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.

PA not work when ANT switch connected?

Status
Not open for further replies.

cybq

Junior Member level 3
Joined
May 31, 2003
Messages
31
Helped
2
Reputation
4
Reaction score
2
Trophy points
1,288
Location
sz
Activity points
299
:oops: I designed a PA with MMIC(FETs internal) at 1.9GHz, which have the negative gate voltage produced within internal OP. It works normal when connected capacitor coupled to spectrum analyzer instead of antenna switch(MMIC 6pin). Unfortunately, it didn't work when loaded with antenna switch via the same capacitor coupled, and produce very large cuurent. Why like this? I just cann't understand and solve it indeed. Help me!! Thks in advance.
 

You should check the match first. To study the antenna switch data sheet. Check antenna switch input impedance.
And check your PA impedance. Maybe you need a match circuit.
 

The only reason why the DC current changes with a AC coupled load is that the PA oscillates when terminated by the antenna switch. If this is the case, you need carefull oscillation analysis.
 

yep.. looks like an oscillation to me...

check the input impedance of the switch, maybe it is not good enough for the PA...
I wonder why you do not try to put an isolator between the PA and the switch..? size/weight or cost?...
if you cannot afford putting an isolator, try to match you switch input impedance to what the PA wanna see to be happy :)

Cheers...
 

I have carefully checked the antenna switch datasheet that S11=-25dB @ 2GHz. In theorily, such case should be the same result as directly with 50ohm cable, while it work normal capacitor coupled to cable.

Isolator is not used for the cost, complexity and space, also its additional
insert loss.

When there was very large cuurent, I found its gate voltage was not negative! It seems that the internal OP work abnormal.
 

this is parasytic oscilation caused by a missmatch in the capacitor coupling section and is a very dirty way to couple rf man....!!!
check the rating of the cap carefully also the path to the relay must be straight

for this power level your using a stripline stage for tuning ??? the amp section


i think instead of the cap you should make a small enclosed dog leg stub coupling coil this will then have sutable capacitace while being above the groundplane

other than this use a better load and tune the amp properly to the load the minimise harmonics and current ig, ic,

maybe its feed back to a previous stage
of mutliplications etc

in this case screen appropriatly

try a hatched ground plane above and a solid unbrocken ground plane under side
place more thrus near the output in a measured fasion
even diameter of thru matters
keep to 1000th lamda dia per 32/th lamda spacing minimum above 1ghz

a tip is dont just use a cap this is cheeting and cheeting isnt allowed

filter the output may be a good idea


also filtering the input to the pa may be needed
as a missmatch here will carry thru problems

ensure a clean psu measure the chokes are working by coupling for stray capacitance

and add the unit to a di cast box with mounting posts at each 1" space for 1-10ghz made from phosphor bronze{computer motherboard fixign hardware i find brilient}


for a dummy load
use a single 50R tube made for the job
direct coupled and tap spectrum analyzer to it

mount on a socket as close to the output as possible
only then tapping off a little rf to see what specrum trace it has

so
my advice is also to check for microphonic pickup
as vibration can cause lots of problems
even a blob of flux can be deadly

so clean carefully all parts
and solder with very good quality solder
mostly silver is best using a hotter iron and nice quality board

or the high tin content solder

use a gas filled relay for transmition at > 1ghz
 

!!!MONKEY!!! Why is it the case:
...check the rating of the cap carefully also the path to the relay must be straight


... using a stripline stage for tuning ??? the amp section
I don't think you mean that you can mechanically tune stripline lengths, or can you ?

... a small enclosed dog leg stub coupling coil
this will then have sutable capacitace ...
Poor me I don't know what it is. Recommend a book ?


...keep to 1000th lamda dia per 32/th lamda spacing

Recommend a book ?

ensure a clean psu ...
psu ???Recommend a book. Please!

...use a single 50R tube made for the job
direct coupled and tap spectrum analyzer to it

... microphonic pickup
as vibration can cause lots of problems
even a blob of flux ...
??? Maybe I need a EMC book recommendation
Regards,
StoppTidigare
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top