Doherty Power Amplifier questions

[natnoraa]

Hi,

I would like to ask the following:

1) For designing the carrier and peaking amplifers separately, is it correct if i match the o/p of the carrier to 100ohm and peaking to 50ohm, or both to 100ohm (for load pull simulations), before finally combining them altogether? I've already read many theses and papers but they sometimes don't agree on this sadly.

2) For an easier way of designing, can i first simulate the carrier amplifier, find the o/p compression point, then design my peaking amplifier to make sure the current starts to increase (an indication that the peaking amplifier turns on) at where the o/p compression point of the carrier amplifier is so that load modulation takes place?

Natnoraa

 

[chuckey]

1 If you are operating the PA at 100 % carrrier and its taking a certain current, does this current change when the peaking amp comes on? if it does not then the matching does not change so the PA should be matched to 50 ohms.
2 The Po/Pi curve should be linear, if the peaking amp comes on at the 1 dB compression point then there will be a kink in the out/in characteristics causing distortion on the modulation.
Frank

 

[natnoraa]

Hi Frank!

Thanks for your invaluable insights.

1) The conventional and most simplified way is to have only 1 transistor/1 stage for each carrier and peaking amplifier. However I cascaded mine to obtain a higher power gain because it's hard to get gain at mm-wave frequency. That said, when i simulate the current change for the peaking amp, do i do it for the current taken in by the o/p stage or the ENTIRE peaking amplifier with other stages?

2) Thanks! so intuitively we have to turn it on BEFORE the compression point of the carrier amplifier. got it!

Natnoraa