[NLRS] Path loss calculations
Mike A. King - KM0T
Mike A. King - KM0T" <[email protected]
Fri, 5 Mar 2004 09:41:07 -0600
Hi all, Thanks Gary for the equations.
The last few nights I have been working on a spread sheet modeled after the
equations in the UHF/Microwave manual to calculate free space loss, water
vapor and oxygen attenuation to get the total path loss. I looked over many
sites as well for the equations trying to determine if they are somewhat
universal. It appeared that the total path loss equation - was fairly
consistant.
The only issues left are detremination of attenuation from humidity very
accuratly, so the references that Gary indicated may be of some help, I will
check into that.
However, the changes to the total path loss from water vapor based on say
95% relative humidity from one chart to another at 24 ghz will not change
things a whole lot.
As I ran my calcs, it was apparent (if the ARRL slug of equations for total
system preformance with the RX noise power and and antenna power gain are
correct) that there was a large difference between what came out for a
predicted Signal to Noise ratio and what the real world try at the 222 km 24
ghz qso distance produced.
I began to wonder if indeed I had been attempting a tropospheric scatter
contact in lieu of what I thought might have been a true "Line of Sight"
qso. (this was based on Gene being at 1950' ASL and me being at 1450')
I dug thru my stuff and came up with the article "will this rover make the
path" from Microwave Update 2000 proceedings article. In there it states
that the total path loss calculation was for only true LOS contacts. If it
was not LOS, then obstructions and distances too then had to be taken into
account from both sides. This then results in a "scattering angle" and then
the common area between for the the scattering to take place to make the
contact.
This of course is pretty tough to model with a simple equation, so the
article called out an old dos program that exists to calculate all this.
Bob Atkins - KA1GT wrote it and one can input all the data needed for
obstructions, their distances from the station and the heights of the
obstructions. The program also calls out for antenna gain, power, NF,
feedline losses, antenna temp, etc.
The link to download the ms dos program is
www.bobatkins.com/radio/scatter.html but it appears to be broken. I do have
it if someone wants it.
When I inputted all my data, instead of getting a +40 dB or S/N ratio from
my ARRL equations, I got around 17 dB. Now, this program indicates that it
does not account for water vapor losses. Just the freespace and
obstruction, troposcatter losses.
Now, I figured that we should have been easy to do CW with that, as SSB
comfortable communication appears from what I have read occures around 10db
S/N.
So, I am thinking of looking at the results from the MSdos program, then
adding the water vapor attenuation and see how close I get to say 0 db to -5
dB S/N. As we were hearing each other sporadically, knew they where there
type of single levels, but could not copy anything of substance. Would -5
or so dB S/N be a good approximation of that?
The one last factor that I can attribute to this, and I cant find a way to
put an numeric value on it is attenuation from HAZE.
The day we did the QSO it was Hazy as all get out on both sides of the path.
>From what I have read that this is an additional attenuation factor at 24
Ghz, so this would have to be added to free space, oxygen and water vapor
(humidity) losses.
The ARRL UHF/Microwave book talks about fog attenuation and haze is similar,
but not the same thing I think. If it could be modeled as "light fog",
great, but the data in the UHF/Microwave manual is sketcy at best and hard
to interpet where one needs to figure the numbers.
So, the question is IF that was the final nail in the coffin for my 24 ghz
attempt, would it have attributed for a number db of S/N?, even more than
100% RH attenuation figures? Good question. More information required to
be sure.
In addition, Matt Reilly - KB1VC has this plot path server where you plug in
the lat/long of both locations, the heights ASL at both and it is supposed
to kick back a GIF file on a 4/3 earth curvature showing all obstruction
heights between the path. I have tried to do this a few time, but never got
anything back. Anyone ever have luck with this? This is the link...
http://users.rcn.com/acreilly/
So I modeled my obstructions with Topo USA and put my data in from there on
the ms-dos program. It appeared that it was almost a knief edge diffraction
propagation issue that I was dealing with as there was one hill inbetween us
that was the critcle obstruction. So perhaps I need to look there for
equations that may relate to that. Apprently though, KB1VC had this
programed in the old dos program.
So, thats where Im at with it to date. I will keep working on it as time
premits. Thanks for any input.
73
Mike- KM0T
----- Original Message -----
From: <[email protected]>
To: <[email protected]>
Sent: Thursday, March 04, 2004 2:50 AM
Subject: [NLRS] Path loss calculations
> Hi all,
>
> Mike, KM0T, asked about path loss calculations
> that included humidity, antenna gain and NF in the
> calculation.
>
> I have been using some info that I found either in
> the old QST column "The New Frontier" that was
> published in the mid 80's, or the info was in the
> ARRL UHF/Microwave Experimenters Manual.
> Both would be by the same author, Bob Atkins, KA1G.
> I wrote a calculator program to do the calculations.
>
> There are some assumptions made here but the results
> should be accurate enough to give us an idea of what
> to expect over a given path.
>
> Here is how it works:
>
> Transmitted signal:
>
> 1) FSPL = 36.6 + 20 Log (Frequency in MHz) + 20 Log (distance in Miles)
> the result is in dB. FSPL is free space path loss.
>
> 2) EIRP = Pt + Ga -Lf
> EIRP- effective isotropic radiated power
> Pt-- Transmitted power in dBm
> Ga--Transmit Antenna gain in dBi
> Lf--Transmitter total feed line loss in dB.
>
> 3) Pr = EIRP - FSPL
> Pr-- total power of the received signal in dBm
>
> Receiver sensitivity:
>
> 4) Convert NF to Temp in degrees Kelvin
> T(�K) = Tr = 290�K [ 10^(NF in dB / 10) - 1]
>
> 5) Ts = Ta + (Lr - 1)* 290 +Lr * Tr
> Ts--receive system Noise Temperature in �K
> Ta--Antenna Noise Temperature (here assumed to be 290 �K)
> This could be lower if the antenna is looking at cold sky.
> Lr- Receive feed line loss expressed as a RATIO, NOT in dB.
> Tr--Receiver noise temperature in �K. See #4 for conversion
> from NF in dB.
>
> 6) Rs = 10 Log ( receiver bandwidth in Hz) + 10 Log ( Ts) -198.6
> Sensitivity is in dBm for a 0 dB Signal to Noise Ratio.
>
> 7) SNR = Pr (from #3) - Rs - Gr (receive antenna gain in dBi)
> SNR will tell you how much the received signal is above
> or below the receiver noise floor.
>
> 8) Humidity. There is a chart for 24 GHz in QST June 1985
> " The New Frontier column".
> Convert the distance in miles to kilometers.
> Find the excess loss per kM for the air Temperature
> and relative humidity.
> Multiply the loss per kM by the total path length in kM.
> Subtract this value from the SNR.
>
> I had been using a value of 0.55 dB/kM for 100 % Relative
> Humidity at 35 �C. the values in the QST chart are
> slightly less.
> It can be surprising how fast the loss can added up on
> the longer paths.
>
> Please let me know if you find any errors or have any questions.
> I hope it will be useful to you.
>
> 73, Gary WB0LJC
>
>
>
> ________________________________________________________________
> The best thing to hit the Internet in years - Juno SpeedBand!
> Surf the Web up to FIVE TIMES FASTER!
> Only $14.95/ month - visit www.juno.com to sign up today!
> _______________________________________________
> NLRS mailing list
> [email protected]
> http://mailman.qth.net/mailman/listinfo/nlrs