Offline
Full Member
Karma: 1
Posts: 129
|
 |
« on: July 16, 2012, 10:20:09 am » |
This is for several uses on a robot. I am looking for a sensor that provides a code for the angular position of a shaft - such as a steering shaft or a disk that rotates a camera mounted on it. The idea is to determine the angle when the shaft is NOT turning. This is not for detecting high-speed changes. What is available?
Years ago, this was done with a shaft encoder - a metal disk with printed patterns that passed under a light sensor as it rotated. The position coding was obviously not very accurate, and got worse as the disk got smaller in diameter. I would like something that is a bit more modern in technology. What's available?
|
|
|
|
« Last Edit: July 16, 2012, 10:54:00 am by louarnold »
|
Logged
|
|
|
|
|
Massachusetts, USA
Offline
Tesla Member
Karma: 96
Posts: 6364
|
 |
« Reply #1 on: July 16, 2012, 03:05:07 pm » |
If you have access to the end of the shaft you can mount a magnet and use a special Hell-effect angle encoder.
What you want is called an Absolute Shaft Encoder.
|
|
|
|
|
Logged
|
|
|
|
|
Global Moderator
UK
Offline
Brattain Member
Karma: 137
Posts: 19016
I don't think you connected the grounds, Dave.
|
 |
« Reply #2 on: July 16, 2012, 03:31:26 pm » |
special Hell-effect angle encoder "Hall Effect"
|
|
|
|
|
Logged
|
Pete, it's a fool looks for logic in the chambers of the human heart.
|
|
|
|
Massachusetts, USA
Offline
Tesla Member
Karma: 96
Posts: 6364
|
 |
« Reply #3 on: July 16, 2012, 04:42:01 pm » |
special Hell-effect angle encoder "Hall Effect" Oops. 
|
|
|
|
|
Logged
|
|
|
|
|
Offline
Full Member
Karma: 1
Posts: 129
|
 |
« Reply #4 on: July 17, 2012, 10:31:28 am » |
If you have access to the end of the shaft you can mount a magnet and use a special Hell-effect angle encoder.
What you want is called an Absolute Shaft Encoder.
Well, the hall effect sensor requires rotation - it senses speed. I do want an absolute shaft encoder, but not in the old style that uses a coded pattern on a disk. I wanted something using a more modern technology - something that sense changes in fractions of a degree, perhaps. Does such a device exist?
|
|
|
|
|
Logged
|
|
|
|
|
Global Moderator
UK
Offline
Brattain Member
Karma: 137
Posts: 19016
I don't think you connected the grounds, Dave.
|
 |
« Reply #5 on: July 17, 2012, 10:46:32 am » |
it senses speed. No, it senses magnetic fields. You need a shaft encoder. I don't know of any solid-state ones, so a Gray code encoder is probably the only sensible absolute position sensor.
|
|
|
|
|
Logged
|
Pete, it's a fool looks for logic in the chambers of the human heart.
|
|
|
|
Offline
Full Member
Karma: 1
Posts: 129
|
 |
« Reply #6 on: July 17, 2012, 11:10:53 am » |
it senses speed. No, it senses magnetic fields. You need a shaft encoder. I don't know of any solid-state ones, so a Gray code encoder is probably the only sensible absolute position sensor. OK. I'll keep my eyes open. Thanks very much.
|
|
|
|
|
Logged
|
|
|
|
|
|
|
Offline
Full Member
Karma: 1
Posts: 129
|
 |
« Reply #8 on: July 17, 2012, 01:48:49 pm » |
Ahhh, that's a good question. I'm not really sure about angular resolution. Placing a gear around a steering shaft and using a stepping-motor driven worm gear, what would be a realistic resolution? Would 1 degree be enough? I wonder. The sources are good. I'll have to look at them more closely, but now I know that they really exist, so thanks very much. I note they seem to be ICs. How do they connect to a shaft? Can a hall-effect device be made to sense angular position without having rotation - so that when the device powers up, it knows what direction the single tire steering wheel is pointing?
|
|
|
|
« Last Edit: July 17, 2012, 01:58:06 pm by louarnold »
|
Logged
|
|
|
|
|
Massachusetts, USA
Offline
Tesla Member
Karma: 96
Posts: 6364
|
 |
« Reply #9 on: July 17, 2012, 02:29:59 pm » |
> Placing a gear around a steering shaft and using a stepping-motor driven worm gear, what would be a realistic resolution?
If your stepper does 200 steps per revolution and your worm gear has 100 teeth that is 20,000 positions. Perhaps you would be happy with 4096 positions.
> How do they connect to a shaft?
Magnet on END of shaft. IC centered over magnet.
> Can a hall-effect device be made to sense angular position without having rotation - so that when the device powers up, it knows what direction the single tire steering wheel is pointing?
Yes. That's why they are called ABSOLUTE angular encoders. They tell you absolutely which way the shaft is pointing.
|
|
|
|
|
Logged
|
|
|
|
|
Offline
Full Member
Karma: 1
Posts: 129
|
 |
« Reply #10 on: July 17, 2012, 03:01:30 pm » |
> Placing a gear around a steering shaft and using a stepping-motor driven worm gear, what would be a realistic resolution?
If your stepper does 200 steps per revolution and your worm gear has 100 teeth that is 20,000 positions. Perhaps you would be happy with 4096 positions.
> How do they connect to a shaft?
Magnet on END of shaft. IC centered over magnet.
> Can a hall-effect device be made to sense angular position without having rotation - so that when the device powers up, it knows what direction the single tire steering wheel is pointing?
Yes. That's why they are called ABSOLUTE angular encoders. They tell you absolutely which way the shaft is pointing.
I think you meant the shaft gear when you said 100 teeth. Geeee, Is that enough? (Just joking. I should have done the math before asking.) For the magnet on the end of the shaft...any pictures, what does it look like when it's attached? I assume the sensor is attached so the magnet rotates under/over it?
|
|
|
|
|
Logged
|
|
|
|
|
Massachusetts, USA
Offline
Tesla Member
Karma: 96
Posts: 6364
|
 |
« Reply #11 on: July 17, 2012, 03:10:43 pm » |
> For the magnet on the end of the shaft...any pictures, what does it look like when it's attached? Instructions for mounting a Magnetic Rotary Encoder 1.0 of a GM3 gear motor: http://reprap.org/wiki/Magnetic_Rotary_Encoder_1.0> I assume the sensor is attached so the magnet rotates under/over it? Yes. Either under or over. They show a cylindrical magnet but it is magnetized across a diameter, not from flat to flat as most cylindrical magnets.
|
|
|
|
|
Logged
|
|
|
|
|
Offline
Full Member
Karma: 1
Posts: 129
|
 |
« Reply #12 on: July 17, 2012, 06:10:46 pm » |
Yes. Either under or over. They show a cylindrical magnet but it is magnetized across a diameter, not from flat to flat as most cylindrical magnets.
Its not quite clear in the picture. The magnet and motor seem to be in a plastic case, but that isn't mounted close to the sensor. I assume that's not easy to show. However, the URL is great to have. I'll book mark it.
|
|
|
|
|
Logged
|
|
|
|
|
Massachusetts, USA
Offline
Tesla Member
Karma: 96
Posts: 6364
|
 |
« Reply #13 on: July 17, 2012, 06:47:38 pm » |
|
|
|
|
|
Logged
|
|
|
|
|
Offline
Full Member
Karma: 1
Posts: 129
|
 |
« Reply #14 on: July 18, 2012, 01:36:00 am » |
That mounting system will certainly work for what I have. I bookmarked all the URLs as well as the FRC home page. Thanks very much! Regards, Lou.
|
|
|
|
|
Logged
|
|
|
|
|
|