|
QUESTION: Cheap Accurate Shaft Position Encoder
Does anyone know of a cheap and accurate shaft encoder assembly,
suitable for that "professional rig feel" for spindle position to frequency encoding. I'm thinking about some sort of integrated unit that contains the encoder, code wheel, assembly, bearings etc. Can anyone help.... Tim |
In article .com,
wrote: Does anyone know of a cheap and accurate shaft encoder assembly, suitable for that "professional rig feel" for spindle position to frequency encoding. I'm thinking about some sort of integrated unit that contains the encoder, code wheel, assembly, bearings etc. HP used to make very nice optical rotary encoders - e.g. the HEDS-7500 "digital potentiometer" (which has its own shaft and mounting), and the HEDS-5000 series optical encoders (which mount on an existing shaft). These are from a 1989 catalog, and the current versions or equivalents are no doubt different. www.newark.com lists a bunch of them. I would not call them "cheap". According to Digikey's catalog, Grayhill and Clarostat and Iwatsu and Bourns all make similar rotary optical encoders with reasonably high pulse-per-revolution counts (128 or 256). It looks to me as if you're probably facing a price of $40 - $60 per piece, in single quantities, for any commercially-made rotary encoder of this caliber. For anything very much less expensive, you'll probably have to homebrew something (e.g. a couple of simple interruption-type photosensors, and a code wheel laserprinted on a piece of plastic) and mount it on an existing spindle/shaft. -- Dave Platt AE6EO Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior I do _not_ wish to receive unsolicited commercial email, and I will boycott any company which has the gall to send me such ads! |
Dave Platt wrote:
In article .com, wrote: Does anyone know of a cheap and accurate shaft encoder assembly, suitable for that "professional rig feel" for spindle position to frequency encoding. I'm thinking about some sort of integrated unit that contains the encoder, code wheel, assembly, bearings etc. HP used to make very nice optical rotary encoders - e.g. the HEDS-7500 "digital potentiometer" (which has its own shaft and mounting), and the HEDS-5000 series optical encoders (which mount on an existing shaft). These are from a 1989 catalog, and the current versions or equivalents are no doubt different. www.newark.com lists a bunch of them. I would not call them "cheap". According to Digikey's catalog, Grayhill and Clarostat and Iwatsu and Bourns all make similar rotary optical encoders with reasonably high pulse-per-revolution counts (128 or 256). It looks to me as if you're probably facing a price of $40 - $60 per piece, in single quantities, for any commercially-made rotary encoder of this caliber. For anything very much less expensive, you'll probably have to homebrew something (e.g. a couple of simple interruption-type photosensors, and a code wheel laserprinted on a piece of plastic) and mount it on an existing spindle/shaft. Or take apart a computer mouse as they each have two. Bill K7NOM |
But nothing like 256 pulses/rev.
Cheers, Tom |
Tim:
There used to be a website with info on using surplus stepper motors as precision shaft encoders. I have been doing Google searches without success. Perhaps someone else will be able to provide a URL. It was Australian, I believe. Good luck. Roger K6XQ |
"Roger Leone" ) writes: Tim: There used to be a website with info on using surplus stepper motors as precision shaft encoders. I have been doing Google searches without success. Perhaps someone else will be able to provide a URL. It was Australian, I believe. Good luck. Roger K6XQ There was an article in "Radio Electronics" (or maybe it had morphed into "Electronics Now" by that point), I'd say around 1994 or so. The concept isn't much more than taking the outputs of the stepper and putting them through comparators to get a binary waveform. But of course, picking the right stepper is important since you need the fine steps, and I know when I brought this up before, someone had something to say against the concept, though I sure can't remember what they felt was wrong with the concept. Michael VE2BVW |
Steppers have no output unless the shaft is moving. Steppers used this
way are really a velocity encoder and not a position encoder. Regards, Glenn AC7ZN |
Steppers have no output unless the shaft is moving. Steppers used this
way are really a velocity encoder and not a position encoder. Regards, Glenn AC7ZN Glenn: I'm sure that limits stepper motors to certain uses, but I think this thread/post may be incorrectly titled. Unless I misunderstand the intended application, I don't think a "position encoder" is what he is looking for. Rather he wants something that, when rotated, feeds pulses to an up/down counter for frequency synthesis. The position of the shaft is not important as long as its rotation can be used to generate pulses for the counter. Roger |
Roger Leone wrote: Steppers have no output unless the shaft is moving. Steppers used this way are really a velocity encoder and not a position encoder. Regards, Glenn AC7ZN Glenn: I'm sure that limits stepper motors to certain uses, but I think this thread/post may be incorrectly titled. Unless I misunderstand the intended application, I don't think a "position encoder" is what he is looking for. Rather he wants something that, when rotated, feeds pulses to an up/down counter for frequency synthesis. The position of the shaft is not important as long as its rotation can be used to generate pulses for the counter. Roger Roger: Yes you're right, that's exactly what I'm looking for... any ideas as to where I might get such a beast? Tim |
In article om,
I'm sure that limits stepper motors to certain uses, but I think this thread/post may be incorrectly titled. Unless I misunderstand the intended application, I don't think a "position encoder" is what he is looking for. Rather he wants something that, when rotated, feeds pulses to an up/down counter for frequency synthesis. The position of the shaft is not important as long as its rotation can be used to generate pulses for the counter. Roger Roger: Yes you're right, that's exactly what I'm looking for... any ideas as to where I might get such a beast? Usually known as a "rotary encoder". They normally have two outputs, in a phase-quadrature arrangment. These can be decoded to create up/down/clock pulses using dedicated ICs (HP makes 'em), or via a small collection of discrete TTL logic chips, or via a simple software routine in a PIC micro or similar (which is how I'd probably do it these days... I wrote a simple state-table routine for an 8051 some years back which worked out quite well). You'll probably want at least 64 counts per revolution, and probably 256, to get a nice smooth "feel" to the synthesizer tuning. Most such use an optical code wheel and a pair of optosensors. Digi-Key catalog lists quite a few such (all with their own shafts, ready for panel mounting), but they aren't cheap. Mechanical rotary encoders are less expensive, but less precise (fewer counts per revolution) and possibly not as reliable or long-lived since they use mechanical contacts. -- Dave Platt AE6EO Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior I do _not_ wish to receive unsolicited commercial email, and I will boycott any company which has the gall to send me such ads! |
|
On Thu, 14 Jul 2005 01:00:45 +0000, Roger Leone wrote:
Tim: There used to be a website with info on using surplus stepper motors as precision shaft encoders. I have been doing Google searches without success. Perhaps someone else will be able to provide a URL. It was Australian, I believe. Good luck. Roger K6XQ That was probably on Richard Hosking's (VK6BRO) web-page. The page seems to have disappeared from the web. Hopefully it has moved to a new URL. Are you lurking here Richard? This page has some info: http://www.webx.dk/oz2cpu/20m/encoder.htm 73, Ed. EI9GQ. -- Linux 2.6.12.1 Remove 'X' to reply by e-mail. |
"Dave Platt" wrote in message
... In article om, Digi-Key catalog lists quite a few such (all with their own shafts, ready for panel mounting), but they aren't cheap. Mechanical rotary encoders are less expensive, but less precise (fewer counts per revolution) and possibly not as reliable or long-lived since they use mechanical contacts. Depending on the application, mechanical encoders are pretty good, and getting better. As a tuning control, in most cases something on the order of 50 pulses per rev is pretty useable and avaliable in a mechanical encoder. Much more than that and you are getting into the optical encoders which do get kind of pricey. They -feel- really nice, though. Most of the mechanical encoders Digikey carries are detented, so watch out for that. The few that aren't are pretty decent. Mechanical encoders are typically under $5, while the opticals are more like $50. But the nice ball-bearing feel might be worth it, depending on your project. ... |
Dave Platt wrote:
You'll probably want at least 64 counts per revolution, and probably 256, to get a nice smooth "feel" to the synthesizer tuning. Most such use an optical code wheel and a pair of optosensors. For something like a receiver tuning control, even 256 steps/rev would sound 'jumpy'. However, you could gear it up mechanically so that even a small movement of the control produces several pulses from the encoder. That is exactly what happens in a mouse. Try an old-fashioned cord drive, only in reverse. -- 73 from Ian G/GM3SEK 'In Practice' columnist for RadCom (RSGB) http://www.ifwtech.co.uk/g3sek |
"Ian White G/GM3SEK" wrote in message
... For something like a receiver tuning control, even 256 steps/rev would sound 'jumpy'. However, you could gear it up mechanically so that even a Actually, 256 is getting too high in most cases. The "jumpiness" comes from the size of the step, i.e., the number of Hz per step. Encoders between 50 and 100 are probably the easiest to deal with. You generally want to make the step size small, like 1 or 10 Hz, but you don't want hundreds of turns to cover the band. At very high resolutions, the pulses can come very fast, so it gets tricky to distinguish the closure from noise. Of course, higher counts can be made to work, and in principle, can be made to work better. But generally, one is dealing with low level software and finite compute resources to read the encoder, so the very high resolutions can actually become somewhat problematic. ... |
From: xpyttl on Jul 17, 5:56 pm
"Ian White G/GM3SEK" wrote in message ... For something like a receiver tuning control, even 256 steps/rev would sound 'jumpy'. However, you could gear it up mechanically so that even a Actually, 256 is getting too high in most cases. The "jumpiness" comes from the size of the step, i.e., the number of Hz per step. Encoders between 50 and 100 are probably the easiest to deal with. You generally want to make the step size small, like 1 or 10 Hz, but you don't want hundreds of turns to cover the band. Ahem, have you guys agreed on what defines "jumpiness?" :-) I'm into the actual receiver portion of a PLL-LO-controlled HF SW BC receiver tuning in 1 KHz steps. The PLL and its controlling circuitry are done and a 256-step Grayhill shaft encoder is used with a shaft encoder decoder circuit from Dr. Robert Dennis. That decoder circuit uses 3 HCMOS DIPs for Up/Down counters having separate Up and Down clocks. One more HCMOS DIP gate package is required for Up/Down counters having a single clock input and having an Up or Down mode control pin. At very high resolutions, the pulses can come very fast, so it gets tricky to distinguish the closure from noise. I didn't find it so. The Dennis Decoder will allow shaft encoder rotation rates of 150 RPM with a 240-step shaft encoder, no problem. Pressed faster, it can handle 300 RPM or 5 revolutions per second...quite fast tuning. It is easily comparable to the "old" tuning control on my Icom IC-R70 which has an estimated 200 steps per revolution. Of course, higher counts can be made to work, and in principle, can be made to work better. But generally, one is dealing with low level software and finite compute resources to read the encoder, so the very high resolutions can actually become somewhat problematic. While the BCD Up/Down counter ICs are getting rather scarce, it takes only three more packages to "see" three-decimal-digit resolution. Add another IC for four-decimal-digit resolution. 4-stage binary Up/Down counters are still being made, no problem. Ain't no software for this critter...it's all hardware. The Up/Down counter drives the PLL divider directly (through a EPROM translator, preprogrammed, or through some more logic gates). Grayhill shaft encoder output is TTL/HCMOS level at +5 VDC into the encoder. Very fast jumps in handling the shaft encoder knob don't upset the counter chain. If you want some nice decoder circuit schematics and a set of decoder waveforms, let me know at signature address below and I can attach the ZIP file (19 K) to e-mail. If someone wants an Absolute position control, that could be done with an additional optical sensor and track that resets the Up/Down counter at all-zeroes. Not the same as a Gray Code multiple track and sensor encoder with a collection of Exclusive-Or gates to make it straight binary, but cheaper. bit bit |
Seems to me that 256 or 512 is reasonable. Then decide how many
kHz/rev you want, and that tells you the step size. If you want 10kHz/rev, 512 steps would give you just under 20Hz/step, which seems reasonable to me. I wouldn't want it any coarser than that, and I'd actually prefer finer. Processors are very cheap; you can easily process the quadrature step info up to dozens of revolutions per second. If you budget ten instruction cycles to process each step, and you're using a slow processor at 1usec/instruction cycle, you can process 100,000 steps a second, if the processor has nothing else to do. You can process 10,000 steps a second, or 200kHz/second, with just ten percent of the processor's time. A nice "plus" is to accelerate the tuning when the steps come fast, so you might bump the tuning up to higher Hz/step as the steps come faster. Expect to spend some time on the algorithm to get a smooth "feel" to it, though. The accelerated tuning when the knob is turned more quickly is a feature commonly found in test instruments. I haven't ever had a problem with the HP/Agilent encoders with respect to the output looking like "noise." The outputs are very clean digital signals with no "bounce" if you're turning in one direction. Cheers, Tom PS...Tim, did you find an encoder yet? Drop me an email if not... |
One you can find at http://ru3ga.qrz.ru/UZLY/encod.htm
Follow the links on that page for more pictures. Sorry, the text is not in English. If anybody wants more details, let me know. 73, Ivan VE3IVM "Roger Leone" wrote in message ... Tim: There used to be a website with info on using surplus stepper motors as precision shaft encoders. I have been doing Google searches without success. Perhaps someone else will be able to provide a URL. It was Australian, I believe. Good luck. Roger K6XQ |
From: xpyttl on Jul 17, 5:56 pm
"Ian White G/GM3SEK" wrote in message ... For something like a receiver tuning control, even 256 steps/rev would sound 'jumpy'. However, you could gear it up mechanically so that even a Actually, 256 is getting too high in most cases. The "jumpiness" comes from the size of the step, i.e., the number of Hz per step. Encoders between 50 and 100 are probably the easiest to deal with. You generally want to make the step size small, like 1 or 10 Hz, but you don't want hundreds of turns to cover the band. I'm into the actual receiver portion of a PLL-LO-controlled HF SW BC receiver tuning in 1 KHz steps. The PLL and its controlling circuitry are done and a 256-step Grayhill shaft encoder is used with a shaft encoder decoder circuit from Dr. Robert Dennis. That decoder circuit uses 3 HCMOS DIPs for Up/Down counters having separate Up and Down clocks. One more HCMOS DIP gate package is required for Up/Down counters having a single clock input and having an Up or Down mode control pin. At very high resolutions, the pulses can come very fast, so it gets tricky to distinguish the closure from noise. I didn't find it so. The Dennis Decoder will allow shaft encoder rotation rates of 150 RPM with a 240-step shaft encoder, no problem. Pressed faster, it can handle 300 RPM or 5 revolutions per second...quite fast tuning. It is ADDENDA: The decoder circuit I used was from: http://www-personal.umich.edu/~bobde...re_decoder.pdf Two internal time-constants were about 10 times faster than what I'm using for the tuning control on my receiver project. I deliberately lengthened the two internal pulses used in decoding to better observe them on a scope. The Dennis Decoder is indicated as originating in 1998. [University of Michigan, not Michigan State University. :-) ] |
Well, "disgustingly cheap" but requiring MORE steps than a mouse
wheel can be done by a spare gear of around 40 or more teeth. I just tried that with an 80-tooth gear that's been in the junkbox for decades. Used an optical interrupter but had to Dremel open the spacing (cut the base of the plastic mounting). Gear teeth are cut with precision and even rejects are good enough for this kind of adaptation. About the only problem is that the openings aren't as abrupt as the mouse wheel...but that is easily overcome by a Schmitt trigger inverter...and perhaps some shielding from ambient light into the interrupters' photosensor. Used a Fairchild interrupter costing about $2, left over from another project. I mentioned the Robert Dennis circuit for its simplicity in using ALL the transitions possible of a sensor wheel. That's twice as many as a more common dual-D FF decoder circuit. Takes three packages versus one to get all the transitions for Up/Down outputs for direction sensing. NOT the space shuttle, yes. I've worked on the instrumentatin for the space shuttle main engines (at Rocketdyne) and understand the complexity and reliability needs for man-rated spaceflight. :-) |
| All times are GMT +1. The time now is 10:55 PM. |
Powered by vBulletin® Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
RadioBanter.com