PA amplifier mismatch load

[klick]

Yes, you are right ,without load is extreme condictions

normally it works whit antenna...

but, if I cover the antenna whit hand (is possible because it is a portable trasmitter), it is difficult to hope the antenna is perfectly matched...

 

[courcirc8]

If you put your hand on a antenna, the impedance can be almost anywhere on the smith chart.
I will not debate if you need better IDD control versus load or not, but in case you really need, I made a small pdf for you.
I did not simulated it but it should at least improve the situation. Tell me if it helps or if you need more precisions.

 

[klick]

but, I don't understand if the problem is the DC bias or IN-OUT matching.....

 

[courcirc8]

I proposed you a new biasing scheme able to control the average current going inside the RF transistor whatever the output impedance is.

You can try modifying your matching, but there will always be a certain load impedance increasing the PA consumption. (ok you can add serial parallel resistor to limit the SWR but this is not an efficient solution).
If this is acceptable for your application, keep it like that!

You can play with the matching network to solve the problem for a certain load condition, if you need to guarantee a maximum consumption for the whole smith chart, then you have better to modify the biasing.

 

[ChaoFan]

You design a coupler to measure the mismatch and then to control the PA bias to shut off it or not and save battary.