Hall effect sensor help

[Mark Schilling]

Hey guys. I'm trying to build a handbrake for a friend using an old xbox 360 wired controller as the interface.
What I wanted to do was replace either one of the triggers, or one of the thumbstick axis with a hall effect sensor.

These are the ones I got..

https://amzn.eu/d/gPdWpIj

Can't get it to detect though. Anything I should know to get this to work, or is it not possible? I replaced the throttle pot in my fanatec csl elite pedals in the past, so thought that it should work in principle anyway ‍

 

[drvrn]

I used some hall effect sensors for my seat belt tensioner and from the batch that I bought from Amazon, 2 out of 10 were faulty. So better try with some others from that batch in case.

And of course, check wiring of the 3 wires.

And check with other magnet and distance to see if signal is working.

 

[Mark Schilling]

I used some hall effect sensors for my seat belt tensioner and from the batch that I bought from Amazon, 2 out of 10 were faulty. So better try with some others from that batch in case.

And of course, check wiring of the 3 wires.

And check with other magnet and distance to see if signal is working.

I'll try that. Thanks

 

[Neilski]

I used some hall effect sensors for my seat belt tensioner and from the batch that I bought from Amazon, 2 out of 10 were faulty. So better try with some others from that batch in case.

And of course, check wiring of the 3 wires.

And check with other magnet and distance to see if signal is working.

Yes, and if it hasn't been tried already then I suggest the testing should include a basic bench test without the controller involved, just to make sure the sensor is happy.
I have no idea what voltage that controller supplies to the original pots, but the Hall sensor above has a fairly wide range so I'd have guessed that it should work OK. (It's not as sensitive as something like the A1324 but it is likely still plenty good enough.)

 

[Mark Schilling]

Yes, and if it hasn't been tried already then I suggest the testing should include a basic bench test without the controller involved, just to make sure the sensor is happy.
I have no idea what voltage that controller supplies to the original pots, but the Hall sensor above has a fairly wide range so I'd have guessed that it should work OK. (It's not as sensitive as something like the A1324 but it is likely still plenty good enough.)

Excuse my ignorance, but how would I go about doing that? I have a multimeter, but not sure how I'd test these sensors.

 

[Neilski]

Excuse my ignorance, but how would I go about doing that? I have a multimeter, but not sure how I'd test these sensors.

No problem, everybody starts out ignorant until they learn stuff somehow.
I would suggest hooking a 5 V supply (e.g. straight from a USB port or similar) to the GND and VCC pins, and then measure the voltage between GND and the output pin with a multimeter as you bring a magnet from far away to touching the front or back face of the sensor. With no magnet nearby you should get 2.5 V and it should go up or down with a magnet close by, depending on the orientation of the magnet.

 

[Mark Schilling]

No problem, everybody starts out ignorant until they learn stuff somehow.
I would suggest hooking a 5 V supply (e.g. straight from a USB port or similar) to the GND and VCC pins, and then measure the voltage between GND and the output pin with a multimeter as you bring a magnet from far away to touching the front or back face of the sensor. With no magnet nearby you should get 2.5 V and it should go up or down with a magnet close by, depending on the orientation of the magnet.

Cheers for the info mate. I'll test that

 

[Mark Schilling]

I get 1.6v with no magnet and when the Magnet gets right against it it's 2.5v

 

[Neilski]

I get 1.6v with no magnet and when the Magnet gets right against it it's 2.5v

That's with a 5 V supply? Bit surprising but at least the sensor is doing something. It would make total sense if the supply was more like 3.3 V though.
Did you buy more than one sensor? If so you might want to see if they all behave the same.
I guess that in your shoes, the next thing I'd want to check is if the in-circuit voltage across the sensor is more or less the same as the open-circuit voltage, i.e. the voltage across the GND/VCC wires then the sensor isn't hooked up (to check the really low probability that the sensor is trying to draw more current than the wires to it are able to provide, given that it pulls more current than say a 10 k pot). You can also monitor the voltage on the output pin while the controller is running and see if it changes on the multimeter but not on the PC.

Also I'm belatedly realising that I may have misinterpreted you when you said:

Can't get it to detect though.

Did this mean that nothing happened on screen when it should have, or that the controller literally wasn't detected by the PC when you installed the sensor?

 

[GeekyDeaks]

I get 1.6v with no magnet and when the Magnet gets right against it it's 2.5v

looking at the datasheet for the OH49E, that would be the expected output for a 3.3v supply as @Neilski mentioned. Try flipping the magnet over to the other pole and making sure the voltage drops to about 0.6v to check the full range of the sensor. I'm guessing you are replacing a pot? If so, along with the suggestions provided, you might want to measure the min/max output of the original pot to determine the range you need for the controller first, then determine how to arrange the sensor and magnets to provide a similar signal.

Also, @Neilski has a very good point about the power draw. The datasheet says approx 4.2mA at 5V, so that would give an internal resistance of about 1.2k which might cause some issues

 

[Mark Schilling]

That's with a 5 V supply? Bit surprising but at least the sensor is doing something. It would make total sense if the supply was more like 3.3 V though.
Did you buy more than one sensor? If so you might want to see if they all behave the same.
I guess that in your shoes, the next thing I'd want to check is if the in-circuit voltage across the sensor is more or less the same as the open-circuit voltage, i.e. the voltage across the GND/VCC wires then the sensor isn't hooked up (to check the really low probability that the sensor is trying to draw more current than the wires to it are able to provide, given that it pulls more current than say a 10 k pot). You can also monitor the voltage on the output pin while the controller is running and see if it changes on the multimeter but not on the PC.

Also I'm belatedly realising that I may have misinterpreted you when you said:

Did this mean that nothing happened on screen when it should have, or that the controller literally wasn't detected by the PC when you installed the sensor?

Controller was detected, but no movement registered on that axis.
I'm tapping the voltage from the point the red wire connects to the controller. I assumed it was 5v, but when I just put the multimeter from that point to ground it says 3.3v

 

[Mark Schilling]

With the magnet flips it drops to 0.8v when touching the sensor

 

[GeekyDeaks]

With the magnet flips it drops to 0.8v when touching the sensor

That sounds like it's working fine to me. The voltage swing might be too narrow for the controller though. Are you able to test with the original pot?

 

[Mark Schilling]

I tested the analog stick. Centered is 1.6v. Fully deflected is 3.3v on one end and 0.1v on the other

 

[GeekyDeaks]

Ouch, that sounds like you are not going to get the full range then, but you should still be getting something. I'm guessing you have already tried testing the sensor voltage on the ouput pin whilst it is connected to the controller? If so, and the voltages are changing with no difference in the output can I ask what you are using to monitor the controller values?

 

[Mark Schilling]

Ouch, that sounds like you are not going to get the full range then, but you should still be getting something. I'm guessing you have already tried testing the sensor voltage on the ouput pin whilst it is connected to the controller? If so, and the voltages are changing with no difference in the output can I ask what you are using to monitor the controller values?

I was using the default windows usb game controllers profiler to measure if I was getting a response.

I've ordered an Arduino pro micro. Hopefully with that and simhub I can setup the sensor as an axis and specify the output range.