how to adjust the agc of dvb-s tunner

[yubingz]

i need to design a dvb-s receiver.
but there have a problem about agc control.

most of the dvb-s tunner have 2 agc control point - rf agc and baseband agc.
ang before bb vga there should have a LPF.

then the power estimation of the signal of the ADC should not be include outband of LPF. so if i adjust the rfagc according to the result of power estimation, then LNA should saturate.

is that right? if it is, how to solve this problem?

thank you very much!

 

[tony_lth]

Can't understand your question.

 

[yubingz]

915-2150M----LNA----- LPF---30M----VGA------ADC-----power estimation
......................^----------<-----------^------------<-----------|

OK?

 

[vfone]

The LPF actually is the channel filter (assuming a Direct Conversion receiver).
So, the RF AGC should be active (theoretically) only for signals inside of the channel filter.
But there are situations (adjacent/alternate channel or blocking test) when have to see the performances of the AGC for signals out-of-channel.

 

[yubingz]

yes, if i tunned the tunner to an empty channel or the channel had not signal, the power estimation should output an very low value. so the agc controller would adjust the gain higher. then LNA would been saturated.

 

[kspalla]

LNA can not saturate when there is no channel signal and/or blocker signal what ever you do for AGC.
you need to derive algo with thresholds and set at what signal level which devise should operate AGC.

 

[tony_lth]

The LNA can't saturate. If No input signal, the link gain should be max, you can see the floor noise is raised very high in the IF in SA, compared to nth input to SA. If there are some signal, the floor noise decreased much.

 

[yubingz]

Thanks for your answer!
May i think that the power of signal from LNB should not make LNA saturate, although the gain of LNA was max?

 

[tony_lth]

LNA should be included in LNB.
The signal maybe make LNA saturate, but noise can't

 

[yubingz]

the problem is i measure the power in LPF passband, but adjust the gain in whole rf band.

 

[tony_lth]

that is normal.

 

[yubingz]

normal? so when the power level in LPF passband is very low, the LNA will saturate?

 

[tony_lth]

the problem is i measure the power in LPF passband, but adjust the gain in whole rf band.

I mean this is normal. Normally there is ALC in IF, and adjust gain in RF and IF.
You can use linkbudget tools to analysis which part 1st to saturate, RF or IF? Then you can use some OpAmp to control the AGC logic. Say 1st use RF AGC, if can't reach the goal when RF AGC is limit, then use IF AGC.

 

[kspalla]

Do not forget to do link budget for not only gain but also IP3.

 

[yubingz]

thank you for your answer. i'm not familiar with RF circuit. i'm digital signal processing engineer. so i just try to understand the circuit. pls explain it detail a little? i can't find the anwer in the book.

 

[RCinFLA]

For DBS dvb-s there is seldom issues with strong interference. The receiver must be able to handle excessive LNB gain. DVB-T is a different story and is much more concerned with intermodulation possibilities.

Generally AGC is done to optimize Carrier to Noise ratio throughout the gain reduction. Dynamic range analysis is done to ensure intermod and overload conditions are met for each stage. Your first issue is usually avoiding overload of the ADC's and BB gain reduction is done to avoid this.