On Wed, 2005-12-21 at 23:02 -0800, Matt Ettus wrote:
> Lee Patton wrote:
> > I don't understand why the gain range returned by the DBSRx ranges from
> > 0 to 104 when the PGA gives only 0 to 20 dB. Obviously, the answer lies
> > in db_dbs_rx.py, but I haven't a clue as to what the code is really
> > doing. Can anyone (Eric) explain? I read about Registers 2/3 in the
> > AD9862 data sheet, but I couldn't correlate that information with the
> > code.
> >
>
> There are 2 components to gain control in the RX daughterboards:
>
Thanks. Matt. I didn't realize there were variable gain amps on the
daughterboards.
> 1> The PGA gain, which ranges from 0-20 dB in 1 dB steps
>
> 2> The gain control amplifiers on the daughterboard itself.
>
> On the BasicRX there is no amplification, so you just get the PGA
>
> On the dbsrx there are 2 variable gain amps. One is controlled by an
> analog signal from the low-speed DAC output. The other is controlled
> digitally over the I2C port.
>
After looking at the data sheets for all components on the DBSRX, I
gather that the RF LNA and IF low-pass filter gain control, both of
which are in the MAX2118, are the two variable gain amps to which you
are referring. Is that correct?
I quote from the data sheet:
"Each IC includes an LNA with gain control, I and Q
downconverting mixers, and baseband lowpass filters
gain and cutoff frequency control. Together, the RF and
baseband variable gain amplifiers provide more than
79dB of gain control range."
daughterboard gain (79dB) + PGA gain (20dB) = 99dB. The DBSRX returns
104dB. Where does the extra 5dB of programmable gain come from? (It
looks like it is the "dl" variable in your code, but I don't know what
that corresponds to. Please see below)
> On the TVRX there are separate IF and RF AGC amps, both are controlled
> by the low-speed DACs
>
> The FlexRF has a single AGC amp, again controlled by the low-speed DAC.
>
> Matt
>
To minimize noise, I would assume that when setting the gain it is
preferable to fist max out the RF LNA, then the IF gain control, and
finally the PGA. Is that correct?
Would you please add some comments to the following code snippet from
db_dbs_rx.py? After looking at the MAX2118 data sheet, I am starting to
recognize your variables like gc1 and gc2, but I would appreciate your
insight.
181 # Gain setting
182 def _set_dl(self,dl):
183 assert dl == 0 or dl == 1
184 self.dl = dl
185 self._send_reg(4)
186
187 def _set_gc2(self,gc2):
188 assert gc2<32 and gc2>=0
189 self.gc2 = gc2
190 self._send_reg(5)
191
192 def _set_gc1(self,gc1):
193 assert gc1>=0 and gc1<4096
194 self.gc1 = gc1
195 self._u.write_aux_dac(self._which,0,int(gc1))
196
197 def _set_pga(self, pga_gain):
198 assert pga_gain >=0 and pga_gain <=20
199 if(self._which == 0):
200 self._u.set_pga (0, pga_gain)
201 self._u.set_pga (1, pga_gain)
202 else:
203 self._u.set_pga (2, pga_gain)
204 self._u.set_pga (3, pga_gain)
205
206 def gain_range(self):
207 return (0, 104, 1)
208
209 def set_gain(self,gain):
210 if not (gain>=0 and gain<105):
211 raise ValueError, "gain out of range"
212 gc1 = 0
213 gc2 = 0
214 dl = 0
215 pga = 0
216 if gain <56:
217 gc1 = int((-gain*1.85/56 + 2.6)*4096/3.3)
218 gain = 0
219 else:
220 gc1 = 0
221 gain = gain - 56
222 if gain < 24:
223 gc2 = int(round(31 * (1-gain/24)))
224 gain = 0
225 else:
226 gc2 = 0
227 gain = gain - 24
228 if gain >= 4.58:
229 dl = 1
230 gain = gain - 4.58
231 pga = gain
232 print "Gain %f" % pga
233 self._set_gc1(gc1)
234 self._set_gc2(gc2)
235 self._set_dl(dl)
236 self._set_pga(pga)
237
Thanks,
- Lee
_______________________________________________
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
