Seek clarifications on load pulling technique

A

amal banerjee

Guest
Could some electronics guru please
help clarify the following. I am a
bit rusty on load pulling. I am
currently working on some ideas
and testing them with SPICE.

Suppose that a RF transistor( e.g.
BFU520) is configured as a single
stage RF amplifier, with a simple
voltage divider base biasing network,
a collector resistor, and DC blocking
capacitors at the base and collector.
The source resistor is set at 50.0
Ohm, and the load resistor is varied.
The RMS input and output power(in
Watts) is measured for each value of
the variable load resistor. The input
signal frequency is set to 500 MHz.


Now suppose that using this
configration and SPICE transient
analysis, the input power is deliveed
to the load when the variable load
has a value of 150 Ohm.


Now the questions:
1. The output impedance of the
transistor is approximately 150 Ohm
at 500 MHz. Is this correct ?


2, Is it safe to state that at 500
MHz, the input impedance is 50 Ohm,
for the given configuration.


3. The BFU520 datasheet provides
S parameters, and it is possible
to compute the Zin, Zout from
S11, S22, S12 and S21. But it is
not clear if the supplied S parameters
are small signal or large signal.
Spice transient analysis is large
signal, so is the supplied SPICE
device model, so if the supplied
S parameters ought to be latge
signal, but there is no clear
indication.

All hints, suggesyions are welcome.
Thanks in advance.
 
On a sunny day (Sat, 31 Aug 2019 21:16:38 -0700 (PDT)) it happened amal
banerjee <dakupoto@gmail.com> wrote in
<f84089c1-846a-4cfb-ae93-218d2b74d053@googlegroups.com>:

>All hints, suggesyions are welcome.

Buy the transistor, build the circuit, test it.

This has 2 advantages over using spice:
1) you are sure you have reality.
2) you actually have a working circuit you can use.

As a bonus it will likely save you money and time,
as when you build your spice circuit later it may not work.

At 500 MHz things like PCB layout, wire length, parasitic capacitances, etc
already become very important.
Building a circuit will manifest those mysterious effects that spice won't.

RF spice is a bit like Hollywood movies about space travel and aliens.
It looks real, but...
It iS MosTlY DeSIGneD tO seLL MoRE COmpUteRS
 
On Sunday, September 1, 2019 at 11:34:56 AM UTC+5:30, Jan Panteltje wrote:
On a sunny day (Sat, 31 Aug 2019 21:16:38 -0700 (PDT)) it happened amal
banerjee <dakupoto@gmail.com> wrote in
f84089c1-846a-4cfb-ae93-218d2b74d053@googlegroups.com>:

All hints, suggesyions are welcome.

Buy the transistor, build the circuit, test it.

This has 2 advantages over using spice:
1) you are sure you have reality.
2) you actually have a working circuit you can use.

As a bonus it will likely save you money and time,
as when you build your spice circuit later it may not work.

At 500 MHz things like PCB layout, wire length, parasitic capacitances, etc
already become very important.
Building a circuit will manifest those mysterious effects that spice won't.

RF spice is a bit like Hollywood movies about space travel and aliens.
It looks real, but...
It iS MosTlY DeSIGneD tO seLL MoRE COmpUteRS
Click to expand...

I totally agree with what you say. However, high frequency
parasitic (capacitive, inductive) effects can be simulated
in ordinary Spice(HSpice, Ngspice in my case) by, e.g.,
having a very low value non-ideal inductor in-between the
terminals of two components. In fact, the SPICE device
models of most RF transistors include terminal inductances, etc., I have tested these in the past, and the differences
in the results is significant. The problem is that the
values selected for the components of the non-ideal
capacitors etc., are educated guesses, not the real
ones.
 
On Sat, 31 Aug 2019 21:16:38 -0700, amal banerjee wrote:

> All hints, suggesyions are welcome. Thanks in advance.

Post your proposed circuit! Your written description leaves out a lot of
specifics.




--
This message may be freely reproduced without limit or charge only via
the Usenet protocol. Reproduction in whole or part through other
protocols, whether for profit or not, is conditional upon a charge of
GBP10.00 per reproduction. Publication in this manner via non-Usenet
protocols constitutes acceptance of this condition.
 
On Sunday, September 1, 2019 at 4:54:54 PM UTC+5:30, Cursitor Doom wrote:
On Sat, 31 Aug 2019 21:16:38 -0700, amal banerjee wrote:

All hints, suggesyions are welcome. Thanks in advance.

Post your proposed circuit! Your written description leaves out a lot of
specifics.




--
This message may be freely reproduced without limit or charge only via
the Usenet protocol. Reproduction in whole or part through other
protocols, whether for profit or not, is conditional upon a charge of
GBP10.00 per reproduction. Publication in this manner via non-Usenet
protocols constitutes acceptance of this condition.
Click to expand...

I would be happy to supply the text SPICE netlist. Is that
what you are looking for > Please specify. Plain text
descriptions are often vague, I admit.
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top