firmware, why on earth would you pick the former. Don't get me wrong, the device is a beautiful and elegant piece of engineering to solve the given problem. But the problem doesn't exist, and the solution introduces entirely unnecessary fragility. Or am I missing how crap microcontrollers were in the 90's?

@erikdenhouter 8 месяцев назад

Even today, many designs with microcontrollers are going to the edge of them, using all kind of tricks to get more functions out of the already available micro's. E.g. modern washing machines, or the cheap weather stations, show a fixed "amount" of functions, still different functions per model to choose from, but often the same amount. And I think that wasn't any different then. If you as a manufacturer could simply mechanize a part, it would be beneficial to a simple controller, and sell.

@falkmachtsachen 7 месяцев назад

Normal encoders in a ESD gun...

@kristyskirt9015 8 месяцев назад

A dit dit dit dit encoder no DA

@johnclawed 6 месяцев назад

Maybe she dropped your gerber tool into the paint can.

@Roobotics 8 месяцев назад

Whelp, totally having to eat my prior comment on this one, granted it's not a quadrature encoder, but an encoder is an encoder all the same, so you very much could have just replaced this with singular push-buttons for the up-down adjustments as you'd said. TIL!

@pr191148 8 месяцев назад

Und die Kandidaten raten

@borstenpinsel 8 месяцев назад

The thumbnail looks like you started playing Minecraft😅

@joelmurphy9369 8 месяцев назад

It seems there is too much chance for failure with this design. Does the datasheet say how many operations this thing can handle? Asking a part to function by slamming its parts together over and over is super old school industrial revolution type stuff.

@jaro6985 8 месяцев назад

Rating is 187k revolutions or 6 million clicks. A normal PEC11 rotary encoder is only rated 30k revolutions. So likely either someone screwed up the design/implementation, or the spec is a bit of wishful thinking.

@falkmachtsachen 7 месяцев назад

@@jaro6985 As my Mercedes has a bunch of EBEs encoder switches: These are reliable but don't like being not used.