Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: physical separation between TX/RX antennas?  (Read 4734 times)
AK4YA
Member

Posts: 106




Ignore
« on: October 17, 2012, 07:39:11 AM »

If I have a radio whose front end can take up to 13dBm of input before overload/damage,  How much distance should there be between a dipole TX antenna and a vertical RX antenna, given no freq offset (both TX and RX are on the same frequency)?

Is there a rule of thumb or something?  I understand I should model both antennas relative to their position to each other, but as far as the distance between them, Im not sure at that point.
Logged
WX7G
Member

Posts: 6136




Ignore
« Reply #1 on: October 17, 2012, 09:51:32 AM »

The missing pieces of the equation are:

1. Transmitter power
2. Frequency
Logged
AK4YA
Member

Posts: 106




Ignore
« Reply #2 on: October 17, 2012, 10:27:44 AM »

The missing pieces of the equation are:

1. Transmitter power
2. Frequency


Lets say 18.1MHz at 54dBm/300watts.  I was hoping for the equation, that way any values could be applied.
Thanks!
Logged
WB6BYU
Member

Posts: 13341




Ignore
« Reply #3 on: October 17, 2012, 11:09:22 AM »

At a first order estimate there would be no coupling because the antennas are cross-
polarized.

That doesn't mean that you'll get infinite isolation, however, but rather it means that
we have to look much more closely at the relative positions and orientations of the
two antennas, as well as other factors (such as coupling through the feedlines,
common mode currents, or the interactions of mutually-coupled objects such
as rain gutters) in order to come up with a reasonable estimate.

For example, if the vertical is placed near the center of the dipole and there is no
extraneous signal coupled via the dipole feedline hanging down through the pattern
of the vertical, the total coupling may be small because both sides of the dipole
will pick up about the same signal, which will tend to cancel each other.   But if
the vertical is near one end of the dipole, you can have capacitive coupling between
the ends of the antennas, as well as pickup along the antenna wire (horizontal
dipoles respond to vertically polarized signals off the ends) that aren't the same
on both halves of the dipole.


If it were me, I'd start by modeling the two antennas in EZNEC or another software
program, driving one antenna and measuring the current at the feedpoint of the other.
This will give you a starting point, and allow you to experiment with different
placements, but the result probably will still need some derating due to coupling via
other paths as well.

Then put up the two antennas, apply power to one and measure the actual output
power from the other before connecting your receiver.
Logged
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #4 on: October 17, 2012, 12:02:15 PM »

Lets say 18.1MHz at 54dBm/300watts.  I was hoping for the equation, that way any values could be applied.
Thanks!

Be real careful with this problem. A perfect dipole is ONLY horizontally polarized exactly broadside to the dipole. As we move off broadside toward the ends, polarization tilts progressively more vertical. This means the perfect horizontal dipole antenna is only perfectly horizontal in two directions along the earth, and becomes more and more vertically coupling prone as we move to the ends.

If I have a perfect dipole and perfect vertical spaced a reasonable distance apart to get out of near field or induction fields, the maximum null between the two is exactly broadside to the dipole. There is increased coupling as we near the end of the dipole to the vertical, and then, depending on geometry and distance coupling can null again (although generally reduced from the broadside null) as we move exactly off the ends.

So we see.... the rough answer is generally the best null is usually when the dipole is centered on and **broadside** to the vertical. In the real world, a problem like this would require an auctal measurement of coupling power or coupled loss.

There is some measured data here:

http://www.w8ji.com/antenna_coupling.htm

13 dBm seems extremely low. That is only 20 milliwatts, and I routenly make IMD, desense, and IP3 measurements at +20 dBm or more (100 milliwatts). I've never found a receiver that could be damaged by 100 mW.

What receiver do you have??

73 Tom
Logged
K5LXP
Member

Posts: 4507


WWW

Ignore
« Reply #5 on: October 17, 2012, 02:23:27 PM »

If I have a radio whose front end can take up to 13dBm of input before overload/damage,  How much distance should there be between a dipole TX antenna and a vertical RX antenna, given no freq offset (both TX and RX are on the same frequency)?

A greater distance than a typical residential lot for sure.  54dBm to 13dBm = 41dB.  Even with 0dBi antennas you would need spacing of a couple hundred yards or so.  You would have to take specific measures to minimize the coupling (like the aforementioned polarization difference) to achieve this at a shorter distance.

If it's just damaging levels you're worried about then W8JI is right, your levels seem awfully low.  The level where "damage" occurs is a lot higher than when intermod products start appearing.  Since you can't practically operate duplex (TX freq is RX freq), the AGC recovery time would be pretty long so you would need some kind of muting during transmit.  If you wanted, this mute function could also incorporate some attenuation in the receive path making the antenna isolation requirement moot.  A basic microwave relay with a built-in termination could accommodate this directly.

It's extremely difficult to estimate or model what isolation you'll actually get even if all the particulars of the antennas and patterns are known due to all the imponderables of the installation site you have.  It's pretty easy to measure it though, just a known low level source on the TX antenna and a calibrated receiver S-meter on the receive antenna, and you'll know right away what the levels will be at full-soup TX power.


Mark K5LXP
Albuquerque, NM
« Last Edit: October 17, 2012, 02:27:38 PM by K5LXP » Logged
WX7G
Member

Posts: 6136




Ignore
« Reply #6 on: October 17, 2012, 02:46:16 PM »

I agree with WB6BYU that a NEC model is the best way to approach it. To get started though here are some numbers for dipoles in free space having the same polarization:

1 watt TX
10 meter separation
100 MHz
1 mW RX


The power into the RX is directly proportional to TX power.
The power into the RX is inversely proportional to distance squared
The power into the RX is inversely proportional to frequency squared

« Last Edit: October 17, 2012, 02:59:43 PM by WX7G » Logged
AK4YA
Member

Posts: 106




Ignore
« Reply #7 on: October 17, 2012, 06:33:01 PM »

<SNIP>
13 dBm seems extremely low. That is only 20 milliwatts, and I routenly make IMD, desense, and IP3 measurements at +20 dBm or more (100 milliwatts). I've never found a receiver that could be damaged by 100 mW.

What receiver do you have??

73 Tom
Ya'll have lots of good ideas and starting points for me.  Thanks a lot.  Ill probably start off with a coax relay and use it HDX to begin with and later put up the vertical antenna and take some measurements.  Its a QS1R SDR.  Their webpage has 2 pieces of info:
The QS1R typically overloads at +9 to +10 dBm
You should not exceed ~3V pk-pk into 50 ohms. That is about 20 mW or 13 dBm.

Thanks again.  Hope to finally be on the air by the end of the year.

73, AK4YA
Logged
WX7G
Member

Posts: 6136




Ignore
« Reply #8 on: October 17, 2012, 06:40:47 PM »

90 meters dipole to dipole free space. Add a 10 dB attenuator in front of the RX and it becomes 30 meters. 
« Last Edit: October 17, 2012, 06:51:25 PM by WX7G » Logged
W8JI
Member

Posts: 9296


WWW

Ignore
« Reply #9 on: October 18, 2012, 06:48:25 AM »

Ya'll have lots of good ideas and starting points for me.  Thanks a lot.  Ill probably start off with a coax relay and use it HDX to begin with and later put up the vertical antenna and take some measurements.  Its a QS1R SDR.  Their webpage has 2 pieces of info:
The QS1R typically overloads at +9 to +10 dBm
You should not exceed ~3V pk-pk into 50 ohms. That is about 20 mW or 13 dBm.

Thanks again.  Hope to finally be on the air by the end of the year.

73, AK4YA

That is what I don't like about SDR receivers. The front ends and RF sections have virtually no selectivity, often being mixers coupled without any real selectivity to antennas.

This means EVERY signal within coupling passband from the antenna port adds to the power the mixer has to process. While they test as good as a normal receiver for specs, a normal receiver will only subject the first mixer to signals inside the band range the front end filters are on.

Since the test does not reflect the hundreds or thousands of signals the SDR "mixer" sees, and only subjects it to one or two signals, it falsely appears competitive with filtered receivers.

This is separate from the overload worry you mention, except the overload worry extends to out-of-band signals. With a normal receiver, the system might tolerate 1/2 watt or more out-of-band. With the SDR, lacking band filtering, what it tolerates in-band is what it tolerates out-of-band.

The problem with the overload spec you posted is we do not know if it is just where the SDR system becomes temporarily useless, perhaps by overflowing an A-D conversion or saturating a mixer, or if that level is where non-recoverable electrical damage occurs.

Most receivers do NOT give a destructive power level guideline, but I have yet to find a receiver electrically damaged by my signal generators at +20 dBm or more in two tone tests. That's over 100mW from two generators combined at one time.

73 Tom
Logged
KC4MOP
Member

Posts: 743




Ignore
« Reply #10 on: October 21, 2012, 04:32:44 AM »

I agree with Tom and the lack of selectivity and a real front-end in an SDR.
Antenna separation at HF frequencies is going to be a real challenge. There is an expensive solution offered by Array Solutions and bandpass filters.
Look at a Ham contest station where there are many radios in use and running QRO. What do they do to protect from cross-band interference? Are they always using 'top-of-the-line' radios??

Fred
Logged
K7KBN
Member

Posts: 2814




Ignore
« Reply #11 on: October 21, 2012, 08:21:22 AM »

Navy shipboard communications before the advent of communications satellites involved transmitting and receiving CW, using R-390 receivers and HF transmitters that might only be a couple hundred feet apart and not cross-polarized.  Transmitter powers were at least 500 watts most of the time.

The R-390 front end was essentially bulletproof.  Even with the overload, we operators could easily hear the other station "break" us in between dits and dahs.

And this was on a carrier! Flight deck area was over 5 acres.  The destroyers and other ships in company had the same communications requirements but with a whole lot less real estate to stick antennas.
Logged

73
Pat K7KBN
CWO4 USNR Ret.
KC4MOP
Member

Posts: 743




Ignore
« Reply #12 on: October 23, 2012, 03:29:53 AM »

In the UHF world isolation is so easy. Just 10 feet separation between vert antennas on the same tower leg mounted perfectly in-line with each other. I forget the dB isolation.
And the real radios can deal with close by HF. Separating antennas is a lost cause on HF.
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!