[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: R: Downconverter noise


Nice analysis but I have found that few commercial radios have s-meters
anywhere near calibration...the FT-847 I have averages 4 dB per s-unit but
varies from 3 to 5 dB and is very non-linear (tested with a commercial
signal generator - IFR-120).  Somewhere I have the actual measurements in
dBm, but didn't dig that out.  So to be accurate I would suggest one
actually test their receiver s-meter against a calibrated signal.  Then use
that information with your analysis.

For example the FT-847 has been tested for NF on an HP noise figure meter
and found to be 14 dB with the internal preamp off, and 8 dB with it on.
Looks like the FT-736R beats it (which comes as no surprize).

Ed - AL7EB 

At 06:37 AM 5/28/2003 +0200, i8cvs wrote:
>----- Original Message -----
>From: Scott Townley <nx7u@arrl.net>
>To: Jon Ogden <na9d@speakeasy.net>; Howard Long <howard@howardlong.com>
>Cc: Phil <phil@spiderweb.com.au>; <amsat-bb@AMSAT.Org>
>Sent: Tuesday, May 27, 2003 2:46 PM
>Subject: Re: [amsat-bb] Downconverter noise
>> Available noise power output from any active device can be stated as:
>> FGkTB
>> where F is the noise figure referred to the input (which is how it's
>> usually specified), and G is the available gain.
>Hi Scott,
>I aegree with you but in the above equation F is the noise factor and if the
>noise figure is NF=1 dB than F= 10E(1/10) = 1.258 in factor
>T in the equation is the equivalent noise temperature of a 50 ohm resistor
>at room temperature of 290 kelvin because the input of a device under test
>in the lab must be terminated in to a 50 ohm impedance of a generator so
>that the noise produced by any active device i.e. the downconverter  in
>this case with no antenna connected and open input is a wrong procedure.
>You got a correct available noise power of -102 dBm because you uses
>290 kelvin for T in the above formula and so your input is terminated in to
>a 50 ohm resistor at room temperature.
>> For the AIDC-series, assuming available gain=37dB and the noise figure is
>> ~1dB (but you can see from the equation that a with a gain of 37dB, that
>> even several tenths variation in the NF makes no difference) the available
>> noise power is -102dBm (2.5kHz RX bandwidth).
>I aegree with you and infact if you make your calculations using the above
>equation introducing:
>F= 1.258
>G= 10E(37/10) = 5011.87 time in power
>k= 1.38 x 10E -23
>T= 290 kelvin
>B= 2500 Hz
>Than Pn1 = FGkTB= -101.99 dBm or about -102 dBm
>This is the available output noise power or the output noise floor of
>the AIDC downconverter in the lab when its input is connected to a
>source of 50 ohm at room temperature of 290 kelvin.
>> What your S-meter does with this obviously varies, but -102dBm is
>> significantly higher than your RX noise floor!
>I aegree:
>For example the SSB sensitivity of a RX like a FT 736 in to a BW of 2.5 KHz
>is -15 dB uV for a (S+N)/N= 12 dB
>The above sensitivity can be converted very easily in a noise figure
>NF= 6.27 dB and consequently the computed input noise floor is -133.73 dBm
>when its input is connected to a source of 50 ohm at room temperature
>of 290 kelvin like an attenuator or the output of a downconverter.
>In this conditions, if  the above AIDC downconverter has the input
>terminated to a 50 ohm resistor and if its output is directly connected
>to the  FT 736 input than the S meter will show a noise level that is
>the difference between -101.99 - ( -133.73)= 31,74 dB
>The standardization of the S meter readings for frequency bands above
>144 MHz states that the S-9 reference level is -93 dBm available signal
>power at the receiver input.
>If a noise level power of -101.99 dBm is injected in a receiver with a well
>calibrated S-meter than the signal level reading must be
>-101.99 - (- 93)= -8.99 dB belove the S9 level .
>Since the S-point standardization recommended by IARU to manufacturers
>of equipments is 6 dB for the S-point, than the noise level for a -101.99
>dBm noise level input power will be between S7 and S8
>> With the dish fitted, the available noise power should increase slightly,
>> as antenna/sky noise have been added.  But for a "good" system it will not
>> be detectable on your S-meter.
>I disaegree here:
>When the downconverter input is connected to the feed  than the 50 ohm
>resistor at 290 kelvin in the lab is removed and the downconverter input
>sees only the equivalent noise temperature of the antenna/sky wich is
>generally lover than 290 kelvin so that the antenna/sky noise is added but
>the 50 ohm resistor equivalent noise temperature is subtracted and the power
>noise should decrease, but how much ?
>To better understand what happens i suggest to convert the above formula
>Pn1=FGkTB in the following one:
>Pn2=Gk (T1 + T2) B  where
>T1= equivalent noise temperature of the AIDC downconverter
>T2= equivalent noise temperature of downconverter input source
>Since NF= 1 dB and F= 1.258 than T1= (1.258-1)x 290 = 75 kelvin
>T2= 290 kelvin if the downconverter input is terminated in a
>        50 ohm resistor in the lab.
>T2= 20 kelvin about at 2400 MHz if the dish is looking at the cold
>        sky depending on feed spillover.
>Replacing the numbars the output noise level of the system is still -101.99
>dBm if the downconverter input is terminated in a 50 ohm resistor but it
>decreases to -107.84 dBm if the input is connected to the antenna looking
>at the cold sky and it make an improvement in sensitivity of
>-101.99 - ( -107.84 )= 5.85 dB
>> All that noise is why on some receivers, it's not a bad idea to pad down
>> the IF.  Remember that the bandlimiting devices in your receiver are
>> *after* the RF front end--so if you're the RF amp/1st mixer then B is much
>> much higher, and the noise power seen is proportionally much higher
>> too.  Too much noise power leads to saturation of the mixer and lots of
>> other bad things :-)
>73" de i8CVS Domenico
>Sent via amsat-bb@amsat.org. Opinions expressed are those of the author.
>Not an AMSAT member? Join now to support the amateur satellite program!
>To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org

Sent via amsat-bb@amsat.org. Opinions expressed are those of the author.
Not an AMSAT member? Join now to support the amateur satellite program!
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org