Searching \ for '[EE]: Directional couplers & T/R test sets' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=directional+couplers
Search entire site for: 'Directional couplers & T/R test sets'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Directional couplers & T/R test sets'
2002\07\05@173440 by Brendan Moran

flavicon
face
Hi,

I'm hoping that we've got a few RF engineers floating around that can help
me out with this.  I'm a digital systems type person, myself, but I'm
working in a place that needs me to do work on RF circuitry, so I'm a bit
stuck.

I have a relatively good hybrid coupler that was built as a power splitter.
It has -41dB isolation and -3.08dB ouput ports.

I want to use it to measure scattering parameters of a preamplifier but I
don't know how to connect it.  I know that the HP transmission/reflection
test set, which I am trying to emulate with this device, is a 4-port unit,
and that 2 of the ports go to a network analyser input, 1 port that get RF
in, and 1 port that goes to the DUT.  I do not, however know what the 2
ports that go to the inputs are.

My directional coupler has the following ports: In, 50ohms, 0 degrees and 90
degrees.  What ports do I connect to the network analyser?

I assume that the IN port gets the RF out from the network analyser, and
that the 50-ohm port goes to one of the inputs (but don't know which one),
but I don't know whether the 90-deg or the 0-deg port goes to the DUT.

Anyone done this before?

--Brendan

--
http://www.piclist.com hint: To leave the PICList
spam_OUTpiclist-unsubscribe-requestTakeThisOuTspammitvma.mit.edu


2002\07\05@212523 by Jim

flavicon
face
 "I have a relatively good hybrid coupler"

I, uh, have always used 'directional couplers'
to sample the RF in each direction in order
to test and characterizs devices.

Have you got some new technique that you plan to use
the hybrid in to do the same thing?

As to 'four ports' on a HP transmission/reflection
test unit -

- the old 8475A "S-Param Test Set" I employ has five RF
 'ports':

1) One RF input Type-N. This must actually deliver RF to both
  the 'Reference' input channel on the 8410B/8411A as well as
  RF to the 'test' channel through a suitable a RF 'splitter'.

2) Two ports (APC-7)to the 8410B/8411A harmonic converter
  (which is basically a two-channel amplitude and phase
   measuring 'receiver').
  These are outputs (I have to presume) are from directional
  couplers that a) sample the RF from an RF signal generator
  and also sample the RF on the 'test' channel as per the
  S parameter to be measured.

3) Two (APC-7) ports to the DUT for transmision (S21) and Refl (S11).
  Internal directional couplers in the 8475A do the 'magic' in
  sampling the RF to/through/reflected from the DUT and direct this
  RF to the 8410B/8411A's 'Test' channel.

I use I only deal with three ports - the RF input port and the
two APC-7 ports to the DUT. The 8410B/8411A harmonic converter
'bolts' into position in an area aboard the 8475A S-Parm Test
Set.

This is rather old gear today - but the basics are there!

Jim


{Original Message removed}

2002\07\06@111429 by Bill Colville

flavicon
face
Brendan,
I have never heard of using a 90 degree hybrid for making S
parameter measurements. They are usually used for comparing
impedences. I don't see any practical way to do it.

Lacking a proper S parameter test set, as described by another
poster, the hook-up for using a separate power splitter and
directional coupler is as follows:

The RF output from the network analyzer goes to the input of the
power splitter. One output from the power splitter goes to the
Reference input of the analyzer. When measuring an active device,
such as your preamp, the other output from the power splitter goes
through an attenuator to the input of the directional coupler and the
non-coupled output goes to the input of the DUT. The coupled
output of the directional coupler goes to either the A or B input on
the analyzer. The attenuator is to reduce the input voltage so it will
not overload the DUT while still providing sufficient Reference signal
for the analyzer. The other, A or B, input of the analyzer goes to
the output of the DUT, might require an attenuator depending on
the gain of the device. This set-up will measure S-11 and S-21, to
measure S-12 and S-22 reverse the DUT. Make sure there is no
DC voltage on the input and output connectors of the DUT, if there
is, a DC block or bias T must be used to isolate it from the
analyzer detectors.

HTH,
Bill, W3NMK

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\07\08@150340 by Brendan Moran

flavicon
face
Thanks Bill & Jim
I guess I assumed that I had something that might work, but when I tried to
find info on using directional couplers, I pretty much came up blank.  So,
how does a directional coupler for low RF (10-20MHz) actually work?

--Brendan

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\07\08@153120 by Jim

flavicon
face
10 - 20 MHz?

Ouch!

I've got directional couplers that are 'effective'
as low as 100 MHz - amd as most of my work in
the past has dealt with this range I've been
okay.

I've got some on-going work (a filter) in the range
of .5  to 5 MHz right now.

I am able to measure S21 (Insertion Loss) using a
spectrum analyzer and signal generator, but shortly
I've got to measure S11 (Return Loss or 'VSWR')
as well.

I'm going to use a crude 'bridge' in combination with
a dual-trace scope or a Vector Voltmeter to accomplish
this in a round-about fashion (measure complex Z and
then compute the Return Loss/S11).

http://www.dallas.net/~jvpoll/tech/ph_mag.html

Jim


{Original Message removed}

2002\07\08@154820 by Brendan Moran

flavicon
face
> 10 - 20 MHz?
>
> Ouch!

Sucks, eh?  For the same project, I've had to design and build a preamp with
a 0.5dB NF and a gain of  >= 26dB.  Ended up using components with a minimum
frequency of 450MHz.  They work though...  I just have to fix the matching
now.

--Brendan

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\07\08@172936 by Jim

flavicon
face
.5 dB NF?

Why all that trouble for HF?

Isn't your ambient noise (atmosphereic, man-made, etc)
going to swamp your preamp nose by a good margin?

(Maybe this isn't for use in free-space, eh?)

BTW - if you have to 'fix the matching' - that will,
in most designs, also affect the 'noise'. There is
something called 'best noise match' and it doesn't
always follow the best 'match' or gain point ...

Jim

{Original Message removed}

2002\07\08@174249 by Brendan Moran

flavicon
face
> .5 dB NF?
>
> Why all that trouble for HF?
>
> Isn't your ambient noise (atmosphereic, man-made, etc)
> going to swamp your preamp nose by a good margin?
>
> (Maybe this isn't for use in free-space, eh?)

All goes into a shielding room.

> BTW - if you have to 'fix the matching' - that will,
> in most designs, also affect the 'noise'. There is
> something called 'best noise match' and it doesn't
> always follow the best 'match' or gain point ...

Yeah, I know.  However, since there are no s-parameters available for these
parts at this frequency, I'm pretty much up the creek until I can get some
way of measuring s-parameters (I'd just do a straight impedance measurement,
but it's not as reliable), though the interesting thing is that the current
setup is doing adequately without a proper match.  I have the feeling that a
good match could give me a few dB extra to play with.

--Brendan

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\07\08@232640 by Bill Colville

flavicon
face
Brendan,

The directional coupler that I use is a KDI Electronics model
DCK-1010. It has 10 dB coupling and is rated for use from
1.0 to 1000 MHz. Minicircuits ( http://www.minicircuits.com)
makes a line of very inexpensive directional couplers that have a
frequency range from  0.2 MHz up with various coupling coefficients.
They also make a line of inexpensive power splitters for the same
range of frequencies. They also have application notes on their
Web site that explain the design and use of directional couplers

Bill, W3NMK


> I guess I assumed that I had something that might work, but when I tried to
> find info on using directional couplers, I pretty much came up blank.  So,
> how does a directional coupler for low RF (10-20MHz) actually work?

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\07\09@140012 by Peter L. Peres

picon face
On Mon, 8 Jul 2002, Jim wrote:

>.5 dB NF?
>
>Why all that trouble for HF?
>
>Isn't your ambient noise (atmosphereic, man-made, etc)
>going to swamp your preamp nose by a good margin?
>
>(Maybe this isn't for use in free-space, eh?)
>
>BTW - if you have to 'fix the matching' - that will,
>in most designs, also affect the 'noise'. There is
>something called 'best noise match' and it doesn't
>always follow the best 'match' or gain point ...

May I ask your opinion on adjusting input match for best S/N using a
noise bridge and examining the output only (i.e. no expensive equipment) ?

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email .....listservKILLspamspam@spam@mitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@140024 by Peter L. Peres

picon face
On Mon, 8 Jul 2002, Jim wrote:

{Quote hidden}

Try to make a bifilar wound directional coupler on a toroid or
snap-together ferrite transformer for 5MHz. It will very likely work for
you. 30 turns at 5MHz would be too much I think. Surely you have a core
like this in the junkbox.

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservspamKILLspammitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@140031 by Peter L. Peres

picon face
On Mon, 8 Jul 2002, Brendan Moran wrote:

>Thanks Bill & Jim
>I guess I assumed that I had something that might work, but when I tried to
>find info on using directional couplers, I pretty much came up blank.  So,
>how does a directional coupler for low RF (10-20MHz) actually work?

The same as for any frequency. A correctly terminated transmission line
does that. The easiest way to understand a d.c. is to imagine that it has
one line terminated with Z0 at both ends and the other is not matched and
fed by a generator. This will cause a standing wave in the unmatched line
and thus in the coupler.  Examine what this causes to appear on the
termination resistors of the matched side of the d.c. assuming that its
coupling coefficient is 1. This is really used to measure SWR usually but
it makes for a good example to understand what is going on. Notice that
for a matched d.c. L/lambda is irrelevant but when it is unmatched
L/lambda influences the magnitude of the signals on the matched outputs.
d.c.  for HF are usually made by winding bifilar (or more) wire on a
toroid to achieve a resonable proportion of lambda. (L is the electrical
length of the d.c. and lambda the wavelength at the frequency used).

hope this helps,

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email .....listservKILLspamspam.....mitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@142316 by Jim

flavicon
face
  "Try to make a "

'cause this technique works, is perfectly
valid AND works from DC to as high as your
construction techniques will allow (making
it largely 'frequency independent') ...

Jim

{Original Message removed}

2002\07\09@151225 by Peter L. Peres

picon face
On Tue, 9 Jul 2002, Jim wrote:

>   "Try to make a "
>
>'cause this technique works, is perfectly
>valid AND works from DC to as high as your
>construction techniques will allow (making
>it largely 'frequency independent') ...

Hey, you're the guy who advocates buying suitable equipment from Ebay.
What would you use for this measurement ?

Peter

PS: BTW, I did 'try' and make what I proposed, for 27MHz.

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email EraseMElistservspam_OUTspamTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@152413 by Sean H. Breheny

face picon face
I'm not going to weigh in one way or another on the directional coupler
discussion, but I will say that buying equipment from eBay is great! I've
been working on a microwave project (weather radar) and I went from having
no microwave test equipment to having everything I need for this project
(signal generator, 40W dummy load, 20W power amp, low-loss microwave coax,
power detector, and directional coupler) for only a few hundred dollars.

Sean

At 10:11 PM 7/9/2002 +0300, you wrote:
{Quote hidden}

-------------------------------------------
Introducing NetZero Long Distance
Unlimited Long Distance only $29.95/ month!
Sign Up Today! http://www.netzerolongdistance.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email @spam@listservKILLspamspammitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@154824 by Jeremy Walworth

flavicon
face
DIY weather radar or a commercial product?

-J

At 03:23 PM 7/9/02 -0400, you wrote:
>I'm not going to weigh in one way or another on the directional coupler
>discussion, but I will say that buying equipment from eBay is great! I've
>been working on a microwave project (weather radar) and I went from having
>no microwave test equipment to having everything I need for this project
>(signal generator, 40W dummy load, 20W power amp, low-loss microwave coax,
>power detector, and directional coupler) for only a few hundred dollars.
>
>Sean

Jeremy S. Walworth
"I fly, I sail, I throw caution to the wind"  -Jimmy Buffett
http://www.jeremyanddarcy.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email KILLspamlistservKILLspamspammitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@160324 by Jim

flavicon
face
Peter penned: "What would you use for this measurement ?

Upon further reflection (no pun intended) -
  - an HP Vector Impedance Meter.

HP made a very nice one with a top end of 108 MHz.

The circuit I 'dreamed up' over a ten years ago
while moonlighting as a broadcast engineer in the
form of that resistive bridge was a product of
inspiration at the eleventh hour -

- as I was afraid a certain GR Z bridge I had planned
on borrowing might be not be available after all.

The GR Z bridge (or some suitable substitute) was
needed for the measurement of a most critical
parameter: the Z (impedance) as seen at the base
of 240 plus foot 1/4 wave guyed vertical for 1.06 MHz.

I had been contracted at the time to install and
tune a new ATU - Antenna Tuning Unit - at a 10 KW
AM broadcast station.

As it was - I did get to borrow the GR bridge, but
the inspired circuit *still* played a critical
part - I had to re-tune the ATU and part of the
process calls for finding the parallel resonant
point ... my "R" bridge works optimally for this
purpose - as one can easily see the phase change
most radically with frequency as one tunes with a
signal generator through the resonance point on a
tuned circuit (the ATU in this case) - at this point
the composite Z (R+jx) remains fairly constant and
causes little change in the amplitude of a signal, but
the phase stands out and identifies the resonant
point like a sore thumb.

Whilst your suggestion is most likely valid - I doubt
you can assure accurate numerical results without
some difficulty (calibration charts etc) over several
octaves (unto a decade) of bandwidth ...

Jim

{Original Message removed}

2002\07\09@160954 by Peter L. Peres

picon face
On Mon, 8 Jul 2002, Jim wrote:

>I'm going to use a crude 'bridge' in combination with
>a dual-trace scope or a Vector Voltmeter to accomplish
>this in a round-about fashion (measure complex Z and
>then compute the Return Loss/S11).
>
>http://www.dallas.net/~jvpoll/tech/ph_mag.html

Nice page.

Peter

PS: I think that if you substitute a pulse generator for the RF source you
can do cable TDR with your setup ;-)

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservTakeThisOuTspammitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@162736 by Jim

flavicon
face
HP 4815A Vector Impedance Meter

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=1722056183

http://www.tucker.com/NCCatalogs/SrchExp/sexpnc372.asp

http://www.e-test.com/TABL1-10.HTM

o Hewlett Packard 4815A Vector Impedance Meter
o Measures impedance magnitude and phase
o 500 kHz to 108 MHz in five bands
o Impedance range 1 Ohm to 100 kohms
o Phase angle 0 to 360 degress in two ranges
o Internal calibration 100 ohms resistive
o Analogue outputs for magnitude and phase indications

Went for a paltry US $137.50 ...



HP 4191A RF Impedance Analyzer
1 - 1000 MHz (this instrument seems like it's more suitable
for component testing)

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=1747099222

www.testequipmentdepot.com/usedequipment/hewlettpackard/impedanceanal
yzers/4191a.htm

Weighing in currently at US $900


or


HP 4193A RF Impedance Analyzer
400 KHz to 110 MHz (a replacement for the 4815A)
None currently on eBay ...
http://www.testequity.com/products/149/

Jim


{Original Message removed}

2002\07\09@163150 by Peter L. Peres

picon face
On Tue, 9 Jul 2002, Jim wrote:

...
>I had been contracted at the time to install and
>tune a new ATU - Antenna Tuning Unit - at a 10 KW
>AM broadcast station.

Ok. 10kW is way out of my league ;-).

{Quote hidden}

I aggree but as you would surely say, it takes cojones to do that with
what amounts to diy equipment and then turn on the juice (even slowly).
100W RF has pretty impressive built-in fault indicators afaik. 10kW from a
tube PA more so ...

>Whilst your suggestion is most likely valid - I doubt
>you can assure accurate numerical results without
>some difficulty (calibration charts etc) over several
>octaves (unto a decade) of bandwidth ...

Of course not. You need to spot-check the d.c. itself from time to time at
the working frequency. However the d.t. is a SWR meter and it takes a
couple of seconds (and four ganged good quality 100ohm cermet trimmers).
This is harder to explain than to do. I got the right kind of wire in 3
tries. The best kind of winding for wideband is with progressively more
spaced-out turns, but this does not work on a d.c. so I just put in the
turns (deliberately disordered winding).

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email spamBeGonelistservspamBeGonespammitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@163815 by Peter L. Peres

picon face
Ok, thanks. I am not going to buy any of these things now but it is good
to know.

Peter

On Tue, 9 Jul 2002, Jim wrote:

{Quote hidden}

>{Original Message removed}

2002\07\09@164702 by Peter L. Peres

picon face
(Jim?) does anybody know how small a signal can be while the HP4815A still
reads right ? I.e. can it be used to measure verly low signals (say 10mV
pk-pk) and give meaningful results ?

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email TakeThisOuTlistservEraseMEspamspam_OUTmitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@171825 by Jim

flavicon
face
I checked the specs for the 4815A (in the '73 catalog)
and it doesn't say - the specs for the 4193A regarding
output appears thusly:

 "Test Level: constant current source"

and shows a small chart describing the RF currents used
on each.

Briefly, the max voltage appears to be 1 V RMS for the
100K Z range (given a constant RF I of 10 uA) -

- to 1 mV RMS for a current of 100 uA on the 10 Ohm Z range.

Of course, when measuring the low-level input Z of a
pre-eamp *any* setup of equipment has to be verified
NOT to be driving the pre-amp into non-linear region
of operation ...

I measured the input Z of some popular 2-Meter rigs once
upon a time over 20 years ago using an HP 8405A - and
kept the RF level as low as possible to avoid over-driving
the first RF stage. As I recall - the input match was horrid!

On the order of 10 to 11 dB RL at 146 MHz! (On a Yaesu FT-227
"Memorizer" I think it was.)

Jim

{Original Message removed}

2002\07\09@174247 by Brendan Moran

flavicon
face
Perhaps I should refer to what I am actually trying to do here.  I have this
same preamp.  The input is really not much of a question since the source
for it is a >5k + j0 ohm coil.  It's tough to match, but I have something
that works.  Once again, magnetic components for this particular project are
a nono.  It also needs to be small in size.  This eliminates the use of
large value inductors, and large trimmer caps.  It's the output that's
giving me grief.  I need to get a measurement of the current impedance
before I can do much to fix it, since the only specs I have for the active
component are for a long way outside of my freq. range.

The test equipment I have available is as follows:
1xHP4395A network analyser
1xHP54645A DSO
1xHP43961A Impedance test set
1x Nice function generator.
1x Not so nice function generator.

1x Hybrid power splitter

With that, and something I can build easily, how can I measure s-parameters
or impedances accurately?

--Brendan

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservspamTakeThisOuTmitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@174945 by Brendan Moran

flavicon
face
I forgot to mention that the circuit which Jim showed looks useful.  I'm
going to try it.
> With that, and something I can build easily, how can I measure
s-parameters
> or impedances accurately?

--Brendan

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listservEraseMEspam.....mitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@190155 by Jim

flavicon
face
It looks to me like you've got a lot of
what you need in that "HP 4395A network
analyzer".

I would take my active device and build a
small fixture and simply characterize it
using that HP 4395A at my frequency of
interest.

Does that particular 4395A have any options
(like the Impedance Measurement Function
and RF Impedance Test Kit)? (I don't own
one - I'm looking at it in the HP catalog ...)

Oh - I see you have the HP43961A Impedance test
set. (Too bad you don't have the S-Parameter
Test Set for the 4395A.)

With the Impedance Test Set you can get the Z
looking into a 'network' (like your pre-amp).

Now you need a test set-up to measure Insertion
Loss (and Phase change) and come up with gain
(or loss) of a network.

What are those four ports on the HP 4395A:

1) RF Output
2) R(ef) Input?
3) A and
4) B inputs?

I think you are closer than you think to a
solution!

Jim


{Original Message removed}

2002\07\09@191319 by Brendan Moran

flavicon
face
> It looks to me like you've got a lot of
> what you need in that "HP 4395A network
> analyzer".
>
> I would take my active device and build a
> small fixture and simply characterize it
> using that HP 4395A at my frequency of
> interest.

Ah, yes, it is that fixture that seems to be the problem

> Does that particular 4395A have any options
> (like the Impedance Measurement Function
> and RF Impedance Test Kit)? (I don't own
> one - I'm looking at it in the HP catalog ...)

It has impedance analyser, spectrum analyser and network analyser modes.

> Oh - I see you have the HP43961A Impedance test
> set. (Too bad you don't have the S-Parameter
> Test Set for the 4395A.)

That's what I said.

{Quote hidden}

Correct

> I think you are closer than you think to a
> solution!

I expect I am.  However, this, unfortunately, falls under the "so close, yet
so far" range, I think.

I have just been having trouble figuring out how to measure reflection
coeficients.  Transmission loss is not a big concern to me, since that one's
easy to measure with a network analyser.

Thanks again

--Brendan

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email EraseMElistservspammitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@210806 by Jim

flavicon
face
Brendan - you've got the capability to measure
the reflection coefficient - the Impedance Test
Set you've got can be used to measure Input
Impedance numbers that can then be converted into
S-Parameter values (if that's what's needed for
design).

As to characterizing the device - I would build
a quickie little breadboard with the ability
to 'bias-up' the part (including the Drain) and
then terminate the output (AC-wise) with a
50 Ohm term and then measure the input Z (which
can then be converted into an actual S11 value
if need be). You will still have to be careful - this
device will *still* have gain across a wide range
of frequencies (and it may want to oscillate!).

You might need an inductor (RF choke) on the
drain in order to provide DC feed while making
that point isolated for AC - then AC couple (using
a capacitor) into a 50 Ohm load (or one of the
channels of the NA for S21 values!).

Jim


{Original Message removed}

2002\07\09@222955 by Jim

flavicon
face
For future reference (note to self!):

Agilent AN 154
S-Parameter Design
Application Note

http://www.sss-mag.com/pdf/AN154.pdf

Jim

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservEraseMEspamEraseMEmitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@224802 by Sean H. Breheny

face picon face
Hi Jeremy,

DIY. For a web page on it that is now way behind the current level of
progress (I am about 1 month away from testing the system), see:

http://www.rocket-roar.com/BT/wsr.html

Sean

At 03:32 PM 7/9/2002 -0400, you wrote:
{Quote hidden}

-------------------------------------------
Introducing NetZero Long Distance
Unlimited Long Distance only $29.95/ month!
Sign Up Today! http://www.netzerolongdistance.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistservTakeThisOuTspamspammitvma.mit.edu with SET PICList DIGEST in the body


2002\07\09@232328 by Jim

flavicon
face
The follwoing was excerpted from:

T&MW Online-- What are S-parameters, anyway?
http://www.tucker.com/Manufacturer/TMWorld/TMFEB01_Spara.pdf

Excerpt:

The measurement examples and corresponding S-parameter
matrices shown in Figure 6 are based on a vector network
analyzer and S-parameter test set having 50-W characteristic
impedance.

If you connect a 25-W resistor to port 1 and leave port 2
unused (Figure 6a), the S-parameter matrix reduces to a
single parameter (the resistor is a one-port network requiring
a 1x1 matrix), which represents the reflection coefficient.

The following equation (Ref. 3) describes the reflection
coefficient due to a load impedance ZL with respect to
characteristic (system or line) impedance Z0:

rho = (ZL - Z0)/(ZL + Z0)    (Eq. 21)

For the values shown in Figure 6a,

rho = s11 = (25 W-50 W)/(25 W+50 W) = -0.333,

or 0.333 with a 180° phase angle.


* * * * * * * * * * * * * * * * * * * * * * * * * *

Further ref:

http://www.sss-mag.com/pdf/AN154.pdf
http://www.sss-mag.com/pdf/hpan95-1.pdf

* * * * * * * * * * * * * * * * * * * * * * * * * *

Jim


{Original Message removed}

2002\07\10@142123 by Peter L. Peres

picon face
I really don't know the answer but maybe you could use a spectrum analyzer
and a noise source. This will not tell you the impedance but you should be
able to improve your circuit by trial and error. It may also show you any
resonances in band, and noise peaks. A dense 'fence rail' signal may
replace the noise generator and make the noise floor easier to see on the
specan when you tweak the circuit.

Peter

On Tue, 9 Jul 2002, Brendan Moran wrote:

{Quote hidden}

--
http://www.piclist.com hint: To leave the PICList
RemoveMEpiclist-unsubscribe-requestKILLspamspammitvma.mit.edu


2002\07\10@144207 by Brendan Moran

flavicon
face
> I really don't know the answer but maybe you could use a spectrum analyzer
> and a noise source. This will not tell you the impedance but you should be
> able to improve your circuit by trial and error. It may also show you any
> resonances in band, and noise peaks. A dense 'fence rail' signal may
> replace the noise generator and make the noise floor easier to see on the
> specan when you tweak the circuit.

Well, I have been developing a sort of a theory of operation when working
with analogue circuits:  Never use trial and error where calculations will
do the job.  So, I admit that I don't particularly like the idea of using
trial and error to solve this problem.

Aside from that, I didn't list the noise figure meter because I didn't think
it was relevant.  The problem I've had with it, though, is it measures
consistently higher than I expect.

For example: I have an antenna pre-amp with a rated NF of 0.5dB.  It has
50-ohm in and 50-ohm out ports.  When I hook it up to the NF meter, (which I
have calibrated using a noise source and the same 2 cables off the noise
source, only with a F-BNC to F-BNC connector, that I use with the antenna
pre-amp) It gives me 11dB or so.  Hardly an acceptable error.

It's primarily for that reason that I didn't mention the NF meter.

Anyways, got any clues on that one?

--Brendan

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestSTOPspamspamspam_OUTmitvma.mit.edu


2002\07\10@180007 by Peter L. Peres

picon face
On Wed, 10 Jul 2002, Brendan Moran wrote:

>> I really don't know the answer but maybe you could use a spectrum analyzer
>> and a noise source. This will not tell you the impedance but you should be
>> able to improve your circuit by trial and error. It may also show you any
>> resonances in band, and noise peaks. A dense 'fence rail' signal may
>> replace the noise generator and make the noise floor easier to see on the
>> specan when you tweak the circuit.

BTW the noise peaks and the noise floor slope can let you infer something
useful about the imaginary part of the input. If you are careful you
should be able to read at least the modulus of the imaginary component
from the specan if not the phase.

Peter

--
http://www.piclist.com hint: To leave the PICList
spamBeGonepiclist-unsubscribe-requestSTOPspamspamEraseMEmitvma.mit.edu


2002\07\10@180017 by Peter L. Peres

picon face
On Wed, 10 Jul 2002, Brendan Moran wrote:

>> I really don't know the answer but maybe you could use a spectrum analyzer
>> and a noise source. This will not tell you the impedance but you should be
>> able to improve your circuit by trial and error. It may also show you any
>> resonances in band, and noise peaks. A dense 'fence rail' signal may
>> replace the noise generator and make the noise floor easier to see on the
>> specan when you tweak the circuit.
>
>Well, I have been developing a sort of a theory of operation when working
>with analogue circuits:  Never use trial and error where calculations will
>do the job.  So, I admit that I don't particularly like the idea of using
>trial and error to solve this problem.

If you don't have the parameters for the circuit the next best thing is to
get some real world data to give you a starting point.

{Quote hidden}

Nope. I know too little about your circuit to be able to help. Sorry.

Peter

--
http://www.piclist.com hint: To leave the PICList
KILLspampiclist-unsubscribe-requestspamBeGonespammitvma.mit.edu


More... (looser matching)
- Last day of these posts
- In 2002 , 2003 only
- Today
- New search...