Reply with Quotehelpful link to understand the linear encoder principle of working.
Electromagnetic Linear Encoders to the Rescue | Control Engineering
Idea - DIY Linear Magnetic Encoder
Hi,
Can the old-timer type magnetic cassette and head be used to make a DIY linear encoder?
Since commercial linear encoders work on the principle of alternating north and south poles on the strip and a head that reads the pitch between two magnetic poles on the strip. This pitch represents a certain linear distance and is calibrated into the head electronics to give an output signal equal to the linear distance (mostly digital signal)
Look at the magnetic cassette, head and the commercial Magnetic strip type linear scale. Don`t you see a possibility ?
Similar Threads:
- HOW TO CHANGE ACTIVE ENCODER FROM LINEAR SCALE TO MOTOR ENCODER IN SIEMENS 840D.
- Need Help!- Error 1170...MCS magnetic encoder alaram OSP E100L Macturn
- Cheap magnetic DRO as linear encoder?
- Need Help!- Magnetic Linear Encoder Reading Head - with 5 Micron
- Any one have experiance with linear magnetic encoders?
helpful link to understand the linear encoder principle of working.
Electromagnetic Linear Encoders to the Rescue | Control Engineering
This is an Hall effect sensor manufactured by Honneywell
I like your idea, It should work if the magnetic head/sensor can work at very low speed.Originally Posted by axis_control
Hi,
Can the old-timer type magnetic cassette and head be used to make a DIY linear encoder?
Since commercial linear encoders work on the principle of alternating north and south poles on the strip and a head that reads the pitch between two magnetic poles on the strip. This pitch represents a certain linear distance and is calibrated into the head electronics to give an output signal equal to the linear distance (mostly digital signal)
Look at the magnetic cassette, head and the commercial Magnetic strip type linear scale. Don`t you see a possibility ?
I think you will have 3 problems;
1. resolution. The tape head is designed to have metal poles so it can pick up smaller sized magnetic details (ie high audio frequencies). But the Hall sensor does not have that ability and it's silicon sensor is a couple mm deep inside the body so the smallest magnetic detail on the tape it can pickup is going tobe large (maybe 2mm?).
2. Quadrature. To work properly as an encoder you need quadrature output which means 2 sensors mounted 90' out of phase in comparison to the magnetic detail.
3. Signal strength. The magnetic field on the tape is VERY small! Tape heads use very sensitive pickup coils and massive amounts of audio AC gain. The hall sensors are meant to pickup a much larger but constant (DC) magnetic fields. It's going to be difficult to get it reliable.
There are also some issues of the tape being very delicate and the tape field degrading over time.
A better option might be to use a fine metal "rack" with teeth and 2 hall proximity sensors that can detect the rack teeth. Like the tape you may still be limited to about 1mm or 2mm resolution but it would be a much better system.
Considering you can buy cheap DROs for $30 or so and just hack the signal from one that is still the best option as you can get high resolution.
First of all, I am Thankful to you for pointing me to the right direction.
Reading your reply, I have concluded that the idea of using the cassette tape and head is useless and there is nothing to do with it.
If we are looking for 1 micron precession, then I guess a mechanical rack teeth design won`t be able to give that unless we had nano tech to fabricate the profile.
I checked on ebay for DRO and mostly the cheap ones are second hand. If I can get a new one for $30 then I would go banana`s. I am sure you are referring to a second hand one. I got a quote from a local manufacturer (he says so) for a 300 mm travel Glass scale linear encoder for $ 305 (USD). The accuracy specified was 5 microns.
Apart from the above I checked out some Chinese products on alibaba and some other business to business websites. The thing that scares me most about buying these products is the reliability and the other is getting a different product when you have ordered one. I have seen some people being cheated for good.
heres one I was looking at
Buy Linear scale, linear glass scale, linear encoder scale, FREE SHIPPING Linear scale DC10 301-500mm at Aliexpress.com
Most companies buy components from large manufacturers with resources and assemble it at their site and put a brand name of their own and sell it as their product. Using the same concept I think it is possible to assemble a magnetic or glass linear scale system.
I am searching on the internet for suppliers of
1] Encoder strip.
2] Encoder strip sensor.
I am getting good results (at least for now and the accuracy is unknown)
Do printers have magnetic linear scale inside them?
I have been getting quite a lot of sites pointing at printer components when I am looking for linear magnetic scales.
If Mitotoyo is available in your part of the world, then check on the measuring scales linear tape with quadrature output.
Particularly the ST24/44 and ST36.
http://www.mitutoyo.ca/files/2009%20...0section_0.pdf
Al.
CNC, Mechatronics Integration and Custom Machine Design
“Logic will get you from A to B. Imagination will take you everywhere.”
Albert E.
Al Thanks for the info.
By the way I loved your Avatar before this one, the baby with a cigarette
It was outstanding and one could remember you with that Avatar.
The new avatar is good too and projects one of your work side that is watching out. I know you do much more than watching out and that is helping so many like me on this forum.
No, that's new. I received a junk mail a while back that had Chinese 500mm DRO for about $30 (if I remember right)....
I checked on ebay for DRO and mostly the cheap ones are second hand. If I can get a new one for $30 then I would go banana`s
...
I expect it would be as accurate as any of those Chinese verniers with digital read outs, the electronics bit looked like the same type of thing but with a longer "stick".
Not an expert, but many printers seem to have an optical linear scale inside.
Tom
all inkjet printers do have linear or disk encoders.
None of the laser printers have such thing.
1. why not using the original tape head? if it can read audio frequency (20kHz). If I've made the computations right: 20kHz with original speed of 5cm/sec means 1 pulse/2.5um - way lower than needed in home-made CNC.
2. using stereo tape head you can create the quadrature pulses. It's easy to create two square signals in quadrature and record them to the tape.
3. I have doubts only about very low-speed movements. The lower the speed the higher the gain needed in the amplifier for the tape head.
I think it worth a try.
I was almost on cloud nine while reading the start of your reply until I reached the chinese vernierIt brought some old memories aka wounds back. I remember a friend of mine being exited about getting a digital vernier cheap and I got exited hearing that and decided to buy one even before I had seen one. It was later that my friend started complaining about this vernier that always had different readings for the same component and working in R&D and designing machines with such a vernier was a wrong decision then. It made us go in a loop making mistakes with GD&T and ruining some machined components.
The reason I am hesitating to buy the chinese ones is because my requirement is severe. I am currently designing a cnc lathe with ID & OD grinding attachments on a parallel integrated bed. I want to make the grinding process a closed loop system. I plan to take the linear scale reading and integrate it with the control system. I still need to figure out a lot in that.
I am in a debut situation. I know if I compromise on quality I am gonna get a kick on my butt later, but some times I tend to be a cheap assor may be not, and I am just trying to figure out the possibilities. Who knows if it works out and a whole community here benefits out of it
edited :
Ok , I think I would take a look at the email in case you can post a link here. Who knows about the possibilities ..lol
Tomorrow morning I am off to get myself an old scraped printer. Lets see whats inside. May be it can turn out useful if the accuracy is available.
Good going there eSilviu. What resources will one need to do an experiment like that?. I am not an electronics eng, but can pull some strings to arrange couple of things in order to get started. We pass or fail, no matter but lets give it a try.
Yes you could make such a system work BUT and it is a very big but you need to source some non stretching magnetic tape and to accurately mark the tape with the reference signal or signals for a quadrature or coded position signal. The tape use for audio recordings stretches under modest tension. In the early days of magnetic recording fine iron wire was used but this will mean building special recording heads as the magnetic flux will be many time higher than that required for modern audio tape. The poor high frequency response of the wire recording will limit the accuracy obtainable as will your ability to lay down an accurate reference track.
Sorry but suggest you use optical sensors.
Regards
Pat
Last edited by wildwestpat; 04-12-2011 at 11:47 AM. Reason: Typos
The tape head produces a signal caused by tape speed, it will produce no signal with a stopped or very slow tape.
Interesting, but see 1. above, also there will be a phase displacement of the signal which is the very last thing you need with a quadreature sensor.
I think it's not. Feel free to prove me wrong.
I'll also throw in a problem 4; The tape length changes quite a lot with temperature. In a casette this does not matter as the pinch roller ensures relative tape:head speed and there are tensioning systems. In a delicate tape stretched across a machine it might be nasty...
Take a look at this artical from 1977...
Popular Science - Google Books
