Has anyone here built a GOOD antenna/network analyzer?

From what I can determine, the MFJ stuff doesn't resolve the sign of
reactances, nor does it have a computer interface. I was looking at Analog
Devices' AD8302 IC (http://www.analog.com) for this application. This is a
single-chip device that measures the gain and phase difference between 2 RF
inputs up to 2.7 GHZ with very high accuracy. It occurred to me that this
chip, together with a PIC that has A-D conversion, and a computer-controlled
station transceiver, could form the basis for a very nice programmable
antenna/network analyzer that would run rings around the MFJ units.

Comments?

Joe
W3JDR

W3JDR wrote:

Has anyone here built a GOOD antenna/network analyzer?

From what I can determine, the MFJ stuff doesn't resolve the sign of
reactances, nor does it have a computer interface. I was looking at Analog
Devices' AD8302 IC (http://www.analog.com) for this application. This is a
single-chip device that measures the gain and phase difference between 2 RF
inputs up to 2.7 GHZ with very high accuracy. It occurred to me that this
chip, together with a PIC that has A-D conversion, and a computer-controlled
station transceiver, could form the basis for a very nice programmable
antenna/network analyzer that would run rings around the MFJ units.

Comments?

Joe
W3JDR

Hi Joe,
Look at:
http://www.njqrp.org/antanal/
Perhaps your IC could be combined with this controller.

Dale W4OP

Dale,
Actually, I was thinking that the controller is a PC or laptop. It would
control the station transciever as a swept low-power frequency source, and
get the phase and gain readings from the PIC that would be paired with the
Analog Devices part (and of course a directional coupler of some sort...even
a simple bridge would suffice). This design could result in a very simple,
low-cost, and powerful vector network analyzer for the average ham.

Joe
W3JDR


"Dale Parfitt" wrote in message
...


W3JDR wrote:

Has anyone here built a GOOD antenna/network analyzer?

From what I can determine, the MFJ stuff doesn't resolve the sign of
reactances, nor does it have a computer interface. I was looking at
Analog
Devices' AD8302 IC (http://www.analog.com) for this application. This is
a
single-chip device that measures the gain and phase difference between 2
RF
inputs up to 2.7 GHZ with very high accuracy. It occurred to me that
this
chip, together with a PIC that has A-D conversion, and a
computer-controlled
station transceiver, could form the basis for a very nice programmable
antenna/network analyzer that would run rings around the MFJ units.

Comments?

Joe
W3JDR

Hi Joe,
Look at:
http://www.njqrp.org/antanal/
Perhaps your IC could be combined with this controller.

Dale W4OP

The chip you're looking at does have a 180 degree ambiguity unresolved.
Makes it tough to do the job you're asking of it. Wait a couple weeks and
look on Google for N2PK. You'll find a real vector network analyzer good
from 50 KHz to 60 MHz with narrow band application to 2 meters and 432 MHz.

W4ZCB

Harold,

The chip's phase measurement transer function has a positive slope for phase
angles less than 0deg and a negative slope for phase angles greater than
0deg. So it seems to me that resolving the 180 deg ambiguity is just a
matter of knowing which slope you're on. This might be done by switching in
a small known reactive component and observing which direction the phase
output changes. Once you know which slope you're on, I think you're home
free...no??

Joe
W3JDR
"Harold E. Johnson" wrote in message
news:Yhiab.496033$Ho3.84151@sccrnsc03...
The chip you're looking at does have a 180 degree ambiguity unresolved.
Makes it tough to do the job you're asking of it. Wait a couple weeks and
look on Google for N2PK. You'll find a real vector network analyzer good
from 50 KHz to 60 MHz with narrow band application to 2 meters and 432
MHz.

W4ZCB

The chip's phase measurement transer function has a positive slope for
phase
angles less than 0deg and a negative slope for phase angles greater than
0deg. So it seems to me that resolving the 180 deg ambiguity is just a
matter of knowing which slope you're on. This might be done by switching
in
a small known reactive component and observing which direction the phase
output changes. Once you know which slope you're on, I think you're home
free...no??

Do it the way you want to.

W4ZCB

The chip's phase measurement transer function has a positive slope for
phase
angles less than 0deg and a negative slope for phase angles greater than
0deg. So it seems to me that resolving the 180 deg ambiguity is just a
matter of knowing which slope you're on. This might be done by switching
in
a small known reactive component and observing which direction the phase
output changes. Once you know which slope you're on, I think you're home
free...no??

Do it the way you want to.

W4ZCB

The chip's phase measurement transer function has a positive slope for phase
angles less than 0deg and a negative slope for phase angles greater than
0deg. So it seems to me that resolving the 180 deg ambiguity is just a matter
of knowing which slope you're on. This might be done by switching in
a small known reactive component

You can do this with an MFJ or any other analyzer to resolve the sign ambiguity
of X. Switch in small known reactance, see which way X goes. Also, hooking a
computer interface to an MFJ would probably not be that much more difficult
than interfacing with the AD chip. So, no I don't think this would run rings
around an MFJ. However, if you do go this route, I would be very interested in
how it turns out.
73 Gary N4AST

The chip's phase measurement transer function has a positive slope for phase
angles less than 0deg and a negative slope for phase angles greater than
0deg. So it seems to me that resolving the 180 deg ambiguity is just a matter
of knowing which slope you're on. This might be done by switching in
a small known reactive component

You can do this with an MFJ or any other analyzer to resolve the sign ambiguity
of X. Switch in small known reactance, see which way X goes. Also, hooking a
computer interface to an MFJ would probably not be that much more difficult
than interfacing with the AD chip. So, no I don't think this would run rings
around an MFJ. However, if you do go this route, I would be very interested in
how it turns out.
73 Gary N4AST

Harold,

The chip's phase measurement transer function has a positive slope for phase
angles less than 0deg and a negative slope for phase angles greater than
0deg. So it seems to me that resolving the 180 deg ambiguity is just a
matter of knowing which slope you're on. This might be done by switching in
a small known reactive component and observing which direction the phase
output changes. Once you know which slope you're on, I think you're home
free...no??

Joe
W3JDR
"Harold E. Johnson" wrote in message
news:Yhiab.496033$Ho3.84151@sccrnsc03...
The chip you're looking at does have a 180 degree ambiguity unresolved.
Makes it tough to do the job you're asking of it. Wait a couple weeks and
look on Google for N2PK. You'll find a real vector network analyzer good
from 50 KHz to 60 MHz with narrow band application to 2 meters and 432
MHz.

W4ZCB