How does an SWR-Meter work?
Andreas Aigner
Who can tell me, how an SWR-Meter works? I am looking for good literature
and/or Web-Site.
tnx, 73 de Andreas.
Matt Erickson
Two seperate rectifier diodes convert the RF into DC voltage, which drives
the meter movement to a level that is calibrated be resistors on the meters
circuit board.
the forward reading is compared to the reflected level, to get the SWR
reading.
Al, N2NKB
The coupler has a mainline, which connects the rig to the antenna.
It also has a forward port, and a reverse port.
Suppose you have a 100 watt signal with a 40/40 dB coupler.
The forward port would pick up 100 watts (50 dBm) minus 40 dB, which is 10 dBm, or 10 mW.
The reverse port would pick up nothing if the coupler directivity was perfect, and the load is perfect.
If you unplug the antenna, all of the signal would reflect back to the reverse port, and you would
read the 10 mW.
But let's suppose that your antenna VSWR is 1:1.5, or a 14 dB return loss.
This would reflect back 50 dBm - 14 dB minus 40 dB, which is - 4 dBm or 400 uW.
As Matt indicated, you just peak detect the two ports with a diode and filter cap, and compare
their values on two meter movements, or by using a single meter movement, you adjust the forward
port to the cal position, and switch to the reverse position.
Good luck Andreas. You will find that you can easily build your own dual directional coupler simply
with a copper circuit board and an Exacto knife.
Al
Matt Erickson wrote in message <38DBE1AD...@hotmail.com>...
Reg Edwards
First thing to remember there is no such thing as a genuine SWR meter.
Beware of adverts.
Such a theoretical meter measures the ratio of the voltages (or
currents) at two different places along a transmission line spaced
1/4-wave apart. Unless such a meter has simultaneous access to these
two different locations along the line then it cannot indicate SWR.
And as you may already be aware, the common or garden SWR meter has
access only to one location on the line - the place where it is
fitted.
The location in the line of the usual SWR meter, ie., immediately at
the output of the transmitter, actually indicates the SWR on a
non-existent line - ie., the non-existent line between the meter and
the transmitter.
How can a line of zero, or even very short length, have standing waves
?
--
Regards, Reg.
Al, N2NKB
As a sanity check, you can measure the SWR normally. Then, reverse the mainline input/output connections.
You should see the same readings, but swap the forward for reverse.
If they don't agree, then something is wrong.
Al
Reg Edwards wrote in message <8bh1b7$fe2$1...@plutonium.btinternet.com>...
Hamm4fun
>Dear Andreas,
>
>First thing to remember there is no such thing as a genuine SWR meter.
>Beware of adverts.
>
>Such a theoretical meter measures the ratio of the voltages (or
>currents) at two different places along a transmission line spaced
>1/4-wave apart. Unless such a meter has simultaneous access to these
>two different locations along the line then it cannot indicate SWR.
>And as you may already be aware, the common or garden SWR meter has
>access only to one location on the line - the place where it is
>fitted.
>
>The location in the line of the usual SWR meter, ie., immediately at
>the output of the transmitter, actually indicates the SWR on a
>non-existent line - ie., the non-existent line between the meter and
>the transmitter.
>
>How can a line of zero, or even very short length, have standing waves
>?
>--
>Regards, Reg.
>
>
>
>
>
>
>
>
I like to view SWR meters as impedance ratio meters. They indicate the ratio of
the impedance of the line/load at the point they are placed in the line
compared to the design impedance of the meter(usually 50 ohms).
Carl Clawson
powers at a single point in the line. This can be done with directional
couplers, or by independently measuring current and voltage at that point.
I'll write down the math for the latter if anyone wants to see it.
Matt's explanation was just fine. And you can certainly do the measurement
as described below, but it's not the only way. Such an instrument does
exist; it's known as a slotted line. They're used at UHF and microwave
frequencies, but are not practical at HF.
The problem with SWR meters that we can afford is one of accuracy, not that
there is any fundamental difficulty defining SWR in terms of measurements at
one point in a line. I'll agree that if a line is extremely short
(length/diameter not large) then impedance and SWR are ill-defined.
Regard, Carl WS7L
"Reg Edwards" <G4fgq...@btinternet.com> wrote in message
news:8bh1b7$fe2$1...@plutonium.btinternet.com...
Russ
scale on the meter converts to swr.
Hamm4fun wrote:
>
> >
> >Dear Andreas,
> >
> >First thing to remember there is no such thing as a genuine SWR meter.
> >Beware of adverts.
> >
> >Such a theoretical meter measures the ratio of the voltages (or
> >currents) at two different places along a transmission line spaced
> >1/4-wave apart. Unless such a meter has simultaneous access to these
> >two different locations along the line then it cannot indicate SWR.
> >And as you may already be aware, the common or garden SWR meter has
> >access only to one location on the line - the place where it is
> >fitted.
> >
> >The location in the line of the usual SWR meter, ie., immediately at
> >the output of the transmitter, actually indicates the SWR on a
> >non-existent line - ie., the non-existent line between the meter and
> >the transmitter.
> >
> >How can a line of zero, or even very short length, have standing waves
> >?
> >--
> >Regards, Reg.
> >
> >
> >
> >
> >
> >
> >
> >
>
> I like to view SWR meters as impedance ratio meters. They indicate the ratio of
> the impedance of the line/load at the point they are placed in the line
> compared to the design impedance of the meter(usually 50 ohms).
--
*******************************************
* Russell Shaw, B.Eng, M.Eng(Research) *
* Electronics Consultant *
* email: rus...@webaxs.net *
* Australia *
*******************************************
Andreas Aigner
Thank you for your answers. I think I know how an directional coupler works.
The problem is the geometric dimension. At VHF, UHF and so on, it will work
fine. But at Shortwave? Carl wrote anything about 'independently measuring
current and voltage at that point' (see below). I know some SWR-Meters,
where the line goes through an 'transformer'. Are these of this type? My
original question concerned such SWR-Meters. Sorry :-) I hope you know what
I mean. And Carl, if you mean this kind of SWR-Meter, I am interested in the
mathematics. Thank you.
tnx, 73 de Andreas.
Carl Clawson <ccla...@transport.com> schrieb in im Newsbeitrag:
1KbD4.170$6S6....@tw11.nn.bcandid.com...
Reg Edwards
>
> I like to view SWR meters as impedance ratio meters. They indicate
the ratio of
> the impedance of the line/load at the point they are placed in the
line
> compared to the design impedance of the meter(usually 50 ohms).
Sorry to disagree. They do not indicate an impedance ratio. If one
impedance is 50 ohms resistive and the other is 50 ohms reactive then
the indicated ratio will not 1 but infinity.
The common or garden SWR meter actually indicates the unbalance
voltage in a bridge where the standard arm of the bridge is a
resistance of 50 ohms. It so happens the unbalance voltage is
directly proportional to the reflection coefficient with respect to 50
ohms.
The magnitude of the Ref.Coeff can vary between 0 and 1. However the
meter does not indicate the phase angle of the RC which is just as
important as the magnitude. Half the information is discarded !
Nevertheless, a more appropriate name for the instrument is
"Reflection Coefficient Bridge".
---
Reg, G4FGQ
Carl Clawson
"Andreas Aigner" <aigner....@aon.at> wrote in message
news:8bksq3$ba6$1...@news.netway.at...
> Hello HAMs ;-)
>
> Thank you for your answers. I think I know how an directional coupler
works.
> The problem is the geometric dimension. At VHF, UHF and so on, it will
work
> fine. But at Shortwave? Carl wrote anything about 'independently measuring
> current and voltage at that point' (see below). I know some SWR-Meters,
> where the line goes through an 'transformer'. Are these of this type? My
> original question concerned such SWR-Meters. Sorry :-) I hope you know
what
> I mean. And Carl, if you mean this kind of SWR-Meter, I am interested in
the
> mathematics. Thank you.
>
> tnx, 73 de Andreas.
>
The transformer-type meter works by measuring current and voltage
separately. It can indeed be thought of as a kind of directional coupler.
The ARRL Antenna book and some editions of the Handbook have good
explanations and some projects. Sadly, in recent years they've dropped the
simplest version, which used a capacitive pickoff to measure voltage, in
favor of one that uses an additional transformer.
The most common SWR meters that I've seen are variations on the following
construction: A length of coax connects the input and output terminals. The
outer conductor is connected at only one of the terminals. The coax passes
through a toroid which has a winding on it. This serves as a transformer
which produces a signal proportional to the current on the center conductor.
There is no current in the outer conductor, and it serves only as an
electrostatic shield to eliminate capacitive coupling to the transformer
winding. A circuit is then attached to the center conductor to pick off the
voltage. These two signals are fed to a bridge that outputs the sum and
difference of them, which are then used to compute SWR, either digitally or
by feeding to a calibrated meter. It is important to note that the sum and
differences are done at RF, not DC, so the relative phases are preserved.
Here's the theory. It can be done without the complex number arithmetic, but
it's much harder that way!
The general form of current and voltage in a lossless line of impedance Z0
at a distance z along the line and at time t is:
V = VF*exp(j*w*t - j*k*z) + VR*exp(j*w*t + j*k*z)
I = (VF/Z0)*exp(j*w*t - j*k*z) - (VR/Z0)*exp(j*w*t + j*k*z)
Here w (usually an omega in the textbooks) is equal to 2*pi*f where f is the
frequency, and k is 2*pi/wavelength. j is the imaginary unit, and exp is the
exponential function. VF and VR are the forward and reverse voltages. They
are complex numbers. In other words, each has an amplitude and a phase.
This equation tells you just about everything you need to know about
transmission lines. You can derive the Smith chart from it if you like.
To examine SWR, assume that the meter is at z = 0. There is nothing special
about this point. To measure at a different point would just change the
phases of VF and VR in such a way that the SWR would be the same, but z=0
makes the math simpler. Having set z = 0, note that everything is multiplied
by exp(j*w*t). We can drop this common factor from the rest of the
calculations.
The equations now read
V = VF + VR
I = (VF - VR)/Z0
Signals proportional to V and I are produced as described above. The bridge
circuit then produces the sum and difference quantities
VF = (V + Z0*I)/2 and VR = (V - Z0*I)/2
SWR is defined as (|VF| + |VR|)/(|VF| - |VR|). The remainder of the problem
is how to come up with this quantity. Three common ways are
1. A meter where you first read |VF| to set the forward power to a "set"
mark, then flip a switch to read |VR| on a scale calibrated for SWR. These
are hard to use because you constantly reset the forward power. This is very
tedious to do while adjusting an antenna tuner.
2. A cross-needle meter. |VF| goes to one needle and |VR| to the other.
Scales are calibrated as PF and PR, and lines of constant SWR are drawn
which can be read at the crossing point of the needles. These are much
better than #1 but it's still a bit hard to focus on the crossing point of
the needles. Your eye is drawn to whichever needle is moving more.
3. The best is a computing SWR meter. Analog or digital circuits are
arranged to perform the sums, differences, and divisions, and results are
displayed as you please on an analog or digital display. My preference is
the kind that uses two analog meters, one reading net power (PF - PR) and
one reading SWR. These are the two quantities you really care about.
Special consideration must be taken at QRP to avoid errors due to diode
drops. Germanium diodes are often used because of their low voltage drops,
and op-amp circuits that linearize diode response have been used. There may
be other techniques.
73 & Happy Tuning
Carl WS7L
Hamm4fun
of the device. I guess I could have found a better way to say it. The fact that
it does not show a 50 ohm resistive load as a match to a 50 ohm reactive load
is all the better. I am not so certain that it will appear as an infinite SWR
regardless of the distance from the meter to the load(keeping this distance
limited to a couple of wavelengths) This is something I have never tried.
Reg Edwards
must be zero. So the magnitude of the reflection coefficient must be
unity with a phase angle of +/- 90 degrees.
The SWR is a function only of the magnitude of the reflection
coefficient. The phase angle of the reflection coefficient only shifts
the locations of the voltage peaks and troughs up or down the line
relative to the location of the reflection.
SWR = (1 + RC) / (1 - RC).
So at the point of reflection the SWR is infinity.
If the so-called SWR meter does not indicate infinity then it must be
due to imperfections in the mter.
--
Regards, Reg.
Tom Bruhns
homebrew and/or the antenna groups.
A quick answer: they sample the current and the voltage at a point on
the line. The samples retain phase information. They effectively ratio
the two, and the ratio chosen is impedance (resistance) for which the
meter is calibrated. For example, in 50 ohm line, the ratio of voltage
to current when there's only "forward" power will be 50 volts/amp = 50
ohms. If there's only "reverse" power, the ratio (because the current
will be in the opposite direction for a given voltage polarity) will be
-50. The measured difference from the calibration ratio will tell you
the apparent amount of power in the "opposite" direction. You'll find
much more detail in the old postings...I've made a few, and others have
as well.
Some may argue that it's not really an SWR meter. But that's the name
we've given it, and as far as I'm concerned, it's useful for telling me
the SWR on a line whose impedance is nominally the same as the meter's
calibration impedance. Saying it's not really an SWR meter is, to me,
like saying the speedometer in a car is really not a speedometer: it
can be miscalibrated, and you might use it to measure the speed when the
car is up on blocks with the wheels spinning. If you misapply it, it
will give you wrong answers. That shouldn't be news. And some of it
comes back to just what you accept as "SWR" on a line: does the line
have to be long enough that you actually have at least one voltage node
and at least one voltage antinode? Or can you accept a generalized
definition that allows SWR to be defined in terms of forward and
reflected power as measured at a single point? I happen to accept the
generalized defintion.
Cheers,
Tom
Carl Clawson
occurred to me.
I agree with you that an SWR meter measures SWR -- some more accurately than
others. If you want better, look into a network analyzer or a vector
impedance meter. You'll be able to measure SWR with far greater accuracy
than you can with ham equipment. But you won't find them for $100 or $200.
They're more like $20,000 - $30,000 and up. And you can't transmit through
them.
Tom's employer will be happy to sell a very nice one to you.
-- Carl WS7L
"Tom Bruhns" <tom_b...@agilent.com> wrote in message
news:38DFCF9B...@agilent.com...
Ken
<ccla...@transport.com> wrote:
>Great summary, Tom. Funny, the speedometer analogy was the same one that
>occurred to me.
>
<snipped a bit>
OK, how about something a little more basic; how do you use the
'standard' single meter SWR meter? I assume (always dangerous) that
you peak the forward reading , then switch to 'REV" and read the SWR
directly? Having never had anyone actually show me, I'm not sure.
Thanks, & Hi Carl :-)
73, Ken KD7FLE
Milwaukie, OR SOC# 254
Ralph Mowery
There are 2 ways that I do it. If I can change the output power of the
rig I set the swr pot to the most sensitive position. Then set the
transmitter to the lowest power, transmitt a carrier and bring the power
up tuil the meter is at full scale, then switch to the reverse
position. If I can not vary the output in a smooth way, I put the
transmitter on the lowest power I can and then use the pot on the swr
meter to get the meter to full scale, then switch to reverse to read the
swr. The transmitter can be set to full power, but that creats more
'trash" on the band.
Sometimes the swr will appear to change if low power vers high power is
used. There are basically 2 reasons for this. One is the transmitter
is generating power (harmonics) at other than the carrier frequency and
the antenna is not matched for the harmonic. The other reason is the
diodes that is used in the swr bridge are not linear for various power
levels.
Jay OmegaC
>Sometimes the swr will appear to change if low power vers high power is
>used. There are basically 2 reasons for this. One is the transmitter
>is generating power (harmonics) at other than the carrier frequency and
>the antenna is not matched for the harmonic. The other reason is the
>diodes that is used in the swr bridge are not linear for various power
>levels.
>
There's one more reason, of course: the high power causes something in the
load to change. The load could be simply non-linear, or it could suffer from
heating effects.
73,
Jay
JJ OKK
It sounds like you are a very considerate operator. Many times,
I've encountered the signal of someone 'tuning up' on the air with a
heckova lot of power. Out of disgust and curiosity as to who and why,
I'd listen to the entire tune up procedure, just to see who the yokel
was, when he started his "CQ". Maybe it is a coincidence, but in every
instance this happened, the operator began speaking in a Spanish
dialect. Maybe this is the approved method in South American countries.
(???) Guy
Carl Clawson
news:20000409163141...@ng-cu1.aol.com...
Or arcing. This happens easily if you try to run high power through a tuner
to a load that has high SWR. The voltages can become quite high. I have
arced across the phenolic insulator of a PL-259 by doing this. A carbon
trail forms and it will arc more easily after the first time. If you need to
run more than 100W into high SWR, consider teflon-insulated connectors, and
check the working voltage of your feedline. Some, such as RG-8X foam, won't
tolerate much SWR at 1500W.
Cheers
Carl WS7L