SWR on vertical goes out of whack... why?
Q
SWR problems, and I haven't been able to figure out why. I'm hoping
someone might have an idea.
The vertical is real simple: 1" Al tubing, held up by polypropylene
rope. Where it meets the roof (the roof is galvanized steel sheet)
it's electrically insulated via a 1/2" PVC pipe with a 1/2" PVC union
to give it a rim to hold up the antenna.
On the lower part of the union, I placed a steel hose clamp from which
3 1/4 wave radials extend. From there, they run on top of the roof at
the roof angle (15%).
What happens is this: no rain, the antenna adjusts to around 1.7 SWR.
It works without any problems. Add a 5 or 10 minute drizzle, and the
SWR goes up to 2.0, which basically makes the transmitter autoprotect.
At the present I have a couple of theories:
1. Water is getting into the coax. However it doesn't seem likely
given that there's a "hood" (a cut off soda bottle) on top of the
whole coax assembly, as well as the typical inverted-U configuration
to keep water out via gravity.
2. The water is changing the "ground" conductivity beneath the
antenna. However it doesn't seem likely that a wet steel roof will
change very much with respect to a dry one. Besides that, on one
occasion I went up and found that the SWR was reading 2.0 and the roof
was visibly dry.
3. The pplene rope is becoming conductive. However, once again I
checked and found no noticeable moisture on the rope, and the SWR was
still reading 2.0
Aside from that the only thing that occurs to me is that the water is
somehow partially shorting the vertical radiator to the radials. But
once again, I'm not sure how much of a problem a film of water on top
of the PVC along 1 or 2 cm length might represent.
Any ideas?
---
Mario - TI2DLL
Wes Stewart
Is the PVC pipe filling up with water?
Why radials and not a direct connection to the roof?
Why is the SWR so high to start with?
Dave Shrader
reasonable. The theoretical VSWR into 50 ohm cable is 1.35:1.
I suspect some water or water vapor is getting into the coax. Water will
'wick up' the coax braid. Use a good Coax-Seal [trade mark] made by
Universal Electronics or equivalent. The Coax-Seal is available from Ham
Radio Outlet, or AES, etc. for about $3.00.
Let the coax dry out then 'squish' on the seal :-)
Deacon Dave, W1MCE
BTW: A VSWR of 2:1 is not bad.
+ + +
Tom Coates
scraps of polystyrene at least two inches long. Or - even better - replace
the rope with heavy monofilament. There's some voltage at the point where
the rope meets the antenna. Moisture and pollutants, whether visible or not,
will have an effect.
Tom, N3IJ
"Q" <mmel...@celeridad.net> wrote in message
news:6298a68d.03051...@posting.google.com...
Wes Stewart
<david....@attbi.com> wrote:
|First, if it's a 1/4 wavelength vertical then a VSWR of 1.7:1 is
|reasonable.
I don't think so.
|The theoretical VSWR into 50 ohm cable is 1.35:1.
Yes but... that's over perfect ground. Anything worse adds loss
resistance in series with radiation resistance to make it closer to 50
ohm.
If this is resonant, and the SWR is actually 1.7:1 then the feedpoint
can't be lower than 50 so it must be higher than 50, which indicates
real trouble.
Bill Turner
news:6298a68d.03051...@posting.google.com:
> I have a 17m vertical mounted on the rooftop which is giving me slight
> SWR problems, and I haven't been able to figure out why. I'm hoping
> someone might have an idea.
>
I'd suggest getting the SWR down to begin with. Use an L network at the
base to match the impedances. The ARRL handbook or antenna book both have
excellent sections about using an L network. If you don't have an MFJ-259
or equivalent SWR analyzer, buy or borrow one. They are invaluable for
this kind of work
And secondly, if you think rain is still a problem, spray different areas
with a garden hose and see what is most sensitive to moisture and take it
from there.
--
Bill, W7TI
Dave Shrader
steel sheet surely makes an excellent counterpoise! IMO, very close to a
perfect ground.
Wes Stewart wrote:
> On Thu, 15 May 2003 00:35:54 GMT, Dave Shrader
> <david....@attbi.com> wrote:
>
> |First, if it's a 1/4 wavelength vertical then a VSWR of 1.7:1 is
> |reasonable.
>
> I don't think so.
<snip> I run 30:1 on 80 meters and have a very efficient antenna. So,
let's not go VSWR crazy. Remember a VSWR of 3:1 which produces a power
loss of 25% is only a -1.25 dB loss of signal and is effectively non
detectable in HF systems.
>
> |The theoretical VSWR into 50 ohm cable is 1.35:1.
>
> Yes but... that's over perfect ground. Anything worse adds loss
> resistance in series with radiation resistance to make it closer to 50
> ohm.
>
> If this is resonant, and the SWR is actually 1.7:1 then the feedpoint
> can't be lower than 50 so it must be higher than 50, which indicates
> real trouble.
I challange the 'real trouble' portion of this statement. Also, see
above statement of 30:1 VSWR.
I have a resonant vertical on 40 meters with three 'poor' wire radials.
The base is 10 feet above ground level and the radials are all over the
place. My MFJ 259B analyzer reports 38 +j0 ohms at the cut frequency of
7.070 MHZ. It's quite easy to achieve reasonable 'ground'.
Deacon Dave, W1MCE
Manager, Textron Antenna Range, retired
Program Chief Engineer, S-118, RS/RV, retired
Mark Keith
> I have a 17m vertical mounted on the rooftop which is giving me slight
> SWR problems, and I haven't been able to figure out why. I'm hoping
> someone might have an idea.
>
> The vertical is real simple: 1" Al tubing, held up by polypropylene
> rope. Where it meets the roof (the roof is galvanized steel sheet)
> it's electrically insulated via a 1/2" PVC pipe with a 1/2" PVC union
> to give it a rim to hold up the antenna.
>
> On the lower part of the union, I placed a steel hose clamp from which
> 3 1/4 wave radials extend. From there, they run on top of the roof at
> the roof angle (15%).
>
> What happens is this: no rain, the antenna adjusts to around 1.7 SWR.
> It works without any problems. Add a 5 or 10 minute drizzle, and the
> SWR goes up to 2.0, which basically makes the transmitter autoprotect.
>
> At the present I have a couple of theories:
>
> 1. Water is getting into the coax. However it doesn't seem likely
> given that there's a "hood" (a cut off soda bottle) on top of the
> whole coax assembly, as well as the typical inverted-U configuration
> to keep water out via gravity.
Adding loss should decrease the SWR. Myself, I'd probably scratch this
one...
>
> 2. The water is changing the "ground" conductivity beneath the
> antenna. However it doesn't seem likely that a wet steel roof will
> change very much with respect to a dry one. Besides that, on one
> occasion I went up and found that the SWR was reading 2.0 and the roof
> was visibly dry.
Could still be damp though, and not really visible. I'm tending to
lean to this one.
>
> 3. The pplene rope is becoming conductive. However, once again I
> checked and found no noticeable moisture on the rope, and the SWR was
> still reading 2.0
Possible if the rope directly touches the radiator. Try adding
insulators as a test, or unhook them temporarily to see what happens.
>
> Aside from that the only thing that occurs to me is that the water is
> somehow partially shorting the vertical radiator to the radials. But
> once again, I'm not sure how much of a problem a film of water on top
> of the PVC along 1 or 2 cm length might represent.
I kind of doubt this one. MK
Wes Stewart
<david....@attbi.com> wrote:
|Wes, the antenna is over a virtual ground plane [perfect ground]! A
|steel sheet surely makes an excellent counterpoise! IMO, very close to a
|perfect ground.
|
|Wes Stewart wrote:
|> On Thu, 15 May 2003 00:35:54 GMT, Dave Shrader
|> <david....@attbi.com> wrote:
|>
|> |First, if it's a 1/4 wavelength vertical then a VSWR of 1.7:1 is
|> |reasonable.
|>
|> I don't think so.
|
|<snip> I run 30:1 on 80 meters and have a very efficient antenna. So,
|let's not go VSWR crazy. Remember a VSWR of 3:1 which produces a power
|loss of 25% is only a -1.25 dB loss of signal and is effectively non
|detectable in HF systems.
I'm not VSWR crazy. You miss the point. And I don't "remember" that
3:1 causes a loss of 25% either. That's mismatch loss and is not a
factor in normal ham operation. If it was, your 30:1 would lose 9 dB.
Sorry, you can't have it both ways.
|
|>
|> |The theoretical VSWR into 50 ohm cable is 1.35:1.
|>
|> Yes but... that's over perfect ground. Anything worse adds loss
|> resistance in series with radiation resistance to make it closer to 50
|> ohm.
|>
|> If this is resonant, and the SWR is actually 1.7:1 then the feedpoint
|> can't be lower than 50 so it must be higher than 50, which indicates
|> real trouble.
|
|I challange the 'real trouble' portion of this statement. Also, see
|above statement of 30:1 VSWR.
"Real trouble" indicates a problem with additional loss, IN THIS
PARTICULAR CASE, not the fact that high SWR is necessarily a bad
thing.
|
|I have a resonant vertical on 40 meters with three 'poor' wire radials.
|The base is 10 feet above ground level and the radials are all over the
|place. My MFJ 259B analyzer reports 38 +j0 ohms at the cut frequency of
|7.070 MHZ. It's quite easy to achieve reasonable 'ground'.
Especially when measuring with an MFJ. All those broadcast station
designers should be using one instead of plowing in those 120 radial
systems.
Owen Duffy
Q wrote:
> At the present I have a couple of theories:
>
> 1. Water is getting into the coax. However it doesn't seem likely
> given that there's a "hood" (a cut off soda bottle) on top of the
> whole coax assembly, as well as the typical inverted-U configuration
> to keep water out via gravity.
>
> 2. The water is changing the "ground" conductivity beneath the
> antenna. However it doesn't seem likely that a wet steel roof will
> change very much with respect to a dry one. Besides that, on one
> occasion I went up and found that the SWR was reading 2.0 and the roof
> was visibly dry.
>
> 3. The pplene rope is becoming conductive. However, once again I
> checked and found no noticeable moisture on the rope, and the SWR was
> still reading 2.0
>
Hmmm, looks like three theories (one more than a couple).
Limiting myself to the issue of the changed situation when wet:
Is water readily available? Could you selectively apply water to a dry
antenna system and explore what needs to be wet for it to behave
differently. Might help identify the element(s) that seem particularly
water sensitive.
Owen
'Doc
Wes Stewart wrote:
>
----(clipped)----
>
> If this is resonant, and the SWR is actually 1.7:1 then the feedpoint
> can't be lower than 50 so it must be higher than 50, which indicates
> real trouble.
You've lost me, why does the impedance of a resonant 1/4 wave
antenna with an SWR of 1.7:1, have to be higher than 50 ohms?
'Doc
'Doc
Obviously, something is changing. It seems that the
most likely 'guess' is #2, the water is changing the
ground system. The next most likely 'guess' is that
either the PVC pipe, or the rope insulators are not
insulating as they should.
If the radials you are using are insulated, I can
imagine that the rain could be causing them to make better
contact with the metal roof when they are wet (water under
the insulation). That would depend a lot on how 'dirty'
your rain is (smog, 'stuff', acid-rain?). The same thing
could be said for the PVC insulator.
I use nylon cord for insulating ropes. I've used some
that was a green color that caused an intermitant problem
such as you are having, because the 'dye' used became
conductive when wet. I have no idea if that may be the cause
of your problem, but it is something to keep in mind when
nothing else doesn't work.
'Doc
Dan Richardson
Because a resonate 1/4-wave monopole does not have a 30-Ohm impedance.
Danny, K6MHE
'Doc
Dan,
What does it have?
'Doc
'Doc
Thought we were talking about a 1/4 wave groundplane,
which can have a 30 ohm input impedance.
'Doc
Wes Stewart
You said:
"Wes, the antenna is over a virtual ground plane [perfect ground]! A
steel sheet surely makes an excellent counterpoise! IMO, very close to
a perfect ground."
This is different from a 1/4 vertical with resonant radials.
Mario said there were three radials that are (if I understand
correctly) laying on the steel roof. I wondered why he didn't use a
direct connection to the roof instead of using radials. This would
remove some of the uncertainty we are worrying about.
Nevertheless, I submit that while the ground is finite and sloped
slightly (15 deg) the antenna more closely approximates a monopole
over infinite ground than it does an elevated vertical with resonant
radials. (Here we seem to agree on the same assumption)
That said, the theoretical resonant impedance would be close to 36
ohm, but never below that figure. To bring this up to an SWR of 1.7:1
would require the equivalent of 50 ohm of loss resistance in series
with the feedpoint. Which is why I questioned the 1.7 number in the
first place.
OC_CAM's Razor
The feedpoint impedance of a 1/4WL groundplane over perfect
ground cannot be lower than ~35 ohms. If the ground is not
perfect, the feedpoint impedance will be higher than 35 ohms.
--
73, Cecil http://www.qsl.net/w5dxp
-----= 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! =-----
Dan Richardson
Around 36-Ohms
Danny
'Doc
Now we are 'assuming' Mario has a 'perfect ground'.
I sure wish someone would 'assume' me a raise on my
check...
'Doc
Wes Stewart
wrote:
|On Thu, 15 May 2003 07:18:00 -0500, 'Doc <w5...@cwis.net> wrote:
|
|| Oops! Where did this 1/4 wave monopole come from?
||Thought we were talking about a 1/4 wave groundplane,
||which can have a 30 ohm input impedance.
|| 'Doc
|
|
Uh Oh.
|You said:
|
|"Wes, the antenna is over a virtual ground plane [perfect ground]! A
|steel sheet surely makes an excellent counterpoise! IMO, very close to
|a perfect ground."
My apologies Doc, I mixed up "Deacons" and "Docs" "Deacon Dave" said
the above.
Richard Harrison
"If the ground is not perfect, the feedpoint impedance will be higher
than 35 ohms."
Yes, and if the radiator is too short, the radiation resistance can be
less than 35 ohms.
Why not make the radiator a little too long and bring the radiation
resistance up to 50 ohms, then tune out the excess inductance with a
relatively lossless series capacitance?
Best regards, Richard Harrison, KB5WZI
'Doc
Wes,
No problem. By the way, I'm the one that's short,
fat, with a Harley hat and hair all over my head...
'Doc
Mark Keith
> Why not make the radiator a little too long and bring the radiation
> resistance up to 50 ohms, then tune out the excess inductance with a
> relatively lossless series capacitance?
Pretty common antenna for 160m and other bands. IE: 3/8 wave, 3/8 wave
inv L's, etc. Also raises the current maximum up the antenna from the
base a bit, which can help reduce ground losses. I used an L , 190 ft
long on 160m for this purpose at one time. For most paths, it did
outdo the 1/4 wave L. But I didn't use too many radials. MK
Mark Keith
> > 3 1/4 wave radials extend. From there, they run on top of the roof at
> > the roof angle (15%).
As one mentioned, if the radials are actually laying on the metal
roof, I would think the radials are quite effected by the roof, and
detuned in comparison to having them elevated over the roof a bit.
Changes in the quality of the roof connections would probably be
noticable in your case because of the strong coupling to the roof. It
would be better to get them off the roof a bit, unless you are trying
to include the roof as the ground system. And in that case, the
radials are really unneeded unless the roof is smaller than the
radials used.
Using the roof as the ground system would require all sheets to make
good connection. I've had many times where I had trouble with just
aluminum antennas in themselves making bad connections after a long
time outdoors. They would intermit as one section would lose
connection. Sometimes I could fire up my amp and give it a blip of
high power and "weld" it together for a short time...But it always
comes back...You end up having to clean it, and relube with conductive
grease. Common problem with beams also..
> >
> > 2. The water is changing the "ground" conductivity beneath the
> > antenna. However it doesn't seem likely that a wet steel roof will
> > change very much with respect to a dry one. Besides that, on one
> > occasion I went up and found that the SWR was reading 2.0 and the roof
> > was visibly dry.
>
> Could still be damp though, and not really visible. I'm tending to
> lean to this one.
Whoa...I goofed up here. I was thinking you had regular shingles for
some reason. Water on the surface of the steel roof should have no
real effect.
It may be possible water is changing the quality of the connections of
any connecting sheets under the antenna though. But seems to me, it
would need to rain a long time, and have a high humidity, being that
normally water shouldn't be running underneath the sheets. "I'm
assuming sheets of some kind are used."
I think I'd quickly rule out the ropes with a quick test unhooking or
insulating them. If it still happens, I'd be thinking roof. It might
just be the ropes...Also as one said, I'd improve the match to give a
wider usable bandwidth and avoid the shutdown problem even with the
small change. MK
Dave Shrader
Me too!! 5'8" short, 250 #, and Santa Claus' beard and hair!! But NO
HARLEY. I haven't developed that passion yet!
Deacon Dave, W1MCE
Jimmy
start. I would start over verifying that everything is good, connections,
length of antenna, coax. I would get rid of the PVC. I personally dont like
this stuff for iinsulators. It contaminates easily and weathers poorly.. In
the past I have run antennas like this and found "long neck" bottles work
great as insulators. Since you have a metal roof I would just use the roof
as a counterpoise. The slight change in SWR during rain is fairly typical
but since you are on the edge of what your transciever can handle it is
causing a problem. My bet would be that your radiator is the wrong length. I
am assuming that the antenna is mounted on the peak of the roof. If so the
SWR should probably be in the 1.1-1.2:1 range. If it is mounted on the slope
of the roof the antenna could have a lower feedpoint impedance accounting
for the higher than expected SWR IF the latter is the case either move the
antenna to the peak or install a 1/4 wl section of 70 ohm cable at the
antenna. This should improve the match between radio and antenna. The 1/4 wl
section should only be used after every other possibility is eliminated and
you have determined your antenna is operating at resonance and all
connections and componets are good.
"Q" <mmel...@celeridad.net> wrote in message
news:6298a68d.03051...@posting.google.com...
Jimmy
impedance could be around 25 0hms. We are only assuming it is mounted on the
peak
"Richard Harrison" <richard...@webtv.net> wrote in message
news:21692-3EC...@storefull-2314.public.lawson.webtv.net...
Richard Harrison
"We are only assuming it is mounted on the peak."
Environment affects antenna impedance. Mario has several problems,
including SWR.
Often the user has tuned out the reactance of his vertical antenna and
asks why it doesn`t now match his coax. Simple. The antenna now looks
like a resistor but that resistor does not match Zo.
See Fig 2 on page 2-3 of the 19th edition of the ARRL Antenna Book. This
figure is the impedance of a dipole, R&X, versus frequency, but a
monopole versus length works the same way, just different numbers, not
different curve shape.
Making the antenna slightly too long makes it inductive and also raises
its radiation resistance. The radiatiation resistance can be adjusted by
the excess length to match Zo. Then when the excess inductive reactance
is tuned out with a series capacitance, the antenna matches Zo.
Rain may affect the influence of the nearby roof. If so, this is solved
by adjusting the roof, moving the antenna away from the roof, or adding
enough radials to hide the roof from the antenna.
Jimmy
> Jimmy wrote:
> "We are only assuming it is mounted on the peak."
>
> Environment affects antenna impedance. Mario has several problems,
> including SWR.
>
OC_CAM's Razor
> CAM, W5DXP wrote:
> "If the ground is not perfect, the feedpoint impedance will be higher
> than 35 ohms."
>
> Yes, and if the radiator is too short, the radiation resistance can be
> less than 35 ohms.
True, but the stated assumed boundary conditions for the discussion
was that the vertical was physically 1/4WL long.
> Why not make the radiator a little too long and bring the radiation
> resistance up to 50 ohms, then tune out the excess inductance with a
> relatively lossless series capacitance?
Good suggestion and it works really well. That is, as well as a
vertical ever works. :-)
Q
> Is the PVC pipe filling up with water?
PVC is open at the bottom, so there's no water accumulating.
> Why radials and not a direct connection to the roof?
That's an interesting question I've been pondering for some time.
There's a couple of mecanical reasons (mainly how to connect the braid
to the roof tin in an electrically effective manner that won't rust
over time), but the real reason I chose the radials was to discard
that the roof should be what's causing the weird SWR readings. Since
the roof has a slightly irregular shape (kind of a T shape) I figured
it might not be the "ideal" infinite ground you'd think it would be.
Now, the question that comes up... what exactly is considered an
"infinite" ground plane? It occurs to me there must be some kind of
relation between wavelength and extension of the plane that would make
radials superior or inferior. Would using a less conductive material
(galvanized steel vs. copper/aluminium) cause any notable attenuation
effects or just the expected variation in antenna impedance?
> Why is the SWR so high to start with?
When I took down the 20m vertical I had installed on the roof I found
out why. Seems the two weren't getting along too well... in fact later
on I took measurements and they were separated about 0.17W on 17m, and
the 20 meter vertical is 25% longer than the 18 meter. Seems the other
vertical was accidentally acting as an incorrectly-calibrated
reflector and that's what was causing the SWR to act up. Probably when
the roof was wet the conductivity changed slightly, but enough to send
the whole "array" out of "adjustment".
Strange thing is, when I ran a model later on (taking into account
roof angles, antenna lenghts, radials, etc) no interaction between the
two came up.
One of those weird interactions that comes up at times I guess.
---
Mario
Q
> Now we are 'assuming' Mario has a 'perfect ground'.
> I sure wish someone would 'assume' me a raise on my
> check...
I guess that if you compare it in conductivity and loss to your normal
backyard soil, the tin metal roof could be considered a "perfect"
ground. The doubts I have would be not so much with the "perfect"
part, but with the other part in theorical verticals: "infinite".
---
Mario
Q
> Why not make the radiator a little too long and bring the radiation
> resistance up to 50 ohms, then tune out the excess inductance with a
> relatively lossless series capacitance?
How easy to do would that be construction-wise? Sounds like a
reasonable alternative, but depending on how exactly you go about
doing it it might turn out it's simply easier to deal with the plain
vanilla flavored vertical impedance.
---
Mario
Q
> He may have his antenna mounted on the slope of the roof. If so his
> impedance could be around 25 0hms. We are only assuming it is mounted on the
> peak
No, it's basically at the peak. Roof slope is 15%, north and south.
---
Mario
Jimmy
be very close to 1 to 1 when its dry. Maybe the radiator is the wrong
length. Have you done any prunning or just calculate,cut and stick it up