Chopper stabilization is an electronic offset correction technique used on many Hall Effect sensors. The Hall Effect bridge element is sensitive to both magnetic fields, and to mechanical stresses put on the element. Mechanical stresses can come from molding the element into its IC package, bending the leads, heating the chip, potting, pressure put on the sensor face, etc. Chopper stabilization is a switching technique that sends current in one direction of the hall bridge and measuring the output, then sending current in the other direction and measuring the output. When these 2 outputs are averaged, the result corrects for the error that comes from the mechanical stresses. This switching occurs at a high speed, called the Chopper Switching Frequency. The outputs need to be turned off during this switching time. They are then updated at the end of the cycle.

What a customer will notice is that with an instantaneous change in the magnetic field, the output may take up to 1 / Chopper Frequency to update. For a typical 62k Hz Chopper Frequency, the output will take up to 16uS to change. At low speeds, this 16uS 'noise' is a tiny percentage. At higher speeds, this noise becomes a larger percentage. This chart shows the percentage of chopper noise vs. the pulse frequency:

Pulse Freq.
Hz
12 tooth
target
60 tooth
target
16uS noise
percent
10 50 RPM 10 RPM 0.0%
100 500 RPM 100 RPM 0.2%
1000 5000 RPM 1000 RPM 1.6%
5000 25000 RPM 5000 RPM 8.0%
10000 50000 RPM 10000 RPM 16%

 

It is important to note that this is noise only, not the error. Error comes from how many pulses are measured to get the speed. If only one pulse is measured to determine speed, the error = the noise. If 100 pulses are averaged to determine speed, the error is the noise / 100. To obtain error, take the noise and divide by the number of pulses used to determine speed. For example a 12-tooth target spinning at 5000 RPM is measured averaging 50 pulses. The noise is 1.6%. When you divide by 50 pulses, the error is .032%.

What you will see is the edges of the pulses appear to jitter at high speeds. Here is a representative scope plot:

It is important to note that this is noise only, not the error. Error comes from how many pulses are measured to get the speed. If only one pulse is measured to determine speed, the error = the noise. If 100 pulses are averaged to determine speed, the error is the noise / 100. To obtain error, take the noise and divide by the number of pulses used to determine speed. For example a 12-tooth target spinning at 5000 RPM is measured averaging 50 pulses. The noise is 1.6%. When you divide by 50 pulses, the error is .032%. What you will see is the edges of the pulses appear to jitter at high speeds. Here is a representative scope plot:

Click Here to return to the Application Notes Index