PCB Antenna Tuning

I am trying to tune an antenna as seen above, using this wonderful Smith Chart Tool. It needs to match 35 + 10j characteristic impedance.

I have found a tool like the Microstrip Impedance Calculator to determine the impedance of the transmission lines.

However, i am unsure how to determine the characteristic impedance of the stubs that don't have a ground plane (green area) behind them. Does anyone know how the stub impedences are calculated?

Do yourself a favour and post the image as an attachment instead of a google link - all I see is an empty gray rectangle.
Once you have attached the image, you can edit your post to use the attachment URL to add it inline again.

Ok, I found a problem with Google Images (sorry, was testing that too). The links worked on my browser, and previews, but didnt work in incognito mode.

I 'think' I figured out a way to make it work.. let me know if the original image looks ok now.

However, I have attached the file, in case it didn't.

brento:
However, I have attached the file, in case it didn't.

So now we have:

PCBAntenna.png

What stubs? I see an antenna and some microstrip joining up some matching components.

What stubs? I see an antenna and some microstrip joining up some matching components.

Are you saying the antenna itself has zero impedance when simulating with the Smith Chart Tool? Ie, that tool is meant only for calculations up until I am connected to the antenna?

It seems to me, in 3 dimensions, that if you viewed the PCB from the top edge of the photo, you could trace a line through metal, through dielectric, and hit the top edge of the ground plane. However, the plane would appear as a line only, from that perspective. Are you saying this has negligible impact on the antenna’s impedance itself?

Is there a source with data for this pcb antenna design? What substrate is it assuming (FR4 isn't the best
at microwave frequencies).

MarkT:
Is there a source with data for this PCB antenna design? What substrate is it assuming (FR4 isn't the best
at microwave frequencies).

So what do all the NRF and ESP boards use?

They match the antenna to the pcb dielectric constant and thickness, whatever it is, my point is one design
isn't optimal for every pcb material. FR4 is often used for cheapness where a few dB more loss isn't an issue.

MarkT:
where a few dB more loss isn't an issue.

But losing 3dB is halving; can halving ever not be an issue?

The layout looks pretty similar to other NRF24L01 designs, with traces on PCB and no Gnd plane



https://www.newegg.com/p/2S7-01JK-0JR61?item=9SIABKSAP31519&source=region&nm_mc=knc-googlemkp-pc&cm_mmc=knc-googlemkp-pc--pla-new+ocean+tech--gadgets-_-9SIABKSAP31519&gclid=EAIaIQobChMI5IGuoqWa6QIVEo3ICh0pWQsvEAQYECABEgLrn_D_BwE&gclsrc=aw.ds

This antenna layout, also with no Gnd plane under the traces, looks more like yours.

I am able to calculate the microstrip impedance based on 1.6mm PCB thickness, dielectric constant of 4.5, 2oz copper (2.8 mils) and a trace width of 0.5mm.

The antenna's length is 1/4 wavelength (35mm) for Wifi and Bluetooth, similar to the photos shown. I have no idea what the characteristic impedance is of an etching like that.

I am presuming with no ground plane, that these types of antennas have negligible impedance, unless someone can say they have measured otherwise.

I am presuming with no ground plane, that these types of antennas have negligible impedance,

Where did you get this idea? Obviously, for a successful design, the impedance is reasonably close to matching that of the transmitter output, as required.

TI has some PCB antenna design guides. For example