Coil design verifying

Hello Bjornar,

For the best resolution, the rule of a thum is for the coil diameter to be the size of the target. The useful range would be 1/2 of the diameter of the coil you are using. For example, if your coil is 40mm, the endpoints would be the closest distance as the starting point and around 20 mm as the end point.

You can see from the GUI response what set up provides you the best resolution. I would advise to log the data in a text file and do plots for all the possible set ups to determine which works better. From the logged file, you can see the 'raw data' of L and Rp and plot it versus the distance.

I assume you are using only L measurement, but could you please confirm?

Please, feel free if you have any questions.

Best Regards,

Natallia Holubeva

Application Engineer

Hi Natalia

We will go with a solution which is in accordance with the rule of thumb as you suggest, but we have to verify the correct sizes in testing.

I am actually a bit uncertain of how I use the GUI response to check resolution and to see which works the best. Will I have to check the total raw data change (startpoint value minus endpoint value) for each setup, and if this change is bigger in some cases then that means a better resolution?

How do I determine wether to use the Inductance or the Rp? The target is steel, does that matter?

Regards, Bjørnar

Hello Bjornar,

The use of either L or Rp is usually determined by the application. For some applications, you might use both (for example, if you are trying to determine the composition of the metal). In your case, you are trying the sense the distance only, correct? I would suggest using L measurement since it is sufficient in this case. Additionally, L measurement is more stable with temperature drift. What is the frequency of your sensor?

In the GUI response, your assumpions are correct. Please, proceed as you stated above.

Best Regards,

Natallia Holubeva

Application Engineer

Will L give the same resolution as Rp? It seems as the response in the GUI is much lower for the inductance.

Why is the L measurement more stable? Due to resistivity changes in the target material when temp. changes?

The frequency is around the maximum for the chip, 5MHz.

Best Regards,

Bjørnar

Dear Bjornar,

L is stable with temperature variation (~30ppm/°C) since it is dominated by shift in sensor capacitor temperature coefficient. Rp is subject to temperature variation (3900ppm/°C for Cu). The variation is dominated by shift in conductivity of target. For Rp measurement, temperature compensation may be required, which introduces more work for you.

Your target is steel, but is there a way you could use Aluminum instead? Certain kinds of steel have ferromagnetic properties that may force you to use Rp instead. What kind of steel is it?

Also, in the GUI streaming tab, please right click on the inductance window and choose 'Toggle Data Type'. You will see the output codes instead of the inductance which you can work with as well.

Best Regards,

Natallia Holubeva

Application Engineer

Hi again Natallia

I understand, Temperature changes will be a factor so I think we will go for an L measurement.

We have chosen steel because that gave the best range and resolution in the coil designer. We can absolutely use aluminium, but random testing of aluminium targets shows a much lower response compared to steel but it may be somewhat due to the metals quality. We used an aluminium ruler, a cut off piece of alu. plate from the local aluminium supplier which is probably construction quality.

Can you verify wether alu. will typically give a lower response compared to steel? The reason we want to optimise the response is that we need to press down the coil and target size in this application as space, size and weight are important design drivers.

The range we need to measure is 1-15mm and the target size should ideally be in the range of 20-30mm

On the coil sizes we have a bit more room, but bigger coils than 40mm could cause problems so you see why we need to optimise the design.

Thanks again for your input to our work.

In general, because Aluminum has a higher conductivity, it will produce a larger response than steel.

The mechanicals you are proposing look fine.

Regards,

ChrisO

ok, good.

Is there also a way to calculate/estimate the power consumption of the circuit? The chip uses 1,7mA but we would like to know the total consumption of the circuit with periferal components and the setup described above with a 250 turn, 110ohm coil and 22pF capacitor.

Rgs, Bjørnar