The distance IC/magnet heavily depends on the choice of the magnet. But for a general answer you can expect a distance between 5 and 25 mm (and more if you would design a magnet for it). Here is an application note that could help www.melexis.com/en/documents/documentation/application-notes/application-note-how-to-use-a-magnetometer-in-a-position-sensor-application

Hi, it won't be possible to simply replace the MLX90363 with the MLX90393. The MLX90393 is a completely different IC. The MLX90393 can only provide X/Y/Z data and uses a reduced and simpler command set, whereas the MLX90363 can also directly provide the angle for rotary, linear position, or joystick applications.

Dear, thank you for your interest in the EVB90395. We advise you to get in touch with our sales team so they can help you further. www.melexis.com/en/contact/sales-contact

You will indeed need some simulations to define the right magnet (size,..) See the application notes on the product page www.melexis.com/MLX90394

In general, you can have a look at this application note here www.melexis.com/en/documents/documentation/application-notes/application-note-how-to-use-a-magnetometer-in-a-position-sensor-application or see for more specific application note on a product page like the www.melexis.com/MLX90394.

Fun fact: when the audio cassette industry faded, some companies producing the magnetic tape needed to reinvent themselves. Therefore, they used a very similar technology to become the main providers of magnetic scales for encoders. The working principle is quite close actually even though the encoding of the position is different and in this case the 'reading head' is a more advanced triaxis hall sensor.

- Good guess. The encoder is absolute. - Due to the bigger size of the Master track which is providing the accurate angle, the effect of the crosstalk coming from the Nonius track is reduced with consequent improved robustness against radial wear-out (which normally causes crosstalk variation). - Magnet can be customized according to specific design rules. Distribution strategy is still being defined. - Noise performances are specified for 1ms refreshing rate. For intermediate product, readout period is 80us. For the final product it will be possible to readout angle and speed at periods as small as 15-20 us.

@@BrunoBrajon Thanks!! Is there a limit on the angular speed of the rotor? A dentist's drill (for example) runs at ~100k RPM i.e. ~1e4 rad/s. I can imagine that the sensors might have trouble reading the angle (albeit with aliasing) as the rotor-magnet would travel almost 50 deg in an 80us window.

@@AdityaMehendale The sensor includes look-ahead angle correction for very high-speed applications. The maximum speed of rotation the sensor can work at depends on the number of pole pairs on the Master track. In general Max rotation speed is equal to 200krpm/Number of polepairs on the Master track. With the magnet shown in the video, this is equal to 25 krpm

Great informations, @arduinomaquinas thank you man, subscribed 😊🇧🇷👏👏👏👏👏👏

I have the same question!

You need range compensation algorithm

Cannot be helped. This kind of sensor is meant for relative temperature changes

PCB less solution makes big advantage in cost first of all. Other big benefits are vibration proof, easiness of assembly process, much more compact size. This IC is hybrid like mentioned in the video - for example it contains additional decoupling capacitors, so no need to add them externally. In case PCB is present, you need to protect it by polymer compounds from moisture and to improve mechanical properties to match automotive requirements, here this is not needed. PCB less solution is very attractive for sensors like magnetic or pressure, especially in automotive hursh environment.

@@vsp_tof Thx!

This saves in the cost of a full sensor. One less item to buy since the caps are built in.

50 percent of normalized graph. We are not interested in absolute values returned from the sensor in this experiment that is why normalisation is done. Ideally the temperature of the object should be high enough to have clear graph. Normalized graph is not much sensitive from the temperature value itself. Definitely, object should be hot enough to be easily distinguished from the background and same time not to saturate the sensor. And yes, as any FIR sensor, it has its internal temperature sensor for compensation. You are right, ideally sensor temperature should be as background, but for most cases with proper object temperature selection, this requirement is not so strict.

Let me nuance that - a "uniform magnetic field" can be rejected by a gradiometer, however the "real life" stray fields are likely caused by small magnets - e.g. in the speaker of a smartphone - can this be rejected too?

The MLX90395 measure the magnetic field in three directions (X, Y, and Z), using horizontal Hall elements combined with and integrated magnetic concentrator, and outputs these measurements as a digital value relative to the magnetic field strength. . The actual position (angular or linear displacement) can be calculated using the data that is provided by the sensor. More information on our IMC (integrated Magnetic Concentrator) can be found here: www.melexis.com/en/tech-talks/triaxis-position-sensing-solution

@@Melexis_Sensors This doesn't really answer my question of if the chip is capable of simultaneous 6dof or not, and none of your materials explain if this is possible either. It's really as simple as a yes or no answer, yes it can do simultaneous 6dof, or no the chip is limited to a certain degree of freedom. I just want to know if I can have both a spatial position of a magnet and 3D rotation (better described as a quaternion) of said magnet, simultaneously. Or does your product lack the amount of sensors that would make this possible. I.e. is this similar to the Fraunhofer IIS product, which does true spatial measurements with a higher number of sensors (IIRC 12 sensors), or is it more similar to the products that Texas Instruments and Infineon produce where it's only one sensor per axis and thus has a more limited degree of freedom?

@@xaytana Thanks for reverting back, please allow us to give a bit more background to the previous answer. As already indicated in your previous reply you were using a reference of the HallinOne approach from Fraunhofer. This HiO cell is nothing more than a combination of 2 vertical hall sensors and 1 horizontal hall to measure the 3 different magnetic field components. In our Melexis solution we are not using vertical hall technology but we are using our IMC technology in combination with planar hall technology our concentrator allows us to convert horizontal field lines into a vertical component which we then use to measure the 3 different field components. So both principles (HallinOne or our Triaxis technology) will give you the raw data signals for the 3 different field components. The Integrated Magnetic Concentrator is offering a certain gain and is resulting in better performance (offset and sensitivity) compared with vertical hall technology. So to answer your question our chip will provide the data for Bx,By and Bz allowing you to use the 3 vectors (simultaneously) for further post processing.

Thank you Diana!

Thank you Nahla!

Thank you Krasimira!

Thanks Bjorn!

Thank you for your nice comment Karel!

Thank you Artem, the video is inspired by you and all the other colleagues!

Hi Doraditya, We can confirm that these accuracies can be achieved after a proper calibration. Depending on the application MLX90363 might be a good choice but MLX90392 is possible too. More info on www.melexis.com/en/documents/documentation/application-notes/application-note-how-to-use-a-magnetometer-in-a-position-sensor-application

Hi Bikash, no, it isn't.

@Tom Vanagt Hello Nick, Actually, I was wondering if it could be used for the same purpose, to detect vehicles. please correct me if i misunderstood The larger the Tesla value, the more the sensing distance of the sensor is, isn't it? So, MLX90395 has 120mT but MLX90392 has 5mT, so MLX90395's sensing range is not higher?

What’s a sleeping giant