Hall effect potentiometer

Hi folks, I have found that the rotary 10k pots I'm using are failing in fairly short timescales (2 - 3 weeks). When I replace them everything works fine for another short while, and then I start to get erratic positioning which gradually gets worse. They are mounted directly onto the back of the actuator shaft. I have used cheap 1 turn pots and continuous turn pots which have all failed.
Now to my question. Have any of our members used Hall effect, continuous turn potentiometers as a straight replacement for rotary linear pots. If so, how effective are they, and did you find any drawbacks using them? I have found "rs-online" stock them but they are expensive - 44GBP. I want to know they will do the job before I part with the cash!

Hi I'm testing 180 degree hall sensor, on 5v, output is 0.4v to 4.8v mounted directly on the back of the gearbox.
also they do not have any mechanical stops, so you cant damage them if the system goes ape.
about £15.00 each from Farnell, If any good I will dig out the details

Great telfel sounds ideal.

Mike

I've tested the Hall effect one,that thanos mention on his first post of AMC 1.6
It work properly in 180 degree long life

hi link to Farnell UK
http://uk.farnell.com/jsp/search/productdetail.jsp?SKU=2319661

regards Terry

Thanks for that Terry. It looks just what I want.

Cheers, Mike

do the hall effect pots hook up the same as regular ones? wiring wise?

I haven’t used these myself but they should function like a pot with a few exceptions. One, you only get 92% range of the input voltage (so 5v input = .2v min and 4.98v max output). Two, they wire like a pot but you can’t invert the positive and negative wires to invert the output, it must be inverted in SimTools if necessary. Third, since it is digital internally, it may technically not be as linear as a pot.

The positives are 12bit resolution with .2% drift which is equivalent to a drift of +.1 or -.1 degrees or so and accuracy to .088 degrees. Also they are available in 360, 180, and 90 degrees turn and each at 12bit resolution. So if you only need 90 degrees and use the 90 degree unit, you will still be using 12bits without any reduction as you would get with only using 90 degrees of a 360 unit (equal to 10bit output). Rotational life is 10 million shaft turns. The rotation is continuous, there are no stops unless you special order them. Therefore you must make sure they wont flip from 360 to 0 or 0 to 360 degrees. Also, since they are digital internally, there is no noise on the output!

I’m going to order some and check them out because they sound like a good option. And a great option over using most pots!

BlazinH pretty much covers it. I have only used them in DIY applications such as custom joysticks that I have made. Using hall effect sensors, rare earth magnets and u-joints the results are extremely accurate with no mechanical "noise". They are very precise.

I do like the idea of using them with SimTools as well, especially the unlimited turning capability. I broke a couple of cheap 180 pots while tinkering with new motor configurations.

I have ordered 2 and will swap out my cheapo pots when they arrive. I'll leave feedback regarding how effective they are.

Mike

Hi guys,
My Hall-effect pots arrived and I have replaced a continuous turn Potentiometer which had failed. I calibrated it using the JRK configuration program, and so far I am very impressed. The movement is very smooth and it seems to keep position extremely well. It is so good that it shows the other mechanical pot to be crap. So I will have to swap that out as well. I'll update this post if there is anything iffy about their performance. At the Moment I would say that they are a very good replacement for mechanical pots. See telfel's post above for the link.

Mike

Thank You Terry for the info and the link!

Thanks for the feedback

Thanks for the info!

So basically, they connects same as regular pot, just keeping eye to connect right wires, not swap them. I didn't fully understand 180* turn, so it does have unlimited turning (full circles) but it gives feedback for just half circle? Is there marked position from where to where is it counting?

Yes, thanks for the info. I had a thought for mine, I use about 1.8 turns of a 3 turn pot because of the gears. So what hall effect would I use? Are there 3 turn hall effects?

I set my actuator arms to a level position and turned the pot until I had the middle voltage - 2.5volts. Then screwed the pot up tight so the body wouln't move. Then I used the JRK config program
to set the limits of movement, about 170 degrees.

I wouldn't think so Dave, as I believe the voltage is generated by the position of the sensor in a magnetic field.
I would think one turn is all you could get. But I'm sure to be corrected if I'm wrong. Mike

I can switch the gears around, so that would solve the problem of turns

@bsft AceofSpies is correct that the output is limited to one turn. The output is limited but the rotation is not. Therefore, it would be possible to use them in multi turn situations. However, it would require the use of custom software or firmware to monitor for a rotation switch over and then mathematically account for it. It would be similar to an incremental encoder that uses an index to keep track of how many revolutions it has made.

Sure you could use gearing to reduce the turns but if they have any backlash you will get additional “slop” added in of course. These are so accurate that it doesn’t take much to register either. However, also because these are so accurate (to about 4095 units per revolution for a 360 degree pot), one turn is quite sufficient for most applications, at least for mine (for the record I use PWM output hall effect encoders, not analog, but specs are the same, 12bit). Again, they are accurate to +- .088 degrees for a 360 degree pot and have a "clean" output.

ps I ordered the 360 and the 90 degree hall effect pots to experiment with. But I don't expect to receive them until sometime later in the week. I plan on trying them on a seat mover design.

I have used them too but found them quiet expensive to buy, I went to 5 turn contactless pots from RS components simualar price but with the winch setup I had them on I want a 5 turn arrangement as in 80 deg of lever movement equalled 5 turns of the pot, work well giving me a very fine resolution of movement that I could not get with a single turn hall effect. To me Hall Effects are good but at the time I was testing them with the Jrks , I was not satisfied with the resolution I was getting from high qualily 5 turn units. Higher the resolution the better you sim movements will be.

I can change the gears around on my motors so I get less of a turn on pot end, easy as.