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.

RF Mosfet Class E PA for Dummies or absolute idiots

Status
Not open for further replies.

berger.h

Member level 5
Joined
Sep 30, 2016
Messages
93
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
2,355
I have about 100W RF PA with MOSFET about class E, and I would like to understand how such a thing is proposed.
I looked into books like this or this
frankly, more I understand from this

Thing simpler and for beginnings i will ignore everything before G MOSFET see in picture.
Frequency of PA is 13,56MHz and all power (sin) go to 50Ohm load.
Screen+Shot.png

RF Choke L1?
If I understand correctly
L1 must form a resonant circuit with the capacitor in MOSFET Coos.
Her I have first problem.
IXFH12N50F have Coss 290pF , but this is for Vds 25V And I have more. How do I find a Coss for 50V?
In any case, L1 is 10 turn on any white coated ferrite OD 24mm and I measure 10uH on 100kHz
And this have resonant frequency about 3MHz?
What can't I see where the problem is?

Impednce Matching networks.
This convert low output impedance of MOSFET Zin to other Zou I thing 50Ohm.
I understand that, but I don't know how to calculate impedance of MOSFET amplifier?
Or if possible how I measure this impedance without expensive instruments like network analyzer?

That's enough for a start.
 

But that amplifier is not Class-E.. It seems Class-A
In order to understand Class-E Power Amplifiers, you have to read Grebennikov's "High Efficiency Power Amplifiers" textbook.There are more but it's good for beginner.
 

How do you judge class A?
Whistleblower that anything before G is just hinted and in reality I would expect the first stage of the amplifier with another transistor.
in any case, just a fool would make a RF amplifier of 100W 13.56MHz (most often induction heating) in class A, if it has a 100W output power, so at a typical class A performance of 35% would have to 185W to heat.
These amplifiers are class C-up.

Also, I do not understand much, but here is a typical case of amplifier in class E.
in my opinion, it is the same as in the black and white picture above. The first look is missing the capacitor Cv, but it is the internal parasitic capacity of the Coos MOSFET transistor.
Plus it's extra Impedance matching network for impedance adaptation of LPF impedance and mainly load impedance.
Without this element, some of the power would change to standing wave.

classE.png
 

Ok that it is an amplifier in class E is just my, perhaps logical reasoning.
The whole amplifier looks something like this
What class does the amplifier do?

Screen+Shot1.png
 

Dear carpenter
Cv is not Internal Parasitic Capacitance of MOS, it's essential of Class-E Amplifiers.
 

I'm not saying it's necessarily, but it can be.
Cv is the sum of all capacities between D S of switching MOSFET (capacity in series with L1).
L1 + CV is designed as resonant LC circuit with resonst frequency = frequency of PA.
And this circuits can be designed with or without external Cv.
In this specific case. IXFH12N50F have Coos 290pF on Vds=25V and this increase with Voltage Vds. Maybe on 50V 330pF (estimate maybe more)
for f=13,56MHz and C 330pF is resonant with L=420nH.
After all, in the referenced document is in Table 1 calculated value Cv and L1 for Class E PA with IRF510 and for band 15M is external Cv missing (C is only IRF510 internal Coos 120pF) and L1 is 500nH 10T T50-6 core.
 

Dear carpenter,

D-S Capacitance is not a fixed capacitor, instead it's voltage depended variable capacitance and therefore using this parasitic element will cause some troubles.
The Switch Mode Power Amplifier Guru Mr. Grebennikov and Nathan Sokal ( founder of Class-E) have both mentioned this phenomena and therefore there is a "controllable" fixed capacitors in their textbooks and papers.
 

In this specific case. IXFH12N50F have Coos 290pF on Vds=25V and this increase with Voltage Vds.
MOSFET Coss always decreases with rising Vds.

- - - Updated - - -

See this IXYS application note for 13.56 MHz class E PA. https://ixys.com/Documents/AppNotes/CO1.pdf
It's using series resonant tank.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top