Wednesday, 5 April 2017

Not all A3114 hall sensors are the same - who knew?

We were playing about with hall sensors again this week. We've used hall sensors a lot in the past, and had a bunch of A3114 sensors left over from previous projects. But there were only a couple left and the massive bag of left-overs was somewhere in a box in the black hole that the bungalow workshop has quickly become.

A few clicks on AliExpress later and we had some more A3114 sensors sent within just five days. We threw the lot together in a little component drawer and got on with making our project.

Hall sensors are often used as limit switches, but don't suffer from the problems that mechanical switches often do in dusty environments - namely there are no moving parts to get gunked up with dust, and no way the switch can get jammed. But when we tried using them, we got some weird results.

Some hall sensors plain simply didn't work.
Some triggered from about three inches away!
Some worked as we expected, triggering when a neodymium magnet approached from about 5mm away. And some acted less like switches and more like variable/analogue devices, with the output increasing in intensity as the magnet was moved closer.

To get to the bottom of things we created a simple hall sensor tester from a battery, an LED and a socket (into which we plugged our different hall sensors to try them out).

The first sensor in the video demonstrates how we expected the hall sensors to work; introduce a magnet and at a certain distance, the sensor acts like a switch and the LED lights up (in the video it appears to fade up quickly, but that's the camera auto-light-adjustment; in real life it switches almost instantly).

The last sensor in the video - although not immediately obvious in the film - appeared to work a tiny, tiny amount; if you looked right inside the LED, a tiny little dot of light was just about perceptible, when the magnet was right up against the sensor.

The second sensor in the video had us puzzled.
Not because it triggers from a long way away, but because it appears to have an almost-analogue-like behaviour - the intensity of the light increases/decreases as the magnet is moved towards/away from the sensor. The reason this was particularly puzzling is because A3114 sensors are supposed to have an inbuilt hysteresis.

The A3114 is supposed to have a "trigger" and "release" magnetic flux density with a "dead band" which reduces any "chatter" that might occur just at the point where the switch would normally activate (similar to the bounce in a mechanical switch).

Yet the second sensor in the video doesn't display either a trigger or a release threshold - the intensity of the LED changes in relation to the distance from the sensor. Which makes us wonder - what on earth kind of sensor is it?!

On closer inspection, we found that the sensors that worked as we expected them to were labelled 3114/515 and 3114/OH15.

The newer sensors are labelled 3114/402.
Which suggests that not all 3114 hall sensors are the same.

Who knew?