Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Designing class AB power amplifier in ADS

Status
Not open for further replies.

Ata_sa16

Full Member level 6
Joined
Mar 29, 2016
Messages
343
Helped
59
Reputation
118
Reaction score
58
Trophy points
28
Location
Milky Way Galaxy, 179° 56′ 39.4″
Activity points
2,221
Hi,

I am trying to design 2 stage class AB power amplifier in ADS. I want to get 30 dbm output power.
I started with loadpul simulation for second stage but I never get 30 dbm.
My VDD is 2.5 since I have max 2 breakdown voltage and I set V=3 for cascode bias. I am setting the gate bias to be 0.75

Pout =1/4 VDD.Imax assume 31 dB considering output matching loss => Imax = 2 A , IDC for class AB is IDC=Imax/3 -> IDC = 0.66

idc.png

max gain and stability

maxg.png

loadpull

LP.png

loadp.png


:bang::bang::bang:
 

Attachments

  • idc.png
    idc.png
    35.5 KB · Views: 291
  • idc.png
    idc.png
    35.2 KB · Views: 307

This is a Cascode amplifier, not AB-Class.
AB-Class Power Amplifiers do not have cascode configuration.They are configured on either transformer based Push-Pull stage or Balun driven Symmetric stages but not Cascode one.
Also, 30dBm is a really remarkable output power for 2.4GHz and it can theoretically be realized with single ended ( pretty difficult ) but most common the designers use transformer based configurations.
There are some academical works with this type configuration but I'm not sure that is realizable and sustainable.
 
I believe being a class AB is not related to cascode or single transistor. It is about bias point.

With single transistor it is even harder since you cant have high drain voltage for high power.

I have read some publications and they reach 1W with a simple structure. It is too weird !!!

photo_2017-07-17_07-45-56.jpg

As shown by the simulation results: at 2.1GHz under 2.5V, the output power is
1W (30.1dBm) while input power signal is 0dBm, the power
gains more than 20dB of broadband within 1.1GHz to 3GHz,
and the power added efficiency(PAE)is 51.98%
 

And you are right, it is not possible to get 30dbm power in CMOS maybe if you have high breakdown voltage.

I also have another problem tke a look at this. It does not matter what the source impedance is, the result is same. It is too weird and I am so baffled !!

LP4.png


Z_s_fund has almost zero effect on load pull !
 

I believe being a class AB is not related to cascode or single transistor. It is about bias point.
With single transistor it is even harder since you cant have high drain voltage for high power.

AB-Class amplifier has an conduction angle between somewhere 180-360 degree so cascode cannot be used for this conduction angle due to its nature.
OK, you bias the first transistor as AB-class and second one may be anything but since the current flow is purely less than 360 degree, the harmonics and additional harmonics due to the non-linearity make this amplifier impractical.
Otherwise it's surely possible theoretically to bias a transistor under any configuration but it should have a meaning.Single ended amplifier configuration is not practical due to those causes which I have mentioned ( non-linearity and limited conduction angle)
But 2 single stage combined with a transformer or balun will give you a good compromise. Efficiency and good non-linearity.
 
yeah I was thinking about single transistor but this time I have very bad drain voltage which keeps my power very low.

- - - Updated - - -

)
But 2 single stage combined with a transformer or balun will give you a good compromise. Efficiency and good non-linearity.

you mean differential structure ?
 

I suppose you could make a class AB cascode (the devices are in series, so they will inherently have equal conduction angle), but I don't see the point of doing so. Especially with low voltage devices, the cascode will kill your efficiency.
 

My breakdown voltage is equal to 2 volts. so Imagine min voltage for the gate of first transistor is close to zero (20 dB gain and Pout=25 lets say, so input voltage will be around 0.5V and considering loss maybe 0.7)

so when gate is zero the drain voltage is max and it should not exceed 2volts ! it means I should have 1 volt VDD and almost no power !!!
 

yeah I was thinking about single transistor but this time I have very bad drain voltage which keeps my power very low.
you mean differential structure ?
If the voltage is low, you increase the the load current by decreasing the load.Review Load-Line theory of the Power Amplifiers.Also check Load-Pull principal that is more important than all
You can convert this low valued load which is seen by the amplifier to 50 Ohm -if you wish-
Let it call Push-Pull because differential has another meaning.
 

If the voltage is low, you increase the the load current by decreasing the load.Review Load-Line theory of the Power Amplifiers.Also check Load-Pull principal that is more important than all
You can convert this low valued load which is seen by the amplifier to 50 Ohm -if you wish-
Let it call Push-Pull because differential has another meaning.

No that is not possible.

Imagine pout=1W and Pout=1/4 VDD Imax (VDD=1) then Imax will be 4A and Rl = VDD^2/(2*pout) then RL=1/2 = 0.5 which is really low value and impossible to match to 50.

I dont need to review loadline theory since it is all in my mind and of course I know I should convert low valued load to 50 (Only in fundamental frequency and harmonics will be trapped buy matching circuit). Those are the basics of PA design.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top