Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
Not applicable


We are using the TLE5012B on one of our products and I am trying to test it functionally using some custom test equipment. The test equipment consists of a jig, for interfacing with the product, and a microcontroller which I use to communicate with the TLE5012B, set up the angle measurement interface, the resolution we want etc and of course to read angle measurements. All of this is working so I don't need any help with that.

My general test strategy is to send the TLE5012B test vectors and then read the angle measurement over the IIF and also by reading the angle value register. The test vectors are working just fine, and I can simulate multiple rotations and count the IFC pulses etc.


In addition to using the test vectors, I also have a permanent magnet installed in the test fixture which I use to take a static measurement of a real magnetic field. The magnet does not rotate. When I installed the magnet, it was placed in at a random angle - my idea was to simply measure the angle with the product and use the reading it gives as a benchmark for the rest of our product going forwards. The angle turns out to be around 268 degrees.

This works, but I find that the measurement error between products can be larger than I was expecting. Below is the result of a couple of days output, which is around 8000 products:


The permanent magnet angle is not taken via the IIF, only by reading the internal angle register.

I have looked through the datasheet and found a few sources of angle error. Since the test fixture is not using a rotating magnet, the auto calibration feature cannot be used. But, even taking all the sources of angle error into account, I can't account for the measurement spread I see here.

Does Infineon have advice about how I might resolve or justify the errors I am seeing here?

Thank you in advance for your time.



Even though I stated that the test vectors are working just fine, I guess I don't know if they suffer the same angle error as is observed with the permanent angle measurement. This is because the permanent magnet is measured first, before I use the test vectors, and if this fails test then it does not proceed to take the test vector measurements.
My current test limits are quite wide, but not wide enough to account for the measurement spread I posted above. So the fails from that measurement spread will only have permanent angle measurement results.

If any one replying to my message is interested in whether the test vectors are also affected, I can go and find some of the poorest performing products and force it to run the entire test routine to collect this information.
0 Replies