[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Linrad] Re: Linrad/sdr-IQ level adkustement?

Hi Dom,

> I ran some experiments yesterday and this morning , and I have dumb 
> question.
> 1) I 1st verified that the hardware really sets the noise floor. No pb I get 
> 18db noise floor increase when the preamplifier is switched off.

> To know more , I also checked whats going on when I switch the SDR-IQ input 
> attenuator :  the preampON/preamp OFF ratio immediately goes on the low 
> side..
> Same test with the SDR-IQ IF gain : even with gain set to0db , preamp noise 
> ratio stays 18 db.
SDR-IQ IF gain? SpectraVue or Windows? Or do you mean any of the the gain 
settings in Linrad?

> 2) Check levels on Winrad :
> There was a contest this w.e , with the original settings , I got heavy 
> saturation on a local friend on Linrad , when my normal rig behaved OK 
> (TS850+tvter). So probably my level is too high.The S meter is at S9+10 with 
> red color on the noise level , I suppose it is not good at all?
No. That is far from reasonable.

> Question : the only way to get the rigth without killing sensitivity is to 
> play with the IF SDR-IQ attenuator, I'm not sure the gain box ( associated 
> with SDR-IQ QRG setting) is effective on the IF gain , someone could confirm 
> pse?On which gain is this control effective?
> If it is not the case , where can I set this? ( Maybe one of the parameters 
> like CICxx?)

For a receive system to have a good sensitivity at the same time as it
has a good dynamic range you must set the gain correctly in all the

The easiest way to optimize is to set too much gain in all stages
as the starting point. Then use preamp on/off as your test signal.
18 dB as you have is fine.

There are three volume controls in the SDR-IQ. Two on the input, both
controlled by the same box as the frequency in Linrad. There is also
an output gain control, "RCF output shift". 

The AD6620 produces 24 bit data but the SDR-14 only sends 16 bits to
the PC. By setting RCF output shift to zero you would send the 16 
lowest bits and thereby get a very high noise floor in Linrad.

Then check by how much the noise floor drops when you unplug the
input cable from the SDR-IQ with the preamplifier in off position.
In case you see more than a 6 dB change, the gain is too high into
the SDR-14. Use the attenuator to reduce the gain until you see
no more than a 6 dB change. There is a price to pay for this.
The noise floors would be like this in linear power scale:

SDR-IQ nothing connected:              1.0
SDR-IQ all connected, preamp off:      4.0
Noise from converter, preamp off       3.0
SDR-IQ all connected, preamp on:       3*63=189

After this you should find that your preamp on/off is not quite
18 dB any more, it should be 189/4 = 16.8 dB rather than 189/3 = 18 dB.
Assuming the SDR-IQ is the point where dynamic range is most critical
it is justified to loose 0.02 dB on your total system noise figure 
by having 1/189 of the total noise floor originating in the SDR-IQ.

Rather than attenuating the SDR-IQ input it would be a good idea
to reduce the gain somewhere by use of noiseless feedback. There is
a risk that the dynamic range of the earlier stages would not be high
enough to make them capable of delivering intermodulation-free signals
through an attenuator into the SDR-IQ.

Once you have the correct input level, best is to use a signal 
generator to send a signal to the SDR-IQ and turn the level up
until the red LED turns on. Set RCF output shift to maximum (=7)
(you may edit par_sdr14) and press A on the normal processing 
screen of Linrad. You might find then that the amplitude margin
of "A/D" (data from SDR-IQ) is 20 dB. With shift = 7 you would
use the highest 16 bits only but with 20 dB margin there are 3
bits that never become used (6 dB per bit.)

That would mean that it would be perfectly safe to set the RCF 
output shift to 4. Do that, then check the noise floor with
preamplifier on and off. You may find that the difference is
quite a bit larger than expected. That would be because the
noise floor of the SDR-IQ is actually a bit below the lowest
bit you allow through the USB interface. That will add a small
amount of quantization noise.

The SDR-IQ is 16 bit on the USB interface so if you want to allow 
really strong signal in the visible passband the above is
the best you can do.

In case the interferer is outside the visible passband you might 
try an RCF output shift of 3. Strong signals taht nearly saturate
the SDR-IQ would then saturate the USB (and Linrad) by something 
like 4 dB.

I hope the above helps:-)


  Leif / SM5BSZ

You received this message because you are subscribed to the Google Groups "Linrad" group.
To post to this group, send email to linrad@xxxxxxxxxxxxxxxx
To unsubscribe from this group, send email to linrad+unsubscribe@xxxxxxxxxxxxxxxx
For more options, visit this group at http://groups.google.com/group/linrad?hl=en