Magnetoresistive vs. Hall effect sensors -- is it possible to tell the difference?
Continuing the topic of camshafts with magnetic encoders, Matthew S. mentioned Magnetoresistive (MR) sensors: diag.net/msg/m6g5jcgbol…
I believe this is the first time the term is used on this site, but from discussions elsewhere I got a feeling there is quite a bit of confusion surrounding this term. In particular, questions such as what attributes of MR sensors are specific to them, how to identify one by looking at it or by connecting a multimeter/oscilloscope, are often answered incorrectly.
The purpose of this post is not to provide exhaustive answers to these questions, but rather to point out how hard these questions are by addressing a few misconceptions that have been seen floating around.
To get somewhere, we need to start with proper definitions. In the context of this discussion, a sensor is a device that uses physical effect(s) to detect a state of the environment, and sends this information out via an electrical interface. In the simplest case, the physical effect and electrical interface are one and the same. For example, in inductive variable reluctance sensors the change in magnetic field induces voltage in a coil of wire, which can then be measured:
Magnetoresistance is the ability of material to change its electrical resistance when an external magnetic field is applied. A special circuit is required to detect such change and convert it into usable signal (the part marked SENSOR in the diagram below is the actual magnetoresistive element):
Thus, the electrical interface has nothing to do with the physical effect (magnetoresistance). For example, KMI25 sensors mentioned above use 7-28mA current interface. Armed with this information, we can address the following statements:
"Operation of an MR sensor can be checked with an ohmmeter while rotating the wheel/shaft".
FALSE. An MR sensor will not even operate without power supply.
"The ABS module monitors resistance between the MR sensor leads".
FALSE. For two-wire active digital sensors the ABS module monitors the current that is pulsed by the sensor.
Now, let's throw Hall effect sensors into the mix. The Hall effect is the production of voltage difference across an electrical conductor when magnetic field is applied perpendicular to the current in the conductor. Just like with MR sensors, a special circuit is required to amplify the Hall voltage and convert it into usable signal. And, again, the electrical interface can be whatever is suitable for the application. For example, for wheel speed sensors, the two-wire current loop interface is popular, but for crankshaft and camshaft sensors three wires (power, signal, ground) are typically used.
"MR sensors use two wires, while Hall effect ones are three-wire sensors".
FALSE. This Hall effect sensor from Infineon, TLE4941plusC, uses two-wire current interface:
"MR sensors are used with magnetic encoder wheels, while Hall effect sensors are used with gear wheels."
FALSE. There are two versions of MR KMI25 sensors from NXP, one for magnetic wheels, and the other for geared wheels:
Likewise, there are Hall effect sensors that can be used with both magnetic and gear wheels. For example, TLE4983C from Infineon is specifically designed for camshaft applications:
Well, it looks like any simple rule of thumb to distinguish between MR and Hall effect sensors has plenty of counterexamples. Unless you have disassembled the sensor and checked chip markings or found a trustworthy description in the service information, or it is a very special application that calls for a specific choice of sensing technology, how can you be sure?
Which begs a question: is it even necessary to distinguish between the two? Engineers used the one they thought fits the application best. From a diagnostic standpoint, it is important to know whether the wheel/shaft speed sensor is:
- digital (data is transmitted in digital format rather than analog);
- active (requires power to operate);
- using current loop (two-wire) or voltage signalling (three-wire) electrical interface;
- using a magnetic encoder wheel or a gear wheel.
The examples above show that the name of the physical effect used does not quite help with questions 3 and 4. So, is it not easier just to call such sensors "digital active", identify their electrical interface and encoder type, and form your diagnostic approach based on that information, or there are some benefits to knowing the underlying physical effect as well?
What do you think?
This is awesome, Dmitriy. I recently bought the Bosch 10th Edition Automotive Handbook and while flipping through it I came upon the section with MR sensors and Hall Effect sensors. I started confusing myself trying to truly understand the difference between them. Your explanation using a bit less "engineer speak" makes it very clear. When scoping a MR sensor, is it "proper" to measure across…
Thank you, Jj! If Matthew is 100% sure the sensors in his post are of MR type then yes, there are more curveballs out there! He writes: "There are two physical cam sensors, but 4 signals total. The sensors sit between the two cams and have 4 wires; power, ground, two cam signals." As for scoping the two-wire sensors, are you using Pico4425? -- it has isolated grounds, so I guess it is OK to…
Hi Dmitriy. Personally, I don't need to know much more than how the circuit is configured and whether an expected output is analog or digital. Both Hall Effect and Magneto Resistive sensors have analog signals within the sensor, but are digital output square waves and that is all that we really need. You are correct that there is confusion in the number of wires used, so a proper description of…
Thank you, Martin, these are all very relevant examples!
Martin I really like your term “mythinformation”. And there sure is a lot of it out there.
Hi Marty: I no longer have access to the captures for show & tell but the WSS on the 2004 Malibu Maxx was an interesting critter. As you know, GM's testing procedure is as you describe. But if you put a scope across both terminals, you would get an A/C-type signal. Nothing special there. The reason I grabbed the captures is because you could tell which way the wheel was turning. Turn the…
Thanks for the info! I always wondered about 2 wire Hall's or 3 wire MR's. The categories you broke down would be much easier to use for identification. Bernie Thompson has a great article on MR operation here. searchautoparts.com/automechanika…
Wow although a lot of the terminology is over my head that article is great!! Lots of great diagnostic information.
Thanks, Sean! Bernie's article is very comprehensive, a good read for anyone not intimidated by the terminology.
Thank you for this information! I am on a quest to learn more about basic electronics. Reading Bernie's article and yours I am seeing the term resistive element, that led my brain to the word resistivity, that I learned last as the way we can calculate the resistance of a conductor based on its length and cross sectional area. My question is are magnetoresitive systems more prone to errors…
Brian, are you talking about extra resistance in the connectors and wiring to the module? If so, it will depend more on details of the circuit design, rather than on whether the sensor is magnetoresitive or Hall effect one. To make things more interesting, it is not just about signal making it to the module, but also immunity to electromagnetic interference. It would be a good topic for an…
Yes, that is what I was talking about. When you say circuit design is that taking into account the sensor itself along with the ecm internals that monitor it, OR would the code & software be able to compensate? I am working with my arduino and automotive hall effect sensors , I will see if I can come up with some magnetoresistave sensors and experiment as well. Thank you again for sharing…
He's not too shabby for a guy who just watched us all on YouTube and then came here to educate us all eh? Awesome as always Dmitriy
Brian, circuitry in both the sensor and the module. And, thanks for mentioning that, on top of this there is how the software interprets the signal. Here is an example of a perfect storm for a digital active wheel speed sensor, "reporting" 158mph: youtu.be/QWPo5F_zt6w Interestingly, Keith's tug test makes an appearance at 2:40 in the video... Oh, hi Keith, thanks for the shout-out!