2
\$\begingroup\$

I did a layout of a NFC PCB inductor/"antenna"(I know it is not one exactly) drawn only on top side copper (nothing drawn on other layers at inductor area), and I wonder if there will be a considerable loss of working distance range on the opposite side due the PCB FR4 core itself is an obstacle... I need to read tags to both sides of the board at approximated distances.

I will share some screenshots.

Schematic of RF side, based on ST ST25R95, 230mW output power, QFN32:

enter image description here

Rant is 1.65 ohms and was calculated HERE

Lant is 1.09uH and came from ST's online tool eDesign Suite

enter image description here

This is my current drawing for the PCB coil/inductor. 10.5 x 26mm, 7 turns.

enter image description hereenter image description here

I wonder if I need (if it is suitable) to continue the "antenna" drawing on the bottom layer, and sum the inductances and resistances? (Or some other equation to calculate the inductance).

Also I wonder if I do that (antenna drawn on both sides) if I could reach more working distance (between the reader circuit and the tags) than doing coil only on top side.

If I need to continue the coil on bottom side also, do you have any recommendation about how to draw a good PCB inductor? So that I can reach the best distance range the chip can offer (it is 0.23W output power only, ST's ST25R95).

Regards.

EDIT 1

4 turns x 2 layer NFC coil.

TURN x mm:

1: 11.2 + 25.7 + 10.2 + 24.9 +

2: 9.4 + 24.1 + 8.6 + 23.3 +

3: 7.8 + 22.5 + 7 + 21.7 +

4: 6.2 + 20.9 + 5.4 + 20.1 + (2.7)

side = 251.7mm

total (x2) = 503mm

PARAMETERS of inductor traces:

  • witdh = 0.3mm

  • length = 503mm

  • copper tickness = 18um (1/2 oz final copper thickness)

According to HERE, Rant = 1.66 ohms

So Then I have new input values, Lant = 1.68uH and Rant = 1.66 ohms to input on the ST25R95 calculator spreadsheet. Now I can easily recalculate the NFC circuit, I mean the caps and indutors of matching and filter circuits.

I wonder if I should consider the capacitance created between the traces of spposite sides, they are on exact same place, they just are in opposite board sides. PCB is 1.6mm FR4. dk (dielectric constant) can consider as 4.5.

Single side inductance enter image description here

new inductor drawing.
enter image description here

Rant value enter image description here

Coil "pins" enter image description here

\$\endgroup\$
1
\$\begingroup\$

Also I wonder if I do that (antenna drawn on both sides) if I could reach more working distance (between the reader circuit and the tags) than doing coil only on top side.

and

If I need to continue the coil on bottom side also, do you have any recommendation about how to draw a good PCB inductor?

For near-field magnetic coupling it would be better if the coil was split between top and bottom copper layers with half the turns on the top layer and half on the bottom stacked above each other. That will allow you to use slightly fewer turns overall and get a slightly better coil Q factor. That will increase range slightly and, because coils are top and bottom you would get equal range performance from either side.

Designing coils to couple magnetism some distance requires the best amount of mutual coupling you can get between the windings of the transmit coil. Designs that are spiral are not the best but, given that you might have only one PCB layer to work with then it has to be a spiral but take note; the lower radius windings of the spiral are largely superfluous to the energy transmitted to the receiver and so, if you had (say) four layers, the optimum would be to put stacked turns on each of the four layers (just like a regular electromagnet).

So, I would look for a calculator that allows you to design the right inductance using 2 or more layers of the PCB.

| improve this answer | |
\$\endgroup\$
  • 1
    \$\begingroup\$ I think there will be tools to do this but I can't recommend one off the top of my head. You could assume that a single layer 4 turn inductor of inductance "L" will become roughly "4L" when stacked above another layer of 4 turns. But, it will probably be more like 3.5L due to imperfect coupling. \$\endgroup\$ – Andy aka May 26 at 12:36
  • 1
    \$\begingroup\$ Somewhere between the two and it's really a guess. Inductor values are really hard to accurately calculate from physical layout. \$\endgroup\$ – Andy aka May 26 at 13:19
  • 1
    \$\begingroup\$ I'm actually excited about this and look forward to what you reveal. \$\endgroup\$ – Andy aka May 26 at 13:36
  • 1
    \$\begingroup\$ Yes, that looks good. I expect the capacitance will be more and a simple area and distance calculation would probably be enough to give you a number. \$\endgroup\$ – Andy aka May 26 at 15:59
  • 1
    \$\begingroup\$ I would also add a capacitor underneath C23 - you don't need to use it but, from experience you might require it. In the past (hand wound coils) I've made provision for three capacitors just so I can tweak things. Also, in your bottom diagram, what is all the flood of green about? \$\endgroup\$ – Andy aka May 26 at 16:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.