QO-100


Andy G4JNT
 

Are the value for G/T and transponder gain for the QO-100 narrowband transponder still Tippy-Toppy-Secret?
Or perhaps could someone be persuaded to release the figures?


DD1US
 

Hi Andy,

 

I am afraid they are still secret and will remain secret.

 

Kind regards

 

Matthias

 

www.dd1us.de

 

 

Von: UKMicrowaves@groups.io <UKMicrowaves@groups.io> Im Auftrag von Andy G4JNT
Gesendet: Mittwoch, 2. Juni 2021 16:08
An: UK Microwaves groups.io <ukmicrowaves@groups.io>
Betreff: [UKMicrowaves] QO-100

 

Are the value for G/T and transponder gain for the QO-100 narrowband transponder still Tippy-Toppy-Secret?

Or perhaps could someone be persuaded to release the figures?

 

Andy

 


Mike Willis
 

I thought you had measured it? Don't you believe yourself?

Transmitting a known EIRP from a standard horn and measuring the signal above the noise floor of the NB transponder gives you the answer once you calibrate for your receive G/T. With the transponder noise floor 10 dB or so above the noise that correction is already fairly minor. Make a few assumptions about the noise temperature of the earth and atmosphere. We can work out the compensation for our own RX G/T or or less accurately from solar transits. About 24 dB in my case. That should be able to get the figures within a dB or so. It will vary by more than that over the coverage area.
 
--
Mike G0MJW


Andy G4JNT
 

It only works if you have a receive system that is transponder noise limited.  Mine isn't.   I think local trees degrade the signal a bit as my 0.6m dish is 10dB below the Goonhilly websdr.    I could use that as the measurement , but so far haven't tried a decent calibrated measurement with it.   Furthermore, need to make a standard gain horn for 2.4GHz, (simple enough with tinplate from an old engine oil can :-)

A reasonable estimate of G/T can't be too far off.    Assume 1dB noise figure for the satellite receive - pretty typical of LNAs with a bit of preceding loss.   That's seeing mostly Earth's surface with a substantial amount of ocean reflection from cold space, so the antenna temperature will be perhaps 150 - 200K.  On -L-Band in my MSS days, IIRC we used 200K.      So a total Noise Temp, call it 250K
A global coverage antenna from GEO will have to have 18 degree coverage, so if made efficiently should be managing 19.5 - 20dBi

Thus the G/T can't be too far away from -4dB/K.   The transponder gain is of rather more interest, and there are too many variables to get an accurate estimate of that, not least really accurate RMS noise measurement in a precisely know bandwidth
Few SDRs, or more to the point FFT software, give an accurate noise bandwidth/effective bin size.    Some don't even declare what window type they use, so absolute S/N measurement can't really be much better than perhaps 2- 3dB accurate

But your comment has reinspired a bit of interest, so I'll see if I've got the materials to make a SGH for 2.4GHz

 


On Sat, 5 Jun 2021 at 08:11, Mike Willis <willis.mj@...> wrote:
I thought you had measured it? Don't you believe yourself?

Transmitting a known EIRP from a standard horn and measuring the signal above the noise floor of the NB transponder gives you the answer once you calibrate for your receive G/T. With the transponder noise floor 10 dB or so above the noise that correction is already fairly minor. Make a few assumptions about the noise temperature of the earth and atmosphere. We can work out the compensation for our own RX G/T or or less accurately from solar transits. About 24 dB in my case. That should be able to get the figures within a dB or so. It will vary by more than that over the coverage area.
 
--
Mike G0MJW


Mike Willis
 

I am not sure that is true but certainly the errors are larger with a smaller dish. It's hardly beyond your skill set to produce a measurement grade receive system. No excuses accepted from JNT for lack of measurement rigour.

I would say a G/T of -4 a little on the high side. Total noise temperature is going to be higher due to waveguide/coax losses and switches, filters etc. A few 100k, say 500k, 27dBk. Hence a G/T in the maybe -7 dB/k range.

In Solar Transit I saw 10.5 dB of solar lift such that the peak solar level came to be within a fraction of a dB of the normal transponder noise floor. It happens the sun is just a little larger than my antenna beamwidth so we can say the transponder noise floor is around the same power as the solar flux. (QO100 NB transponder is emitting as much noise as the sun, who knew?)

For a quiet sun, that can be estimated in solar flux units SFU. One SFU equals 10^-22 W/sqm/Hz. (i.e. -220dBW/sqm), There is an approximation to the Solar Flux for a quiet sun:

S(f) = 26.4 + 12.4f + 1.11 f^2 in SFU for f =>1 and f <=20 GHz.

For 10.5 GHz that's 280 which equates to -195 dBW/sqm

There is an approximation for G/T for a dish where you take the noise uplift N (dB):

G/T = 10 * log [ 4 π k (10N/10 - 1) / ( λ^2 S) ] . dB/K

For linear polarisation we lose half the solar flux so the G/T needs to be corrected and will be 3 dB higher. 

For me that works out at 21.9 dB, not 24 which is probably a more realistic result. There are a few weaknesses / major conceptual errors in this calculation. Ignores atmospheric losses and noise etc.

There is another method here https://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=4128&context=smallsat which properly takes into account the antenna beam and atmospheric losses. That comes out at 22.1 dB so not a huge difference. It makes a much greater difference with larger dishes.

--
Mike G0MJW


Mike Willis
 

Forgot to add, if the transmitted noise power is roughly the same as the solar flux that gives a handle on the gain of the transponder once you do the link budget taking into account the distance, the G/T and the bandwidth.

--
Mike G0MJW


Andy G4JNT
 

I can't make that G/T equation work with your values.
What is  k   ?

And I assume it should read
G/T = 10 * log [ 4 π k (10^(N/10) - 1) / ( λ^2 S) ] . dB/K



On Sat, 5 Jun 2021 at 12:36, Mike Willis <willis.mj@...> wrote:
Forgot to add, if the transmitted noise power is roughly the same as the solar flux that gives a handle on the gain of the transponder once you do the link budget taking into account the distance, the G/T and the bandwidth.

--
Mike G0MJW


Andy G4JNT
 

I'm also puzzled as to how that can work with arbitrary antenna beamwidth
If the beamwidth is much larger than the sun's disc diameter then only a portion of the solar flux is contributimg,  A very narrow beamwidth, ie << 0.6° will see all the  sun

I shall go and sit in the garden and either :

   A) Try to work it all out from first principles   or  
   B)  Realise the grass needs cutting and really ought to be doing that instead



On Sat, 5 Jun 2021 at 13:02, Andy Talbot <andy.g4jnt@...> wrote:
I can't make that G/T equation work with your values.
What is  k   ?

And I assume it should read
G/T = 10 * log [ 4 π k (10^(N/10) - 1) / ( λ^2 S) ] . dB/K



On Sat, 5 Jun 2021 at 12:36, Mike Willis <willis.mj@...> wrote:
Forgot to add, if the transmitted noise power is roughly the same as the solar flux that gives a handle on the gain of the transponder once you do the link budget taking into account the distance, the G/T and the bandwidth.

--
Mike G0MJW


Andy G4JNT
 

Worked out the equation from first principles now, and it gives the same as Mike's example

k is Boltzman's constant, 1.38E-23
and S in the equation is the actual solar flux in Watts per m^2/Hz,  not the S(f)  from the first equation



On Sat, 5 Jun 2021 at 13:59, Andy Talbot <andy.g4jnt@...> wrote:
I'm also puzzled as to how that can work with arbitrary antenna beamwidth
If the beamwidth is much larger than the sun's disc diameter then only a portion of the solar flux is contributimg,  A very narrow beamwidth, ie << 0.6° will see all the  sun

I shall go and sit in the garden and either :

   A) Try to work it all out from first principles   or  
   B)  Realise the grass needs cutting and really ought to be doing that instead



On Sat, 5 Jun 2021 at 13:02, Andy Talbot <andy.g4jnt@...> wrote:
I can't make that G/T equation work with your values.
What is  k   ?

And I assume it should read
G/T = 10 * log [ 4 π k (10^(N/10) - 1) / ( λ^2 S) ] . dB/K



On Sat, 5 Jun 2021 at 12:36, Mike Willis <willis.mj@...> wrote:
Forgot to add, if the transmitted noise power is roughly the same as the solar flux that gives a handle on the gain of the transponder once you do the link budget taking into account the distance, the G/T and the bandwidth.

--
Mike G0MJW


Andy G4JNT
 

An instinctive feeling is nagging me that the fraction of the sun in the antenna beamwidth doesn't matter when working things out this way.   But it's hard to quantify why, except that using a solar flux density and an antenna aperture assumes nothing about the source, and that it's ideally a point source

SO applicable to small antennas, but what about narrow ones which are , to all intents and purposes, in the sun's near field.
(I know that sounds a bit daft at first sight, but think about it)



On Sat, 5 Jun 2021 at 13:59, Andy Talbot <andy.g4jnt@...> wrote:
I'm also puzzled as to how that can work with arbitrary antenna beamwidth
If the beamwidth is much larger than the sun's disc diameter then only a portion of the solar flux is contributimg,  A very narrow beamwidth, ie << 0.6° will see all the  sun

I shall go and sit in the garden and either :

   A) Try to work it all out from first principles   or  
   B)  Realise the grass needs cutting and really ought to be doing that instead



On Sat, 5 Jun 2021 at 13:02, Andy Talbot <andy.g4jnt@...> wrote:
I can't make that G/T equation work with your values.
What is  k   ?

And I assume it should read
G/T = 10 * log [ 4 π k (10^(N/10) - 1) / ( λ^2 S) ] . dB/K



On Sat, 5 Jun 2021 at 12:36, Mike Willis <willis.mj@...> wrote:
Forgot to add, if the transmitted noise power is roughly the same as the solar flux that gives a handle on the gain of the transponder once you do the link budget taking into account the distance, the G/T and the bandwidth.

--
Mike G0MJW


Dave G8KHU
 

<< ...in the sun's near field. >>

Not sure I completely agree with that Andy.

If the sun is a single source driving a sun diameter antenna, then yes indeed.

If you consider the sun to be an enormous array of coherent radiaters, then again yes.

However if the sun is considered as an enourmous array of non-coherent, independant small radiators then wouldn't the effective near/Fresnel/far field boundaries would be determined by the geometery of the individual radiator rather than the array geometry?

Dasve G8KHU


Andy G4JNT
 

Not sure I do either - in the traditional    2.D²/λ   sense
But when the transmitter is not a point source (not diffraction limited, I think, in optical terms)  there must be some name for it.
The sun is 0.6° wide, so any antenna beamwidth narrower than that gets 'interesting'

 


On Sat, 5 Jun 2021 at 20:17, Dave G8KHU <david@...> wrote:
<< ...in the sun's near field. >>

Not sure I completely agree with that Andy.

If the sun is a single source driving a sun diameter antenna, then yes indeed.

If you consider the sun to be an enormous array of coherent radiaters, then again yes.

However if the sun is considered as an enourmous array of non-coherent, independant small radiators then wouldn't the effective near/Fresnel/far field boundaries would be determined by the geometery of the individual radiator rather than the array geometry?

Dasve G8KHU


Dave G8KHU
 

When it gets very narrow you get pictures like this - it doesn't say at what frequency other than "microwaves"
https://www.spaceanswers.com/space-exploration/celebrating-15-years-of-the-spitzer-space-telescope/

The site for the observatory is here
http://www.ovsa.njit.edu/index.html

and includes schematics on their hardware here which shows it has the operating range of 1 -18 GHz
http://www.ovsa.njit.edu/wiki/index.php/Hardware_Overview#Frontend_System

Dave


Dave G8KHU