Rectifier efficiency reduce after I include matching network

[abx]

Hi!

I just learned about designing rectifier. I simulate the rectifier without input matching impedance and I got 38% of efficiency but whenever I put the matching network the efficiency reduced to 27%. S11 shows better result with the matching network. Does anyone know what is going on?

 

[BigBoss]

How did you design this matching circuit ? I mean, what impedance you have referred for rectifier diode ??
Is this impedance small signal impedance or dynamic large signal impedance ??

 

[abx]

How did you design this matching circuit ? I mean, what impedance you have referred for rectifier diode ??
Is this impedance small signal impedance or dynamic large signal impedance ??

I referred to impedance from hbsp simulation for the impedance matching. I believe that’s the right impedance for the matching, isn’t it?

 

[volker@muehlhaus]

Does your matching network use realistic component models with loss?

 

[BigBoss]

I referred to impedance from hbsp simulation for the impedance matching. I believe that’s the right impedance for the matching, isn’t it?

If you defined the source well, it should be OK. This kind of impedance has to be identified with large signals, not small signals.
If you did that, no problem.
Check this impedance with HB using with Current Meter. Simply divide Voltage over Current (both are complex quantities) to verify the result.

 

[D.A.(Tony)Stewart]

Hi!

I just learned about designing rectifier. I simulate the rectifier without input matching impedance and I got 38% of efficiency but whenever I put the matching network the efficiency reduced to 27%. S11 shows better result with the matching network. Does anyone know what is going on?

I wonder if you considered that a rectifier impedance is time variant because it is nonlinear.
This is a conflict with the ideology of matching impedance with a linear invariant matching impedance.

How do you think it is possible to make a rectifier >99% efficient? ( by choosing a voltage Vp=> 99*Vf)

 

[abx]

Does your matching network use realistic component models with loss?

Yes, I include loss for my matching network

 

[vfone]

Before thinking that the matching network is the problem, check the detector itself that it is able to perform well at the working frequency.
Among others, the efficiency of an RF detector also depends by frequency and by the input power.

https://media.springernature.com/full/springer-static/image/art%3A10.1038%2Fs41598-021-99405-7/MediaObjects/41598_2021_99405_Fig5_HTML.png?as=webp

 

[volker@muehlhaus]

Yes, I include loss for my matching network

Then this might be your problem, and the loss in these components is larger than the mismatch loss without any matching network.

You can test that by a simulation with ideal, lossless LC in your matching network.

 

[D.A.(Tony)Stewart]

show your work

--- Updated Feb 5, 2024 ---

Then this might be your problem, and the loss in these components is larger than the mismatch loss without any matching network.

You can test that by a simulation with ideal, lossless LC in your matching network.

What is the matched impedance for a diode?