purpose of a tower-top amplifier

Dana wrote:

> "Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
> news:f8v2q25qtibnk8r915f7o2vi9thoejt5v1@4ax.com...
>
>>Tony Hwang <dragon40@shaw.ca> hath wroth:
>>
>>
>>>Never saw a coax radiating or picking up signal like crazy in real
>>>world? Want to experiment? Raise the SWR on the transmission line
>>>and scan(sniff)the coax along, what do you see?
>>
>>Wrong. I do that experiment as a demonstration of common electronic
>>assumptions (along with the loss through a mess of adapters, water in
>>the coax, and others).
>
>
> Water in the coax or feedline, how I hate that, especially here in Alaska
> where the water then freezes inside the cable.
> Talk about a gremlin type of problem.

No wonder wave guide feeding antenna is usually pressurized with
Nitrogen gas.

>
> Somewhere along the line, someone mumbled that
>
>>VSWR causes RF to radiate down the outside of the coax shield. It
>>made sense because every time I ran coax to a high VSWR antenna, I had
>>the coax radiate enough RF to light up fluorescent bulbs in the shack.
>>
>>However, that's not the way it works. The previous example is an
>>uncontrolled environment. What's happening is that the coax cable has
>>become part of the antenna system and will radiate as described. But,
>>what if I replace the high VSWR antenna with a high VSWR dummy load?
>>If the reflected signal comes down the outside of the coax, as is
>>commonly suggested, then it should radiate as badly as the high VSWR
>>antenna. It doesn't.
>>
>>You can try your RF sniffer experiment the same way. Put a high VSWR
>>dummy load on the end of a piece of coax and go sniffing for RF. You
>>won't find any. As long as the field between the center conductor and
>>the shield is totally enclosed, you can have a high VSWR termination,
>>but no radiation outside the shield.
>>
>>
>>--
>>Jeff Liebermann jeffl@comix.santa-cruz.ca.us
>>150 Felker St #D http://www.LearnByDestroying.com
>>Santa Cruz CA 95060 http://802.11junk.com
>>Skype: JeffLiebermann AE6KS 831-336-2558
>
>
>

Tony Hwang <dragon40@shaw.ca> hath wroth:

>No wonder wave guide feeding antenna is usually pressurized with
>Nitrogen gas.

Nope. They use dried compressed air. Dry nitrogen is too expensive
and juggling bottles is a maintenance nightmare.
<http://www.tessco.com/yts/partner/manufacturer_list/vendors/puregas/index.html>

--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

Tony Hwang <dragon40@shaw.ca> hath wroth:

>How come a guy miles away picked up my signal when I was folling around
>with matched dummy load? When Z is smatched SWR is low but perfect SWR
>does not gurantee perfect match. You explained it just now.

Your unspecified radio probably wasn't very well shielded. You'll be
amazed at how much RF is conducted out of various holes and cables on
the typical ham radio. Need some entertainment value? Build an RF
sniffer or use a scope with a small loop at the end of a piece of
coax. Put a dummy load on your xmitter and sniff around the case, mic
cable, power cord, and tuner connections and see how much RF is
leaking out. Then do the math to figure out the field strength.
Perverse square law says that the guy that was "miles" away might get
as strong a signal as someone across the country if you used a proper
antenna. Just a wild guess, but methinks that a really good HF radio
can offer about -30dB shielding. 100 watts is +50dBm, so you've got
perhaps +20dBm or at least 100 milliwatts for RF. Welcome to QRP.

>Yes, coax can be leaky electrically. Hmmm,

Do the math. Typical braided coax has about 98% coverage. That means
that somewhat less than 2% of the power leaks out. I'll guess 1%.
(Actually, it varies with frequency and length but I don't wanna get
complicated). 1% = 0.01 = -20dB. If you have 100 watts (+50dBm)
floating around the coax cable, we have somewhat less than 50 - 20 =
+30dBm or 1 watt of power leaking out of the coax. Welcome to QRP
again.

>No wonder mil-spec stuffs have double triple shielding compared to
>commercial counter part.

I prefer foil shielding instead of multiple layers of braid. Braid to
too messy to crimp.

--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

Tony Hwang wrote:
> decaturtxcowboy wrote:
>> Noise floor issues yes. Coax generated noise, no. I don't know what
>> he was smoking when he wrote that article.
>
> Never saw a coax radiating or picking up signal like crazy in real
> world? Want to experiment? Raise the SWR on the transmission line
> and scan(sniff)the coax along, what do you see?

That true, but its not the coax (per se) that is creating the noise where
the classical definition of noise is any unwanted signal - its the mismatch
of impedances creating standing waves.

"Dana" <raff242@yahoo.com> wrote in message
news:12q01dsdlua047d@corp.supernews.com...

| > > On Jan 6, 10:12 am, "Dana" <raff...@yahoo.com> wrote:
| > > > Everything produces noise.
| > > > And in the applications I was describing we had anywhere from 100 to
| 600
| > > > feet of cable to deal with.
| > >
| > > Wrong. Coax cable does not normally generate any noise by itself.
|
| Here is an article that describes noise in RF systems
| http://telephonyonline.com/wireless/...luating_noise/
| All matter at temperatures above absolute zero (0K, about -460F) radiates
| electromagnetic energy. The amount of energy is related to temperature --
| the hotter the matter, the more energy is radiated. This energy is
described
| by Boltzmann's Constant, 'k' (k = -198.6dBm/degreesK-Hz). This constant,
| multiplied by the temperature of the matter a receiver views and the
system
| bandwidth, yields an irreducible background noise against which a desired
| signal must compete. This is thermal noise.
|
| In a cellular system, the receiving antennas are designed to view the
ground
| around the site because that's where the subscribers are. The ground
| temperature varies, but at 80F, it's about 300K (T=300K). RF engineers
| typically use this number as a rule of thumb.
|
| The receiver bandwidth varies depending on the technology, but the same
| principles hold for all technologies. EAMPS, for example, uses 30kHz-wide
| channels. Receiver bandwidth is a bit less than 30kHz, for rejection of
| adjacent channels. Assume the typical EAMPS receiver has a bandwidth of
| 25kHz (B=25,000Hz). By making this assumption, you can calculate the
amount
| of noise an EAMPS receiver will have in its passband if it contributes no
| noise of its own.
|
| This receiver thermal noise floor often is referred to as 'kTB.' In the
| example, assume consistent units:
|
| kTB = -198.6 + 10 Log(300)
|
| + 10 Log(25,000) in dBm
|
| kTB = -129.8dBm
|
| Thus, if you build a perfect EAMPS cellular receiver, it would have -129.
| 8dBm of noise in its passband competing with the wanted signal.
|
| CABLE LOSS Cable, filters and other passive elements exhibit a loss and
| produce thermal noise.
|
| If a cable (or other lossy element) has 10dB of loss, it will attenuate
the
| desired signal as well as the input noise by 10dB. But at the output of
the
| cable, you will see noise at least equal to kTB because the cable itself
| contributes it.
|
| If you put a signal into the cable at -100dBm over a thermal noise
of -129.
| 8dBm, you have a signal-to-noise ratio of 29.8dB at the cable input. At
the
| cable output, the signal has been attenuated by 10dB to -110dBm. The noise
| you put into the cable also has dropped the same amount, to -139.8dBm. But
| the cable contributes its noise floor of -129.8dBm, so the combined
| (uncorrelated) noise terms are -129.8dBm. The resulting signal-to-noise
| ratio is only 19.8dB at the cable output. You sacrifice 10dB of
| signal-to-noise ratio. This is why you spend money on 7/8", 15/8" or
larger
| coax at cell sites to reduce this loss.

In the real world such calculations mean little as with most wireless
systems the problem is not a noise but interference floor. Much of this is
in band interference but out of band can be significant especially in metro
areas.

"Tony Hwang" <dragon40@shaw.ca> wrote in message
news:zAkoh.583485$R63.237354@pd7urf1no...
> Dana wrote:
>
> > "Jeff Liebermann" <jeffl@comix.santa-cruz.ca.us> wrote in message
> > news:f8v2q25qtibnk8r915f7o2vi9thoejt5v1@4ax.com...
> >
> >>Tony Hwang <dragon40@shaw.ca> hath wroth:
> >>
> >>
> >>>Never saw a coax radiating or picking up signal like crazy in real
> >>>world? Want to experiment? Raise the SWR on the transmission line
> >>>and scan(sniff)the coax along, what do you see?
> >>
> >>Wrong. I do that experiment as a demonstration of common electronic
> >>assumptions (along with the loss through a mess of adapters, water in
> >>the coax, and others).
> >
> >
> > Water in the coax or feedline, how I hate that, especially here in
Alaska
> > where the water then freezes inside the cable.
> > Talk about a gremlin type of problem.
>
> No wonder wave guide feeding antenna is usually pressurized with
> Nitrogen gas.

And have dehumidifiers.
And that is if the site in question is even using waveguide to begin with.
>
> >
> > Somewhere along the line, someone mumbled that
> >
> >>VSWR causes RF to radiate down the outside of the coax shield. It
> >>made sense because every time I ran coax to a high VSWR antenna, I had
> >>the coax radiate enough RF to light up fluorescent bulbs in the shack.
> >>
> >>However, that's not the way it works. The previous example is an
> >>uncontrolled environment. What's happening is that the coax cable has
> >>become part of the antenna system and will radiate as described. But,
> >>what if I replace the high VSWR antenna with a high VSWR dummy load?
> >>If the reflected signal comes down the outside of the coax, as is
> >>commonly suggested, then it should radiate as badly as the high VSWR
> >>antenna. It doesn't.
> >>
> >>You can try your RF sniffer experiment the same way. Put a high VSWR
> >>dummy load on the end of a piece of coax and go sniffing for RF. You
> >>won't find any. As long as the field between the center conductor and
> >>the shield is totally enclosed, you can have a high VSWR termination,
> >>but no radiation outside the shield.
> >>
> >>
> >>--
> >>Jeff Liebermann jeffl@comix.santa-cruz.ca.us
> >>150 Felker St #D http://www.LearnByDestroying.com
> >>Santa Cruz CA 95060 http://802.11junk.com
> >>Skype: JeffLiebermann AE6KS 831-336-2558
> >
> >
> >

"NotMe" <me@privacy.net> wrote in message
news:zhvoh.17$Ud6.11@newsfe02.lga...
>
> "Dana" <raff242@yahoo.com> wrote in message
> news:12q01dsdlua047d@corp.supernews.com...
>
> | > > On Jan 6, 10:12 am, "Dana" <raff...@yahoo.com> wrote:
> | > > > Everything produces noise.
> | > > > And in the applications I was describing we had anywhere from 100
to
> | 600
> | > > > feet of cable to deal with.
> | > >
> | > > Wrong. Coax cable does not normally generate any noise by itself.
> |
> | Here is an article that describes noise in RF systems
> | http://telephonyonline.com/wireless/...luating_noise/
> | All matter at temperatures above absolute zero (0K, about -460F)
radiates
> | electromagnetic energy. The amount of energy is related to
temperature --
> | the hotter the matter, the more energy is radiated. This energy is
> described
> | by Boltzmann's Constant, 'k' (k = -198.6dBm/degreesK-Hz). This constant,
> | multiplied by the temperature of the matter a receiver views and the
> system
> | bandwidth, yields an irreducible background noise against which a
desired
> | signal must compete. This is thermal noise.
> |
> | In a cellular system, the receiving antennas are designed to view the
> ground
> | around the site because that's where the subscribers are. The ground
> | temperature varies, but at 80F, it's about 300K (T=300K). RF engineers
> | typically use this number as a rule of thumb.
> |
> | The receiver bandwidth varies depending on the technology, but the same
> | principles hold for all technologies. EAMPS, for example, uses
30kHz-wide
> | channels. Receiver bandwidth is a bit less than 30kHz, for rejection of
> | adjacent channels. Assume the typical EAMPS receiver has a bandwidth of
> | 25kHz (B=25,000Hz). By making this assumption, you can calculate the
> amount
> | of noise an EAMPS receiver will have in its passband if it contributes
no
> | noise of its own.
> |
> | This receiver thermal noise floor often is referred to as 'kTB.' In the
> | example, assume consistent units:
> |
> | kTB = -198.6 + 10 Log(300)
> |
> | + 10 Log(25,000) in dBm
> |
> | kTB = -129.8dBm
> |
> | Thus, if you build a perfect EAMPS cellular receiver, it would
have -129.
> | 8dBm of noise in its passband competing with the wanted signal.
> |
> | CABLE LOSS Cable, filters and other passive elements exhibit a loss and
> | produce thermal noise.
> |
> | If a cable (or other lossy element) has 10dB of loss, it will attenuate
> the
> | desired signal as well as the input noise by 10dB. But at the output of
> the
> | cable, you will see noise at least equal to kTB because the cable itself
> | contributes it.
> |
> | If you put a signal into the cable at -100dBm over a thermal noise
> of -129.
> | 8dBm, you have a signal-to-noise ratio of 29.8dB at the cable input. At
> the
> | cable output, the signal has been attenuated by 10dB to -110dBm. The
noise
> | you put into the cable also has dropped the same amount, to -139.8dBm.
But
> | the cable contributes its noise floor of -129.8dBm, so the combined
> | (uncorrelated) noise terms are -129.8dBm. The resulting signal-to-noise
> | ratio is only 19.8dB at the cable output. You sacrifice 10dB of
> | signal-to-noise ratio. This is why you spend money on 7/8", 15/8" or
> larger
> | coax at cell sites to reduce this loss.
>
> In the real world such calculations mean little

That is not true at all.
We use those calculations when designing our cell sites and microwave shots
(10 to 30 mile shots are not uncommon.

as with most wireless
> systems the problem is not a noise but interference floor.


Noise is interference

> Much of this is
> in band interference but out of band can be significant especially in
metro
> areas.

Yes urban areas create even more issues to overcome.


>
>
>
>
>

Johann Beretta wrote:
> They most certainly use nitrogen gas in military aircraft. And it's no
> maintenance problem. The bottles are easily refilled every few days. Takes about
> 5 minutes.

But then...military may spend five times their flight time having ground
maintenance done.

I've seen both compressed dry air cylinders and cryogenic tanks filled with
liquid nitrogen next to splice cases.

Jeff Liebermann wrote:

> Tony Hwang <dragon40@shaw.ca> hath wroth:
>
>
>>How come a guy miles away picked up my signal when I was folling around
>>with matched dummy load? When Z is smatched SWR is low but perfect SWR
>>does not gurantee perfect match. You explained it just now.
>
>
> Your unspecified radio probably wasn't very well shielded. You'll be
> amazed at how much RF is conducted out of various holes and cables on
> the typical ham radio. Need some entertainment value? Build an RF
> sniffer or use a scope with a small loop at the end of a piece of
> coax. Put a dummy load on your xmitter and sniff around the case, mic
> cable, power cord, and tuner connections and see how much RF is
> leaking out. Then do the math to figure out the field strength.
> Perverse square law says that the guy that was "miles" away might get
> as strong a signal as someone across the country if you used a proper
> antenna. Just a wild guess, but methinks that a really good HF radio
> can offer about -30dB shielding. 100 watts is +50dBm, so you've got
> perhaps +20dBm or at least 100 milliwatts for RF. Welcome to QRP.
>
>
>>Yes, coax can be leaky electrically. Hmmm,
>
>
> Do the math. Typical braided coax has about 98% coverage. That means
> that somewhat less than 2% of the power leaks out. I'll guess 1%.
> (Actually, it varies with frequency and length but I don't wanna get
> complicated). 1% = 0.01 = -20dB. If you have 100 watts (+50dBm)
> floating around the coax cable, we have somewhat less than 50 - 20 =
> +30dBm or 1 watt of power leaking out of the coax. Welcome to QRP
> again.
>
>
>>No wonder mil-spec stuffs have double triple shielding compared to
>>commercial counter part.
>
>
> I prefer foil shielding instead of multiple layers of braid. Braid to
> too messy to crimp.
>
Hmmm.
Let me see, what is typical RF power this COMSATs are using? And
distance it covers? 1W is lot of power for sure. 5GHz range
M/W links on LOS or refraction path I used to deal with had TX power of
3 to 5W. So 1W is lot of power in my book. Then I used to work on TX
power of 100s of Kilo Watts. Ever worked on High voltage in the range of
30KV? Nowadays it's different. Solid state devices don't need such high
voltage.

"Johann Beretta" <invalid@invalid.org> wrote in message
news:bq56q21s8lce8hgirv9icfqqumaid6947v@4ax.com...
> On Sun, 07 Jan 2007 23:46:48 -0800, Jeff Liebermann
> <jeffl@comix.santa-cruz.ca.us> wrote:
>
> >Tony Hwang <dragon40@shaw.ca> hath wroth:
> >
> >>No wonder wave guide feeding antenna is usually pressurized with
> >>Nitrogen gas.
> >
> >Nope. They use dried compressed air. Dry nitrogen is too expensive
> >and juggling bottles is a maintenance nightmare.
>
><http://www.tessco.com/yts/partner/ma.../puregas/index.
html>
>
> They most certainly use nitrogen gas in military aircraft. And it's no
> maintenance problem. The bottles are easily refilled every few days. Takes
about
> 5 minutes.

True enough, we use nitrogen at our microwave sites that have waveguide,
along with dehumidifiers.

"Dana" <raff242@yahoo.com> hath wroth:

>True enough, we use nitrogen at our microwave sites that have waveguide,
>along with dehumidifiers.

From the Andrew Corp data sheets on waveguide and coax cables:

http://www.andrew.com/search/BN_17815.aspx
Dry air is normally used for pressurizing. Dry
nitrogen may also be used. When pressurizing equipment
is connected to a gas port on the waveguide connector,
or whenever pipe fittings are reassembled, threads
must be covered with sealing tape to ensure leakproof
connection.

http://www.andrew.com/search/BN_17800.aspx
Pressurization is needed in air-dielectric cables
because changes in temperature can cause condensation
of moisture from outside air that enters the cable.
This moisture can seriously impair the efficiency of
system operation. Connecting a pressure source of dry
air or nitrogen (dehydrator) to the cable at slightly
more than atmospheric pressure will correct this condition
since the moisture will be removed and air will then
leak from instead of into the cable.

At one community repeater system I maintained in the early 1970's, I
used a compressed air drying filter and a bicycle pump. I ran the
pressure up to about 7 psi. It would leak down to about 2 psi in 3
months. I missed re-pressurizing the coax one winter and got to spend
a back breaking day purging the coax. I built a small pressurization
system from an air brush pump and eliminated the bicycle pump ordeal.
Refridgerated dessicating system are nice, but you can get away with
much less if you don't mind purging the coax every few years.


--
Jeff Liebermann jeffl@comix.santa-cruz.ca.us
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558