Antenna analyzer question
NoSpam...@hotmail.com
antenna. The antenna seemed to be resonating low so I removed a few
inches from each end. Currently the Rs = 50, the Xs = 0, but the SWR =
3.1 at 14.035 mhz. This is up from a 2.2 SWR when I began. Should the
SWR be this high at resonance? Or, is there something else wrong?
Thanks
JGBOYLES
>3.1 at 14.035 mhz. This is up from a 2.2 SWR when I began. Should the
>SWR be this high at resonance? Or, is there something else wrong?
Sounds like the 269 has a problem, the swr should indicate 1.0 for that value
of R and X. You may be getting some local broadcast interference in the
analyzer. If you could run the analyzer to a 50 ohm dummy load and get R=50,
X=0, and swr=1.0 then interference is the problem. If the swr is still high
into a dummy load, the analyzer needs calibration, and it would have to be sent
back to MFJ. Unless you have some good test equipment and a cal. procedure.
If you don't have a dummy load, you could solder a couple of short pigtails
with small alligator clips to a pl-259 connector and connect them to a 50 ohm
non-inductive resistor to check the calibration of the 269. I use this method
to check small capacitors, inductors, and antenna impedance right at the
antenna terminals. Good luck.
73 Gary N4AST
73 Gary N4AST
W6RCecilA
Describe the conditions a little better. Is the SWR reading from the 269?
Describe the feedline? If the 269 is working OK, sounds like common-mode
current problems or RFI.
--
http://www.mindspring.com/~w6rca
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Reg Edwards
The 269 is not accurate enough to show that an antenna is behaving
exactly as theory says it should. But there's probably nothing
wrong with either the meter or antenna.
There are several important things to remember about 1/2-wave
dipole behaviour versus length and frequency.
(1) When the feedline impedance does not match the antenna
feedpoint impedance (it never does) the length of the feed line is
as important as antenna length insofar as the load on the
transmitter is concerned (as indicated by the 50-ohm SWR meter at
the transmitter end).
(2) As frequency is varied, minimum SWR (whether it be good or
bad ) always occurs at the resonant frequency of the antenna PLUS
feed line, ie., of the system as a whole. Trimming the length
either of the line or of the antenna will change the resonant
frequency of the system. But why do you want to change the
resonant frequency ?
(3) When you change the length of line or antenna you not only
change the resonant frequency, ie., the frequency at which minimum
SWR occurs, you also change the radiation resistance and therefore
the antenna feedpoint impedance. And so at another resonant
frequency the minimum SWR will change to another value. It may be
either better or worse than the previous value. In your case it
seems to have got worse. But this is quite normal behaviour. What
actually happens can be influenced by antenna height.
(4) The probability of the feed line input impedance being
precisely 50 ohms at a resonant frequency of the system is so
remote you can forget it. The only way to ensure an SWR of 1-to-1
is to install an antenna tuner at the transmitter end of the feed
line.
To summarise, if the resonant frequency of the dipole is in the
right ballpark, as determined by its length, then leave it that
way. Don't bother about what the SWR on the feedline might be.
Concentrate on adjusting the tuner such that the load on the
transmitter is a nice 50 ohms resistance.
One explanation of your observations is that you may be fortunate
in having a very high antenna with low ground loss. If you lowered
the antenna to increase ground loss you may find the feed line
input resistance is the desired 50 ohms when the resonant frequency
is where you want it to be.
But which do you prefer - an inefficient antenna system with the
resonant frequency on your favourite operating frequency - or an
efficient system which doesn't care what its resonant frequency is.
But don't forget the tuner.
----
***********************************
Regards, Reg, G4FGQ
Free radio design & modelling software
http://www.btinternet.com/~g4fgq.regp
***********************************
Rick Littlefield
>Currently the Rs = 50, the Xs = 0, but the SWR = 3.1 at 14.035 mhz.<
Something is clearly wrong. The resolution of the meter can be a little rough
because of limited processor bit capacity, but it sure isn't THAT bad. One
responder gave you the best advice:
1. Check the meter with a 50-Ohm resistive load. If you see more that 1.2:1
VSWR on a 47 - 51 ohm 1/4-watt resistor at 14 MHz, the meter is likely out of
calibration. To check a 2:1 point, try a 100-ohm resistor. If it's off,
contact MFJ by phone, speak to a real person, and insist on a replacement. If
the SWR is low, your meter's probably fine.
2. Check the condition of the coax by terminiating it at the far end with a
50-ohm load and re-measuring. If it's good with the load connected directly to
the analyzer--but off-the-wall with the coax connected to the load--you're coax
may be bad.
3. If the condition only occurs when the antenna's connected, it may be RF
getting into the analyzer bridge from a local high-power transmitter (any
powerful broadcasters in your neighborhood?). If that turns out to be the
case, you may need to look into getting a MFJ-731 tunable filter to reject the
unwanted RF.
4. Common mode current on your feedline could cause a problem if the feed is
very nearly 1/4-wave or 3/4-wave long (set the meter on an insulated surface,
and if the SW changes when you touch the case, you've got common mode). To
cure that, install a balun, or coil about 5 turns of cable up at the feedpoint
and tape them together to form a choke.
Hope this helps--
73
Rick K1BQT
NoSpam...@hotmail.com
feedline is 36 feet. The balun is coiled coax per the ARRL handbook.
I seem to be getting out FB with the antenna, so maybe it's better than the
analyzer indicates!
73
JGBOYLES
>analyzer indicates!
Yep, the analyzer will not tell you much about how well you will get out, and
with a 3.0 swr you can certainly make a lot of contacts with good signal
reports. The analyzer will tell you all sorts of info about your antenna
system that you can use to make your antenna better, and to check out that new
design to see if it's the way you want.
If the MFJ is telling you the swr is 3.0 and R=50 ohms and X=0, then
something is wrong. Paying the $250 for the 269 I would want to know what was
wrong. If you are going to cut a dipole to formula and string it up and start
transmitting, why bother checking with a very useful (if working right)
instrument?
I have the 259b and it is the single most used piece of test equipment in the
shack. If it told me what your 269 did, I would want to know the reason.
But, the name of the game is getting out, and the analyzer can help, or it
can be of no help at all
73 Gary N4AST
Dan Richardson
The analyzer is telling you the conditions on the feed line. It says
little about the antenna ifself.
That would be something like checking the air pressure in an
automobile's tires to determine how fast it will go.
Danny, K6MHE
Bill Nelson
: I'm using a MFJ antenna analyzer (model 269) to adjust a wire dipole
: antenna. The antenna seemed to be resonating low so I removed a few
: inches from each end. Currently the Rs = 50, the Xs = 0, but the SWR =
: 3.1 at 14.035 mhz. This is up from a 2.2 SWR when I began. Should the
: SWR be this high at resonance? Or, is there something else wrong?
In the first place, the impedance of your dipole at resonance should be
closer to 70 ohms, not 50.
How are you determining resonance? Are you assuming that it is resonant
when the Rs reads 50 ohms? If so, then you are doing it wrong. The
antenna is resonant when Rs is the lowest. Yes Xs should also be 0 ohms.
--
Bill Nelson (bi...@peak.org)
George, W5YR
72/73, George W5YR - the Yellow Rose of Texas QRP-L 1373 NETXQRP 6
Fairview, TX 30 mi NE of Dallas in Collin county EM13qe
Amateur Radio W5YR, in the 55th year and it just keeps getting better!
Icom IC-756PRO #02121 Kachina #91900556 IC-765 #02437
Tom W8JI
Resonance is ALWAYS where X=0
The impedance at the center of a dipole is almost always NOT 70 ohms,
it is usually lower.
It sounds like the meter is miscalibrated or defective. It can NOT be
common mode current. It CAN be a really close by transmitter causing
a false reading.
73 Tom
Reg Edwards
minimum value whatever the actual value of the SWR might be.
Peter O. Brackett
The frequency of the minimum value of VWSR is a good indicator of resonance,
however it is a more complicated process for the meter to come up with the
VSWR reading... just in case something is wrong with the meter...
Wouldn't a "safer" indication of resonance, less subject to errors and fauty
meter operation, be that the antenna plus feedline is resonant when the
measured reactance equals zero. Xs = 0. Of course the resistance will be
whatever it will be at resonance, but the reactance has to be zero.
Just trying to be helpful.
-Peter K1PO
"Reg Edwards" <g4fgq...@btinternet.com> wrote in message
news:9bep3u$9h4$1...@plutonium.btinternet.com...
W6RCecilA
> The antenna is resonant when Rs is the lowest.
Is 0.000001 - j10000 resonant? :-)
W6RCecilA
> It sounds like the meter is miscalibrated or defective. It can NOT be
> common mode current. It CAN be a really close by transmitter causing
> a false reading.
Doesn't a really close by transmitter cause common-mode currents?
John O'Flaherty
> Bill Nelson wrote:
> > The antenna is resonant when Rs is the lowest.
>
> Is 0.000001 - j10000 resonant? :-)
In short, no. :-)
john
Tom W8JI
wrote:
>Tom W8JI wrote:
>> It sounds like the meter is miscalibrated or defective. It can NOT be
>> common mode current. It CAN be a really close by transmitter causing
>> a false reading.
>
>Doesn't a really close by transmitter cause common-mode currents?
Quit trolling. It confuses people trying to actually understand the
problem.
The meter does NOT respond to common mode current, as long as the
shield is connected to the jack on the meter, because it is a
self-contained system inside a closed metal box. It can only respond
to differential mode currents on the two input terminals.
73 Tom
Bill Nelson
I wrote.
: In the first place, the impedance of your dipole at resonance should be
: closer to 70 ohms, not 50.
: How are you determining resonance? Are you assuming that it is resonant
: when the Rs reads 50 ohms? If so, then you are doing it wrong. The
: antenna is resonant when Rs is the lowest. Yes Xs should also be 0 ohms.
Oops, that was a brain fart.
I should have stated that the antenna is resonant when the impedance is
lowest.
My apologies for the confusion this may have caused.
--
Bill Nelson (bi...@peak.org)
Bill Nelson
: W6RCecilA wrote:
: In short, no. :-)
Yeah, that was a brain fart, as I noted elsewhere. That should have
been "when the antenna input impedance is lowest".
--
Bill Nelson (bi...@peak.org)
W6RCecilA
> The meter does NOT respond to common mode current, as long as the
> shield is connected to the jack on the meter, because it is a
> self-contained system inside a closed metal box. It can only respond
> to differential mode currents on the two input terminals.
Why does the reading change when touched by a human hand?
Tom W8JI
wrote:
>Tom W8JI wrote:
>> The meter does NOT respond to common mode current, as long as the
>> shield is connected to the jack on the meter, because it is a
>> self-contained system inside a closed metal box. It can only respond
>> to differential mode currents on the two input terminals.
>
>Why does the reading change when touched by a human hand?
Because you are changing the actual SWR by detuning the antenna
system.
The meter is non-responsive to common mode currents when the shield is
connected, it is the antenna SYSTEM that is actually changing when you
touch the meter case. The meter reads only the differential mode
impedance, and that is what you changed.
73 Tom
Tom W8JI
>Oops, that was a brain fart.
>
>I should have stated that the antenna is resonant when the impedance is
>lowest.
>
>My apologies for the confusion this may have caused.
The system is resonant when the reactance is zero.
The antenna is resonant when the reactance at its terminals is zero.
Once you measure through a feedline, you can not tell if the antenna
is resonant or not without knowing something about the feedline. A
zero ohm reactance (resonant) antenna might not present a zero ohm
reactance load one a feedline is involved, and the lowest impedance
might not be the resonant point of the antenna.
Fortunately, resonance has little to do with system efficiency.
73 Tom
Rick Littlefield
>Because you are changing the actual SWR by detuning the antenna
system.<
Tom is 1000% correct on this point, and I apologize if my reponse implied
anything different. Under the condition I described, the metal exterior
surface of the meter's case is at a voltage point on the "third leg" of the
antenna (the outside of the shield). It requires only minimal capacitive
loading at this point to alter the system's overall tuning.
Believe me, either the meter's way out of cal, or a powerful local signal is
being picked up by the antenna and "fooling" the meter's bridge circuit.
Rick K1BQT
W6RCecilA
I guess that's one way to look at it. But when I can flood the meter
chassis/circuit-common with common-mode currents by touching it, I am
not going to trust the meter readings.
I thought it was not possible to "change the actual SWR" by doing
something at the transmitter. If I touch the SWR meter chassis and
the SWR reading decreases, are you saying that the actual SWR has
decreased?
Wes Stewart
[snip]
>I guess that's one way to look at it. But when I can flood the meter
>chassis/circuit-common with common-mode currents by touching it, I am
>not going to trust the meter readings.
>
>I thought it was not possible to "change the actual SWR" by doing
>something at the transmitter. If I touch the SWR meter chassis and
>the SWR reading decreases, are you saying that the actual SWR has
>decreased?
If the "transmitter" is also part of the "antenna" as it is with common-mode
current on the transmission line, then yes if touching it lowers the SWR, I
would expect the meter to read lower too.
Wes Stewart N7WS
W6RCecilA
> If the "transmitter" is also part of the "antenna" as it is with common-mode
> current on the transmission line, then yes if touching it lowers the SWR, I
> would expect the meter to read lower too.
Wow! A whole new way of matching. If the coax balances the differential
current as it is supposed to do, exactly how does it change the ratio
between the 100 ohm load and the 50 ohm Z0?
Tom W8JI
wrote:
>I thought it was not possible to "change the actual SWR" by doing
>something at the transmitter. If I touch the SWR meter chassis and
>the SWR reading decreases, are you saying that the actual SWR has
>decreased?
>--
>http://www.mindspring.com/~w6rca
Absolutely, you have changed the actual SWR.
Very few SWR meters respond directly to common mode currents. They
would have to be a horrible design to do that....and would be
unreliable in any system. When you see the meter change, it is either
because the null of the meter is not calibrated to the surge impedance
of the cable being used or because the actual SWR in the line is
changing.
If you touch the case of the 259 connected properly to a shielded
cable, and the reading changes, the reason is the outside of the
feedline is part of the antenna and radiating. You have, in effect,
changed the antenna to feedline SWR by tuning the antenna system at
the meter end of the feedline.
The meter indicates the true SWR at that point. Of course if you
remove the meter and install a transmitter, the SWR will likely change
again, because you perturb the common mode path from the antenna to
the feedline.
The cure is to not allow common mode currents to flow on the outside
of the shield away from the antenna. Then what you do to the feeder
won't change the SWR at the antenna end of the feedline.
73 Tom
W6RCecilA
> Very few SWR meters respond directly to common mode currents.
The case of SWR meters are the circuit common for the DC electronics.
If the SWR meter case has RF on it, seems to me that can throw all the
DC measurements off especially measurements done by a microcontroller.
At Intel, I saw environments where RF rendered all the A/D conversion
measurements invalid.
> If you touch the case of the 259 connected properly to a shielded
> cable, and the reading changes, the reason is the outside of the
> feedline is part of the antenna and radiating. You have, in effect,
> changed the antenna to feedline SWR by tuning the antenna system at
> the meter end of the feedline.
Please define that SWR. It's obviously not a 100 ohm load divided by
a 50 ohm Z0 to get 2:1.
Richard Clark
wrote:
>Tom W8JI wrote:
>> Very few SWR meters respond directly to common mode currents.
>
>The case of SWR meters are the circuit common for the DC electronics.
>If the SWR meter case has RF on it, seems to me that can throw all the
>DC measurements off especially measurements done by a microcontroller.
>At Intel, I saw environments where RF rendered all the A/D conversion
>measurements invalid.
>
>> If you touch the case of the 259 connected properly to a shielded
>> cable, and the reading changes, the reason is the outside of the
>> feedline is part of the antenna and radiating. You have, in effect,
>> changed the antenna to feedline SWR by tuning the antenna system at
>> the meter end of the feedline.
>
>Please define that SWR. It's obviously not a 100 ohm load divided by
>a 50 ohm Z0 to get 2:1.
Hi Cecil,
This is another case of shield effectiveness. The outside of the SWR
meter case is not the same electrical point as the inside of the SWR
meter case. The discussion of DC circuitry is immaterial to the
measured response. The SWR circuitry can be effectively isolated from
all grounds and still provide similar service readings already
described in this thread.
Cecil, there is no need for an explicit value of load. Quantification
is not necessary to prove the problem of Common Mode artifacts being
demonstrated by your hand compounding that problem as evidenced by the
SWR shift.
A microcontroller may have problems with RF, but that is simply an
issue of design shortfall. I have done A/D conversion at speeds up to
25 MHz - it would be extremely hard to differentiate the 25 MHz as a
clock or RF if we casually blamed RF as a source of all controller's
problems (they too run in the HF regions, if not VHF/UHF and soon
SHF). I could regularly measure nanowatt signal levels without
interference from the clock driving either the A/D or the
microcontroller. My designs exhibited very low noise levels (quite
close to the cosmic noise level in fact) as I measured those nanowatt
signals to an accuracy of 0.5% or better.
73's
Richard Clark, KB7QHC
Richard Clark
wrote:
>Wes Stewart wrote:
>> If the "transmitter" is also part of the "antenna" as it is with common-mode
>> current on the transmission line, then yes if touching it lowers the SWR, I
>> would expect the meter to read lower too.
>
>Wow! A whole new way of matching. If the coax balances the differential
>current as it is supposed to do, exactly how does it change the ratio
>between the 100 ohm load and the 50 ohm Z0?
Hi Cecil,
No, it is the old fashion way - you change the load, SWR follows. If
you choose to become part of the load by interfering with Common Mode
artifacts with your hand....
The human body presents about 10 pF worth towards added reactance and
a variable resistance based upon if you are telling the truth or not.
An outrageous liar would probably come close to 50 Ohms (a suitable
load for the electric chair). ;-)
Richard Clark
wrote:
>The SWR circuitry can be effectively isolated from
>all grounds and still provide similar service readings already
>described in this thread.
Hi All,
On reflection (no pun intended) this may have been an overstatement
insofar as ground isolation.
W6RCecilA
> A microcontroller may have problems with RF, but that is simply an
> issue of design shortfall.
Does the MFJ259B overcome every design shortfall? I have seen completely
unbelievable readings from mine, e.g. 70 ohms, zero reactance, and the
SWR is 4:1.
Wes Stewart
>Wes Stewart wrote:
>> If the "transmitter" is also part of the "antenna" as it is with common-mode
>> current on the transmission line, then yes if touching it lowers the SWR, I
>> would expect the meter to read lower too.
>
>Wow! A whole new way of matching. If the coax balances the differential
>current as it is supposed to do, exactly how does it change the ratio
>between the 100 ohm load and the 50 ohm Z0?
I don't know where this 100 ohm load came from. Some of these threads kinda
make my eyes glaze over and I don't follow along.
You said:
>"I thought it was not possible to "change the actual SWR" by doing
>something at the transmitter. If I touch the SWR meter chassis and
>the SWR reading decreases, are you saying that the actual SWR has
>decreased?"
If, I repeat, if, there is common-mode current on the outside of the coax (for
whatever reason) then the outside of the coax is part of the radiator. When you
start touching it, you are changing the characteistics of the radiator. So why
wouldn't the indicated SWR change?
Wes N7WS
>--
>http://www.mindspring.com/~w6rca
>
W6RCecilA
> If, I repeat, if, there is common-mode current on the outside of the coax (for
> whatever reason) then the outside of the coax is part of the radiator. When you
> start touching it, you are changing the characteistics of the radiator. So why
> wouldn't the indicated SWR change?
Oh, the indicated SWR would certainly change. Question is, is it valid?
Seems to me, if the assumptions of the SWR/wattmeter designer are no
longer valid, the indication may no longer be valid. Shirley, you
realize that RF can invalidate an A/D conversion result?
Richard Clark
wrote:
>Richard Clark wrote:
>> A microcontroller may have problems with RF, but that is simply an
>> issue of design shortfall.
>
>Does the MFJ259B overcome every design shortfall? I have seen completely
>unbelievable readings from mine, e.g. 70 ohms, zero reactance, and the
>SWR is 4:1.
Hi Cecil,
Does the MFJ overcome every design shortfall? Does it overcome any?
I use my trusty General Radio bridge. No active components, no power
supply requirements, no processors - just unequivocal results.
It is probably 1000 times more difficult to use the first (second,
third, fourth...) time than your MFJ, but once you get into the method
of determination, you will not be handed a bogus answer. However,
much the same could be said of the MFJ. That is, once you get into
the method of determination....
It is arguable that even a loop with rectifier feeding a magnetic
compass could be made to work to determine RF characteristics.
However this is not a particularly suitable choice - just an option.
The problems you describe are more likely due to software and a
general lack of data bandwidth or A/D resolution (almost certainly
this last if not both). These sources of problems bear no relation to
your presumptions of RF getting into the processor. This is a
distinct possibility, but there are other more common issues to visit
first.
I designed the flight data recorder using a military version of the
8080; BUT I also had a 12 bit A/D converter hanging off it measuring
600 analog lines coming from every point in the airframe. So, speed
of the processor and the nature of what is being measured are no less
complex, but the added data bandwidth allowed me to meet FAA
regulations as to absolute accuracy requirements. There are many
considerations beyond hardware, and software and data bandwidth are
the principle first points of failure.
Tom W8JI
wrote:
>The case of SWR meters are the circuit common for the DC electronics.
>If the SWR meter case has RF on it, seems to me that can throw all the
>DC measurements off especially measurements done by a microcontroller.
>At Intel, I saw environments where RF rendered all the A/D conversion
>measurements invalid.
The stuff you saw at Intel is all very nice Cecil, but meaningless for
devices designed to work in RF environments.
Anyone who sold a meter that was affected as you describe would
quickly have them all back from customers.
I've never seen an SWR meter or power meter that responded to common
mode currents, although I imagine some sloppy or unbypassed layout
somewhere might behave that way.
>> If you touch the case of the 259 connected properly to a shielded
>> cable, and the reading changes, the reason is the outside of the
>> feedline is part of the antenna and radiating. You have, in effect,
>> changed the antenna to feedline SWR by tuning the antenna system at
>> the meter end of the feedline.
>
>Please define that SWR. It's obviously not a 100 ohm load divided by
>a 50 ohm Z0 to get 2:1.
I don't follow your question.
73 Tom
Brian Kelly, w3rv
. . . somebody had to get back to the basics . .
>
>73 Tom
>
w3rv
Gene Mason
Could you elaborate on your 'interesting' last paragraph ?
Gene KZ5V
P.S. You might begin with an explanation of the association of FAA
regulations and 'absolute accuracy'.
Was the accuracy .001, 1, or 30 % ?
Tell us about the 'low pass filters' used between the sensors and the ADC.
Tell us how the 12 ADC is better than a 1 bit ADC.
for starters, thanks.
"Richard Clark" <kb7...@home.com> wrote in message
news:3adcb062....@news.seanet.com...
Gene Mason
May I ask a dumb question (s)?
Was the impedance measurement made at the antenna feedpoint, or at the end
of the coax transmission line? And how long was the transmission line, and
what was the frequency? <grin>
Gene KZ5V
P.S. How can so much talent miss the obvious?
"Reg Edwards" <g4fgq...@btinternet.com> wrote in message
news:9bc8mk$4mm$1...@plutonium.btinternet.com...
> > I'm using a MFJ antenna analyzer (model 269) to adjust a wire
> dipole
> > antenna. The antenna seemed to be resonating low so I removed a
> few
> > inches from each end. Currently the Rs = 50, the Xs = 0, but
> the SWR =
> > 3.1 at 14.035 mhz. This is up from a 2.2 SWR when I began.
> Should the
> > SWR be this high at resonance? Or, is there something else wrong?
> ================================
>
> The 269 is not accurate enough to show that an antenna is behaving
> exactly as theory says it should. But there's probably nothing
> wrong with either the meter or antenna.
>
> There are several important things to remember about 1/2-wave
> dipole behaviour versus length and frequency.
>
> (1) When the feedline impedance does not match the antenna
> feedpoint impedance (it never does) the length of the feed line is
> as important as antenna length insofar as the load on the
> transmitter is concerned (as indicated by the 50-ohm SWR meter at
> the transmitter end).
>
> (2) As frequency is varied, minimum SWR (whether it be good or
> bad ) always occurs at the resonant frequency of the antenna PLUS
> feed line, ie., of the system as a whole. Trimming the length
> either of the line or of the antenna will change the resonant
> frequency of the system. But why do you want to change the
> resonant frequency ?
>
> (3) When you change the length of line or antenna you not only
> change the resonant frequency, ie., the frequency at which minimum
> SWR occurs, you also change the radiation resistance and therefore
> the antenna feedpoint impedance. And so at another resonant
> frequency the minimum SWR will change to another value. It may be
> either better or worse than the previous value. In your case it
> seems to have got worse. But this is quite normal behaviour. What
> actually happens can be influenced by antenna height.
>
> (4) The probability of the feed line input impedance being
> precisely 50 ohms at a resonant frequency of the system is so
> remote you can forget it. The only way to ensure an SWR of 1-to-1
> is to install an antenna tuner at the transmitter end of the feed
> line.
>
> To summarise, if the resonant frequency of the dipole is in the
> right ballpark, as determined by its length, then leave it that
> way. Don't bother about what the SWR on the feedline might be.
> Concentrate on adjusting the tuner such that the load on the
> transmitter is a nice 50 ohms resistance.
>
> One explanation of your observations is that you may be fortunate
> in having a very high antenna with low ground loss. If you lowered
> the antenna to increase ground loss you may find the feed line
> input resistance is the desired 50 ohms when the resonant frequency
> is where you want it to be.
>
> But which do you prefer - an inefficient antenna system with the
> resonant frequency on your favourite operating frequency - or an
> efficient system which doesn't care what its resonant frequency is.
> But don't forget the tuner.
> ----
> ***********************************
> Regards, Reg, G4FGQ
> Free radio design & modelling software
> http://www.btinternet.com/~g4fgq.regp
>
> ***********************************
>
>
>
Richard Clark
wrote:
>Richard,
> Could you elaborate on your 'interesting' last paragraph ?
>Gene KZ5V
>P.S. You might begin with an explanation of the association of FAA
>regulations and 'absolute accuracy'.
>Was the accuracy .001, 1, or 30 % ?
>Tell us about the 'low pass filters' used between the sensors and the ADC.
>Tell us how the 12 ADC is better than a 1 bit ADC.
>for starters, thanks.
>
Hi Gene,
Hmm, that's a tall order. A 1 bit ADC is simply an analog comparator
and error could be as high as 100% (you measure no signal, and yet it
exists just below the trip point). A 12 bit ADC has a resolution of 1
part in 4000 so that your error is no higher than 0.025% (with respect
to the ADC reference) if the signal is just below the trip point of
the internal comparator. So, for this example, the best accuracy
would be 0.025% for a perfect reference (no such thing). I used a
precision Voltage source that guaranteed a reference voltage that met
this (probably good to 0.01% or so). This was my "absolute" standard
(I could count on it never being out of that range of error). This is
distinct from "relative" accuracy which is based on a comparison where
you have no concern with what the value is, merely how it correlates
to another value(akin to a dB relationship).
But, this is still not the reported accuracy because the circuit
contributes error (there are many sources of error when resolutions of
voltage begin to measure in the mV). As I've pointed out in other
threads, the level of discussion in shields and Common Mode problems
are truly amateur. There are also errors due to conversions of AC
potentials. I had to monitor all legs of Servo's feeding my design
and compute the angle on the basis of phase differences (this in an
8080 remember). I would compute the angle of the control surfaces
(flaps, rudder, and all), throttle, and so on, all being fed from the
positioners on the airframe. I also had to compute the thrust from
the engines by measuring their temperature (yet another set of
sensors) and I had to have a stiff and robust current source (a
Howland circuit - if I recall accurately). Absolutely every line into
the ADC came through several MUX switches that all exhibited some path
loss (switches have resistance) AND stray capacitance that would
present a lag. On the other side of the switches was a sample and
hold that also contained a capacitor that held that voltage to be
converted within a time frame as it too would exhibit its own decay
from the measured signal (it takes time to convert the input to an ADC
and during that time, the input is slowly draining away). Even at a
conversion rate of around 5 uS, the slope of decay ate away at
accuracy.
One point about my "absolute" reference voltage. This was a four wire
device (by my design) such that I could recover from the card edge
connection of this voltage not being depressed by even the smallest
contact resistance. For those unfamiliar with what I call four wire,
that is simply the supply (two wires) being monitored (by two wires)
such that the system always drives the monitored point to the
reference potential. You would have to find a laboratory grade power
supply to find this, or many open frames supplies do this.
All of these sources of error contribute towards what is called an
error budget. The FAA mandates that this budget not exceed say 1% for
flaps, 0.5% for temperature and so on in that manner. And worse, this
is over all temperatures from -55 to 200 degrees C. And more worse,
over accelerations up to 5 G's in all axis (yes, I measured
acceleration in those three axis too). And it must survive impact and
up to half an hour of 2000 degrees of flame. And, if under water, the
water cannot intrude (and it must turn on a transponder).
The generation of Black Box that I designed was for 64 parameters
(many identified above) all in one second to a continuously running
recorder. I had to not only perform the conversions, I had to place
the data on the bus, and then read it back before I toggled the record
trigger. Then the recorder would put it on tape, back up and read it
back to me for failure testing. (Again, every second, of every hour,
for 24 hours before the tape overwrote itself.) And this done in an 8
bit 8080.
As a caution, do not attempt this at home.
As a warning, the numbers above are culled from the memory of the
design now some 18 years distant. You will note that it is a long way
from the accuracy of my 0.01% voltage reference to the error budget of
1% for flaps - this is the accumulation of error for every
consideration (and there were a ton of issues to consider).
If you have ever ridden in a 757 or 767, there is about a 50% chance
that my design was listening to the airframe you were sitting in.
Gene Mason
Thanks for the prompt and detailed response.
If I might , let me set the record straight, the 12 bit ADC is just 12, 1bit
ADC's as it has the same accuracy or more , since it uses one stage and the
errors are better controlled.a ramp from a precision voltage source would do
the trick.
You indicated resolution of the 12 bit ADC has something to do with the
accuracy. It is a common error to confuse resolution and accuracy, but that
doesnt make it right.
I question some measurement methods, you mentioned, e.g. phase
measurements. Why didnt you use a PSD, phase sensitive demodulator? The
output is a single dc voltage.
You didnt mention whether the sensor low pass filters were implemented,
and if they were active or passive.
Also, did you stagger the scan position on the more accurate
measurements?
Your project sounds very interesting, and you appear proud of your
accomplishments, as well you should. I would recommend that you write a book
about the project. In In your era, most air data computers were analogue.
A new digital system is no small accomplishment, lots of problems to
resolve. My point in summary, measurement accuracies of 1 % are usually
routine, accuracies beyond that are not trivial. Thanks for taking the
time for the explanations.
"Richard Clark" <kb7...@home.com> wrote in message
news:3ae145c2.2049509969@news.seanet.com...
Reg Edwards
helpful.
But it is a mathematical, academic, theoreical, undeniable fact,
that the antenna plus feedline is resonant and will register
minimum SWR versus frequency, whatever the minimum SWR may be, when
the antenna plus feedline is resonant, ie., the feedpoint
resistance is a pure resistance.
--
***********************************
Regards, Reg, G4FGQ
Free radio design & modelling software
http://www.btinternet.com/~g4fgq.regp
***********************************
Peter O. Brackett <ab...@ix.netcom.com> wrote in message
news:9besr7$jit$1...@slb0.atl.mindspring.net...
> Reg:
>
> The frequency of the minimum value of VWSR is a good indicator of
resonance,
> however it is a more complicated process for the meter to come up
with the
> VSWR reading... just in case something is wrong with the meter...
>
> Wouldn't a "safer" indication of resonance, less subject to
errors and fauty
> meter operation, be that the antenna plus feedline is resonant
when the
> measured reactance equals zero. Xs = 0. Of course the
resistance will be
> whatever it will be at resonance, but the reactance has to be
zero.
>
> Just trying to be helpful.
>
> -Peter K1PO
>
>
> "Reg Edwards" <g4fgq...@btinternet.com> wrote in message
> news:9bep3u$9h4$1...@plutonium.btinternet.com...
> > The antenna *PLUS* feedline is resonant when the SWR is at its
> > minimum value whatever the actual value of the SWR might be.
Peter O. Brackett
Agreed!
But... only if the "reference impedance" in one's "reflected voltage meter"
is also a resistance! [Usually the case.]
-Peter K1PO
"Reg Edwards" <g4fgq...@btinternet.com> wrote in message
news:9c215t$7fi$1...@plutonium.btinternet.com...
Reg Edwards
> But... only if the "reference impedance" in one's "reflected
voltage meter"
> is also a resistance! [Usually the case.]
The poor quality of the usual RC meter is due to the effective
reference impedance being anything BUT purely resistive.
Such wierd territory is entered at your peril. Beware of man-traps
such as reflection coefficients greater than unity and negative
values of SWR.
---
Reg.
Wes Stewart
wrote:
[snip]
>
>But it is a mathematical, academic, theoreical, undeniable fact,
>that the antenna plus feedline is resonant and will register
>minimum SWR versus frequency, whatever the minimum SWR may be, when
>the antenna plus feedline is resonant, ie., the feedpoint
>resistance is a pure resistance.
Gad Reg. The Queen's English is more twisted than George Bush's.
"The antenna plus feedline is resonant when the antenna plus feedline is
resonant and the feedpoint resistance is a resistance?"
Are you absolutely sure? <g>
Wes
>
VE9SRB
> Peter, believe me, I do understand you are are trying to be
> helpful.
>
> that the antenna plus feedline is resonant and will register
> minimum SWR versus frequency, whatever the minimum SWR may be, when
> the antenna plus feedline is resonant, ie., the feedpoint
> resistance is a pure resistance.
>
If you are trying to imply that all antenna's exhibit minimum SWR when they
are resonant, this is not correct. The feedpoint impedance being purely
resistive does not guarantee that it also has minimum SWR. Its a function
of Zo and the antenna R at resonance. I can send measured or computed data
on several antennas proving this. I would really like to know who comes up
with all this stuff.
Steve VE9SRB
W6RCecilA
> I would really like to know who comes up with all this stuff.
People who assume conclusions about subsets also apply to supersets.
Tom W8JI
>
>If you are trying to imply that all antenna's exhibit minimum SWR when they
>are resonant, this is not correct. The feedpoint impedance being purely
>resistive does not guarantee that it also has minimum SWR. Its a function
>of Zo and the antenna R at resonance. I can send measured or computed data
>on several antennas proving this. I would really like to know who comes up
>with all this stuff.
Also, just because reactance is zero on the feedline at some point
along the line, it does not mean the antenna is resonant....
You really need to measure at the antenna, unless you know the line
characteristics well and can measure complex impedance accurately.
Lowest SWR or zero reactance down the feedline someplace does not mean
the antenna is resonant!
73 Tom
Reg Edwards
If nothing is done except vary frequency while observing the SWR
meter, when the indicated SWR is at its minimum the system is at
resonance whatever the minimum value might be.
---
Reg.
Tom W8JI
That's true Reg if you talk about the SYSTEM, and not of the antenna,
and if the system passes a point where reactance is zero and
resistance at that point is not shifting rapidly with frequency.
Adding any amount of reactance increases SWR, but it certainly does
not mean the antenna is resonant when the impedance measured elsewhere
has no reactance!
73 Tom
VE9SRB
This is not necessarily true at all.
A minimum SWR does not imply "resonance". Let's just talk about the point
where the measurement is made. Considering impedance, resonance is a
condition where the reactance is zero and the impedance only has a resistive
component. Over a given range of frequencies, the impedance may never go
through resonance but you will still measure a minimum SWR.
This resonance - minimum SWR concept, comes from the fact that for a given
impedance, Z = R + jX, minimum SWR for this value of R will occur when X =
0. The important point is that this minimum SWR is only for this R. This
concept cannot be extended to imply that the minimum SWR of an antenna or a
system will occur when the system is resonant. Many antennas and systems
will have their minimum SWR occur at a frequency where the antenna or the
system is not resonant.
Steve VE9SRB
VE9SRB
"Tom W8JI" <2w...@akorn.net> wrote in message
news:3ae6d7cc...@news.accessunited.com...
> On Wed, 25 Apr 2001 03:07:19 +0100, "Reg Edwards"
> <g4fgq...@btinternet.com> wrote:
>
> >Yes Wes, I made a mess of it. I'll try again.
> >
> >If nothing is done except vary frequency while observing the SWR
> >meter, when the indicated SWR is at its minimum the system is at
> >resonance whatever the minimum value might be.
> >---
> >Reg.
>
> That's true Reg if you talk about the SYSTEM, and not of the antenna,
> and if the system passes a point where reactance is zero and
> resistance at that point is not shifting rapidly with frequency.
>
Measuring a minimum SWR anywhere in the system does not mean the system is
resonant at that point. It only means that the system has minimum SWR.
Many antennas and systems will have their minimum SWR occur at a frequency
other than their resonant frequency.
It is true that for any given impedance R + jX, that minimum SWR occurs when
X = 0 (R being fixed). This is a different concept and issue than whether
or not a system will have minimum SWR at resonance.
Steve
Reg Edwards
> >If nothing is done except vary frequency while observing the SWR
> >meter, when the indicated SWR is at its minimum the system is at
> >resonance whatever the minimum value might be.
Tom wrote -
> That's true Reg if you talk about the SYSTEM, and not of the
antenna,
> and if the system passes a point where reactance is zero and
> resistance at that point is not shifting rapidly with frequency.
Yes, when the SWR meter is next to the transmitter I carefully
specify it is the system I am talking about.
I made the contribution because some people appear to be under the
impression the system is resonant only at the frequency where the
SWR is 1-to-1.
Such erroneous ideas arise from the other misconception that an SWR
meter at the transmitter end of the antenna feedline indicates the
SWR on that feedline. In fact it does not indicate an SWR anywhere.
It does not even respond to an SWR.
The magazine authers, technical editors, meter manufacturers and
salesmen have been fooling us hams for the best part of a century.
And Tom, from your remark that the resistive component of input
impedance should not be changing very rapidly with frequency as the
frequency passes through resonance indicates you understand very
well what's going on.
I agree, my statement is an approximation.
If we take as an example the feedpoint at the centre of a 1/2-wave
dipole, at which point the Q of the antenna is roughly 10, the
reactance changes very rapidly through resonance. The resistive
component of feedpoint impedance is, of course, the antenna
radiation resistance which changes versus frequency relatively
slowly and the approximation is good enough.
----
Reg.