Could a High Temperature Absolute Encoder Fill a Niche Market?

[barryn56]

Hi,
as part of a project I needed a slim (<10mm) small diameter (<30mm) through hole absolute encoder that works at 100 degC. I wanted this to be able to determine a motorized valve position, so it needed to slip onto the gearbox output shaft between the motor and the valve.

Anyway, I couldn't find anything on the market, so I designed and built my own. I'm wondering if this would have a market in its own right? Being slim, you simply slip it onto the output shaft of a motor and it gives you an absolute angular position in 7 bit parallel digital form. The version I have made is 7 bits, so about 3 degrees resolution. If you have an incremental encoder at the other end of the motor, you can use this to improve the resolution, or, alternatively, I have designs for higher bits. Cost to build is under $50.

Given there's nothing on the market that I could find, maybe this is too esoteric - anyone out there know if this would be useful?

 

[berkeman]

Hi,
as part of a project I needed a slim (<10mm) small diameter (<30mm) through hole absolute encoder that works at 100 degC. I wanted this to be able to determine a motorized valve position, so it needed to slip onto the gearbox output shaft between the motor and the valve.

Anyway, I couldn't find anything on the market, so I designed and built my own. I'm wondering if this would have a market in its own right? Being slim, you simply slip it onto the output shaft of a motor and it gives you an absolute angular position in 7 bit parallel digital form. The version I have made is 7 bits, so about 3 degrees resolution. If you have an incremental encoder at the other end of the motor, you can use this to improve the resolution, or, alternatively, I have designs for higher bits. Cost to build is under $50.

Given there's nothing on the market that I could find, maybe this is too esoteric - anyone out there know if this would be useful?

This post could have been deleted as spam, but based on several things, I'm going to leave it posted for now. I'm interested in any responses from the forum regulars...


EDIT -- Can you post some pictures or other application information? Thanks.

 

[barryn56]

Good point, Berkeman - I didn't mean in any way to post this as an offer of sale of any kind, and appreciate you posting it - I'll be more careful in future.

 

[zoanie]

Hi Barryn56,

I'm interested. I'm looking for low cost absolute postion encoder. 7 bits however is a bit too low. I would like to find something with 12-bit resolution and a minimum sample period of perhaps 100ms.

Carlos Navarrette
Los Angeles, CA

 

[barryn56]

Hi Carlos,

there are two ways to get that resolution. You can use the combination of the absolute encoder and a separate incremental - they can be on opposite ends of a motor/gearbox assembly - so the resolution will be based on the resolution of the incremental encoder. Say you've got a 50 to 1 ratio gearbox on the motor, so the incremental encoder does 50 revolutions for every turn of the absolute encoder. If the former has 1200 lines per revolution, then for every 3 degrees you get 500 pulses. So if your absolute encoder changes from 37 to 40 degrees after 101 pulses, you know your actual start angle was 39.394 degrees - the penalty is that you have to wait for the first transition to get an accurate reading.

The other option would be to just use the absolute encoder with the increased number of sensors. I'll see what 12 bits would look like (optical, so shouldn't be an issue) but diameter might need to go up a little to accommodate the extra sensors - do you have a preferred maximum diameter?

Not quite sure what you mean by minimum sample period, but suspect this is to do with incremental encoders (??- let me know if not). With (this) absolute one, when you apply power you get a parallel digital output of the angular position, so you can sample it as fast as you like.

Regards,

Barry

 

[zoanie]

It is an intriguing idea to combine the two types to increase the resolution, however I still like working with a single absolute encoder, its much easier to diagnose when something goes wrong. For a variety of reasons for my purpose (antenna systems) this piece of equipment seems to start at a price around $600. I'm browsing for an alternative.

 

[barryn56]

Hi Carlos,

I've got some preliminary designs checked out for an 11 bit encoder - this will be 30mm disc, so outside housing, say, 40mm and thickness around 10mm. Laser cutting the encoder disc is pretty cheap, so I think construction cost all up is going to be in the $100 range, main cost is the optical sensor/irleds and pcbs. We'll build a few and see what issues we find and let you know how it goes. May take a while as we have some other work on at present.

Due to the high precision needed for the sensor positioning compared with 7 bits, we're looking at incorporating a mask layer as well. I guess your application doesn't really need the high temperature operation, but interested to know what sort of output you would normally need (serial?), and what size shaft would be going through, as this increases the OD in relation- I'd suggest 10mm (or 3/8") as a standard for a 30mm diameter unit.

If you want to take this off-forum, feel free to send me email direct - hope this isn't against the rules...

Regards,

Barry