Bench testing 41F hall sensors

[salvage]

Hello everyone!

I am attempting to bench test some brand new SF41F hall effect sensors I got from aliexpress using Tommycat's thread as a guide.

I followed the diagram and wired up one of the sensors to a 5V wall adapter, which provides stable 5.44V. The sensors supply and ground are directly connected to the power supply. Here I am measuring the power supply's voltage


I combined a bunch of resistors in series to create a 10Kohm resistor as per the guide.


I then connected the sensors signal wire to the 5V supply through the 10Kohm resistor and connected the multimeter leads to the power supply ground and hall sensors signal line respectively.


All of the connections were soldered as to avoid any bad contacts and I tested 5 different sensors, but for all of them I saw no voltage change when putting a magnet (one from a 1000w motor) right next to the sensor in all kinds of different orientations.

I feel like I am making some mistake in the testing methodology or something else is wrong because I doubt that all of the sensors could be defective.

Any help or suggestion would be appreciated, thanks in advance.

 

[Zambam]

The magnetic field has to be perpendicular to the face (or back) of the device according to the data sheet. Did you try that?

 

[fechter]

It looks like you have it wired up correctly. Try flipping the magnet over to expose the sensor to both polarities of magnetic field.

 

[salvage]

Thank you very much for the quick responses.

I did try holding the magnet in all conceivable orientations. I even stack up a multiple magnets in attempt to create a stronger magnetic field, but none of the 10 sensors react.

I attached two videos - one where I test some old 502A sensor and it clearly reacts to the magnets, the other one if of the new 41F sensor which doesn't do anything regardless of how the magnetic field is oriented.

https://imgur.com/a/8mGUrgI

I have to be honest this have driven me a bit mad because this is the first time I am doing any ebike motor repair and I did not expect brand new semiconductor components to simply not work. Currently the only sensible conclusion I have is that the entire batch of the new sensors are defective. Judging by the price (1.80eur for 10pcs), they are some bottom of the barrel quality so I guess I got what I paid for, but I certainly did not expect this. I will try sourcing them from some more reputable source.

 

[TommyCat]

Just for fun does manually shorting the hall sensor’s signal leg to ground leg leads give you your voltage change in your original meter testing set-up? (Right up close to sensor test points, like below…)


Then just to be sure, I'd check the voltages right on/at the hall sensor's leads. Correct power to the sensor with good ground?
Correct voltage at signal leg?


A spring testing clip is handy for testing at these locations...

Try a different magnet with obvious and dependable north and south poles.




Regards,
T.C.

 

[fechter]

Possibly they sent you the wrong part? I've had that happen before.

 

[salvage]

Thank you for all the suggestions. I just bought some new sensors from a more reputable local source (I think they are genuine honeywell ones) and they all work fine using the test setup I used. So it seems that the aliexpress batch was simply shit.

 

[TommyCat]

Glad you got it worked out and not too much treasure lost! (Overlooking the lost time... )
Be fun to crack one of the bad ones open to see if it has any guts... :-/

Best of luck in the future,
T.C.

 

[Zambam]

Open a dispute, send your video and you should get a refund. I bought this magnetic brake cutout Magnetic Brake Sensor For Ebike Cut Off Power-Brake System 2 3 Pin Plug | eBay which did not cutout with a magnet. I had to send the seller a video, after which they promptly issued a full refund. I cut open the sensor, was expecting a reed switch or hall sensor inside but there was nothing.