where I can find an inductor to be used as RFC at 10Ghz ?

[StoppTidigare]

RFC for 10GHz

Hi all I wonder where I can find an inductor to be used as RFC at 10Ghz ?

Kindest regards,
StoppTidigare

 

[vfone]

Re: RFC for 10GHz

The best microwave inductors (lumped) on the market, has SRF at 10GHz.
**broken link removed**
At this frequency you have to take in consideration using microstrip elements.

regards

 

[activewei]

Re: RFC for 10GHz

since your frequency is 10 GHz, Why don't you try using quarter wavelength trasformer with radial stub? The overall dimension will be small and parasitic effect of lumped element is reduced to minimum. Do you need broadband RFC?

 

[boy]

Re: RFC for 10GHz

I would use a radial stub.

 

[lguancho]

Re: RFC for 10GHz

Using microstrip choke has added advantage that it can filter unwanted harmonics of the desired signal.

 

[StoppTidigare]

Re: RFC for 10GHz

Hi all thanks,for the replies.

vfone, I checked up your link to avxcorp. They don't have any inductor for
10GHz. They have selfresonant frequency at 10 GHz. That means that is no inductance left!

I have to go for either spiral inductor or lambda four transformer with accompanied radial stub or a spiral inductor in microstrip.

How do I deal with a radial stub ? I know it is a serial stub but I don't know so much more.

Could somebody tell ?

Kindest regards,
StoppTidigare

 

[activewei]

RFC for 10GHz

StoppTidigare: You can try out spiral inductor using EM simulation. But it is quite time consuming to do that.

FOr radial, it is simple. What you need is a quater wavelength at 10 GHz (calculate based on the substrate that you are using) followed by a radial stub (the radius is closed to quater wavelength at 10 GHz as well). After the radial stub, extend the line slightly longer for bypass capacitor (recommend tantalum).... Simple right?

 

[lguancho]

Re: RFC for 10GHz

At microwave frequency ( 1 or 2 GHz above), you should really considered distributed design, rather than using RFC. Even if a 10GHz RFC is available, it will be expensive.

 

[StoppTidigare]

Re: RFC for 10GHz

Hi all,
activewei I don't really understand the use of the radial frankly speeking..
I'll make a short circuit with a capacitorwhich also will be the de-coupling capacitor for the voltage supply why do I really need a radial stub between
the capacitor and the lambda/four microstrip ?

The textbook for understanding radial stubs is the Mathei book I guess ?

Regards,
StoppTidigare

 

[activewei]

Re: RFC for 10GHz

Hi all,
activewei I don't really understand the use of the radial frankly speeking..
I'll make a short circuit with a capacitorwhich also will be the de-coupling capacitor for the voltage supply why do I really need a radial stub between
the capacitor and the lambda/four microstrip ?

The textbook for understanding radial stubs is the Mathei book I guess ?

Regards,
StoppTidigare

THe paper has been uploaded previously. But I attached here again just for your reference. Refer to figure 6 at pg 2, you will get the idea what I mean. It is really hard to explain without figure....

 

[activewei]

RFC for 10GHz

opps, i would like to emphasize something:

1) the paper recommend to use lumped element to replace radial stub as the dimension of a radial stub is quite large at S/C-band, which is true.

2) However, as your design frequency is higher (at 10 GHz in your case), it is better to use radial stub. It is hard to include all the parasitic effect of lumped component at 10 GHz in simulation (if you still prefer lumped element, use ATC capacitor as the manufacturer provide all S-parameters of the capacitor).

- I designed 4W power amaplfier at S-band using lumped element in my bias network (as explained in the paper) and 12W x-band Power amplifier using radial stub. Prior to active device assembly, i evaluated the PCB board first. The wonderful thing about the radial stub at high frequency (x-band) is the through 50 ohm line (13mm) with bias network incorporated is relatively low. The measured insertion loss of the 13mm line with bias network is just -0.35dB at 8.3GHz (including the connectors losses), amazing right?