2.4Ghz Antenna selection(help needed)

[whollender]

I am not sure what senso is going to do after all. Perhaps it make sense to make small 4 layer breakout board and solder it on top of 2 layer if going 4 layer for entire design is not justified.

When people say "popper equipment" I wonder what do they mean specifically in cases like this? Thing is so tiny. Suppose I have 6GHz VNA and want to "tune it". How do I even approach measuring/tuning it? I sometimes solder pieces of RG405 with SMA connectors at other end right to the track, but here even that won't fit well. 

Whoever has the "proper equipment" how exactly do you guys do it in cases like this? Please share. Even few photos would mean a lot.

Check out the antenna tuning link in the antenna datasheet: http://www.johansontechnology.com/tuning.html

It steps you through the process of tuning in the antenna, but assumes that you know how to do a lumped element smith chart match.

Basically the steps are:

• Cut RF trace and solder in a coax pigtail
• Cal VNA so that cal plane is at connection to pigtail
• Measure return loss with no antenna or matching components (end of transmission line is 'open')
• Measure return loss with antenna and 0 ohm jumper for series matching element
• Perform matching calculation using results from #4, normalized with data from #3

The data from #3 is basically used to shift the calibration plane of the VNA down to the antenna itself, so that you're looking directly at the return loss of just the antenna (with PCB layout) instead of the pigtail, PCB transmission line, and antenna.  With that data you can perform a matching calculation that will give you the component values you need for the matching network.

[hendorog]

I am not sure what senso is going to do after all. Perhaps it make sense to make small 4 layer breakout board and solder it on top of 2 layer if going 4 layer for entire design is not justified.

When people say "popper equipment" I wonder what do they mean specifically in cases like this? Thing is so tiny. Suppose I have 6GHz VNA and want to "tune it". How do I even approach measuring/tuning it? I sometimes solder pieces of RG405 with SMA connectors at other end right to the track, but here even that won't fit well. 

Whoever has the "proper equipment" how exactly do you guys do it in cases like this? Please share. Even few photos would mean a lot.

Other guys who do this for a living have commented about using products like Sonnet to model the circuit using the s-param files published by the component manufacturers. The free lite version of Sonnet doesn't allow this which is a shame as it would be perfect.
(You can download the s-param model for the antenna from the Johanson website.)

There are quite a few free options which may be able to do it, like Qucs, openEMS, Z0lver,.... I can't comment on them as I have not used them.

One other point that hasn't been mentioned yet - the differential traces into the balun should both be the same length. Probably not a big enough mismatch to matter though.

I don't see any reason why a 4 layer board is _required_ for this.

FWIW, simulating the current board with a free simulator is the approach you could take without any test gear. Then build it and test it and tweak it and repeat...

[hendorog]

Turns out I was wrong - you can model this in Sonnet Lite - using a netlist.

If you let me know the board thickness and Er if you know it, and the width of the trace between the balun and antenna I'd be happy to have a crack at doing it.

[ataradov]

Connectors I was talking about are Murata MM8430-2610. They are switch connectors and can be reversed on the board.

[Alexei.Polkhanov]

Connectors I was talking about are Murata MM8430-2610. They are switch connectors and can be reversed on the board.

O, lovely! I like the "push to disconnect feature!
I found a "training module" on DigiKey. I wish I learned about these earlier. Those test probes are $50 to $110 piece. Pricey ..., but I only need 1 or 2.
(Here if anyone interested: http://dkc1.digikey.com/CA/en/TOD/Murata/RFSwitchConnectors/RFSwitchConnectors.html)

Speaking of CONTROLLED IMPEDANCE stuff. How does it work? You send some kind of special PCB art Gerber files to manufacturer and measure it when it is back and then hope everything stays same as soon as you keep ordering same PCBs? Anyone has any example Gerber files for that?

[ataradov]

Those test probes are $50 to $110 piece. Pricey ...

That's RF, everything is expensive.

Speaking of CONTROLLED IMPEDANCE stuff. How does it work?

You send them Gerbers and specify which traces need to have controlled impedance. They will adjust trace width according to their knowledge of their process and materials. They will also generate test tracks of the same width outside of the board. After boards are done, they measure test tracks and generate a report. This assumes that normal tracks are now OK as well. If they missed a lot - they typically re-do the boards.

[senso]

Sorry for the long long delay, other projects got in the middle, but I pulled back the cooper around 0.6mm around the trace after the balun and gained around 20dB of headroom.
This is a stupid question I know, but how would I tune this using a VNA?
Remove the antenna, hook it up to a VNA and then see the signal output and try to throw some inductor or cap from the antenna trace to ground?

Best regards.

[ataradov]

There is an appnote for this - http://www.atmel.com/Images/Atmel-42732-Antenna-Tune-Up_ApplicationNote_AT16802.pdf

You need to keep the antenna in place. You will need to break a trace between the balun and the antenna. That's where those super expensive switch connectors come in handy.

And what you will be measuring and tuning is S-parameters - https://en.wikipedia.org/wiki/Scattering_parameters

BTW, Smith charts are mind-bending technology