[hpsdr] Janus Lives!!

Wed Jul 5 08:17:18 PDT 2006

Phil,

My apologies.  Your measurement method sounds absolutely correct.  The total
system gain in front of the ADC at the time of calibration with raise or
lower the RELATIVE noise floor accordingly.  If you add 10dB more gain in
front, the noise floor will drop another 10dB from where it is now.  I think
the thing that confused me was that either I missed the measurement
conditions are they were not stated.  This is NOT a measurement of system NF
as you stated below but simply shows the relative noise floor of the ADC
under the calibration conditions.  The 500 Hz bandwidth number is not
meaningful unless the calibrated radio is attached with a dummy load.

By the way, the AK5394A part should have a noise figure of 30 dB according
to the manufacturers data sheet, given an unweighted (A weighted less 3dB)
dynamic range of 120 dB in 20 kHz bandwidth.  The CS5381 should be around 33
dB under the same conditions.  The input amplifiers will also have an effect
on system NF that has to be accounted for as well.

Good work.

73,
Gerald

Gerald Youngblood, K5SDR
FlexRadio Systems
Ph: 512-535-5266
Fax: 512-233-5143
Email: gerald at flex-radio.com
Web: www.flex-radio.com

> -----Original Message-----
> From: pvharman at arach.net.au [mailto:pvharman at arach.net.au] 
> Sent: Tuesday, July 04, 2006 3:51 AM
> To: Gerald Youngblood
> Cc: 'High Performance Software Defined Radio Discussion List'
> Subject: RE: [hpsdr] Janus Lives!!
> 
> Hi Gerald,
> 
> If you use kTB to calcluate the noise then this assumes a 1 
> ohm source resistance. In the case of the Janus the input 
> impedance is ~9k assuming it is being driven from an open 
> circuit (which is how I did the measurement).
> 
> Measuing the Janus noise floor with the SDR1000 hardware 
> gives the system noise figure not the Janus Noise Figure, 
> which is dependant on the gain of the SDR1000 and its Noise Figure
> 
> 
> The way I made this measurement was to connect my SDR 
> hardware to the Janus, connect a -50dBm signal to the antenna 
> and calibrate PowerSDR. I then removed the input to Janus and 
> this gives the -160dBm noise floor figure. 
> 
> If this is incorrect then more than happy to learn how to do 
> it correctly. 
> 
> Have a great 4th July up there!
> 
> 73's Phil...VK6APH 
> 
> 
> 
> 
> Quoting Gerald Youngblood <gerald at flex-radio.com>:
> 
> > Hi Phil,
> > 
> > The spectrum looks very clean.  Great work!  
> > 
> > There appears to be one problem with the measurements though.  The 
> > noise floor in the screen shots cannot be accurately 
> calibrated since 
> > the 500 Hz RMS noise measures -159 dBm.  If you subtract 26 
> dB for the 
> > 500 Hz bandwidth, you get a noise floor of -185 dBm/rt Hz.  This is 
> > below kTB of
> > -174 dBm/rt Hz.  That would equate to a negative NF 
> (-11dB).  I would 
> > be interested in seeing the 500 Hz bandwidth shot after calibration 
> > with the radio attached.
> > 
> > 73,
> > Gerald
> > 
> > Gerald Youngblood, K5SDR
> > FlexRadio Systems
> > Ph: 512-535-5266
> > Fax: 512-233-5143
> > Email: gerald at flex-radio.com
> > Web: www.flex-radio.com
> >  
> > 
> > > -----Original Message-----
> > > From: hpsdr-bounces at hpsdr.org
> > > [mailto:hpsdr-bounces at hpsdr.org] On Behalf Of Phil Harman
> > > Sent: Monday, July 03, 2006 7:45 AM
> > > To: High Performance Software Defined Radio Discussion List
> > > Subject: [hpsdr] Janus Lives!!
> > > 
> > > ***** High Performance Software Defined Radio Discussion 
> List *****
> > > 
> > > All,
> > > 
> > > I've just completed the CPLD software for the Alpha Janus 
> board and 
> > > the results are spectacular.
> > > 
> > > The image below is of the Janus sampling at 48kHz.  As 
> you can see 
> > > the noise floor is at a spectacular -159dBm (it actually hovers 
> > > around -160dBm but that is when I captured the screen). 
> As you can 
> > > see there are absolutly no spurs in the passband.
> > > 
> > > The file name is "Janus noise floor.jpg"  for those that have a 
> > > problem with getting files from www.hamsdr.com
> > > 
> > > http://www.hamsdr.com/personaldirectory.aspx?id=348
> > > 
> > > The second screen shot is at 48kHz in spectrum mode. 
> Again no spurs 
> > > and the signal at 0Hz is -120dBm. This drops below the band noise 
> > > when I connect the SDR1000 hardware.
> > > 
> > > http://www.hamsdr.com/personaldirectory.aspx?id=349
> > > 
> > > The file name is "Janus spectrum.jpg"
> > > 
> > > I still have some more tests to do as well as testing  at 
> 96/192kHz 
> > > but so far all is looking great.
> > > 
> > > There are a small number of minor changes to the 
> prototype PCB which 
> > > will be addressed shortly.  My sincere thanks to Lyle 
> KK7P for doing 
> > > such a great job on the PCB design, Bill KD5TFD for the PowerSDR 
> > > software modifictions so Janus will work with it and help 
> with the 
> > > Verilog coding and all those who helped me up the FPGA learning 
> > > curve.
> > > 
> > > Onwards to OZY!!
> > > 
> > > 73's Phil...VK6APH
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > --
> > > No virus found in this outgoing message.
> > > Checked by AVG Free Edition.
> > > Version: 7.1.394 / Virus Database: 268.9.8/380 - Release
> > > Date: 30/06/2006
> > > 
> > > _______________________________________________
> > > HPSDR Discussion List
> > > To post msg: hpsdr at hpsdr.org
> > > Subscription help: 
> > > http://lists.hpsdr.org/listinfo.cgi/hpsdr-hpsdr.org
> > > HPSDR web page: http://hpsdr.org
> > > Archives: http://lists.hpsdr.org/pipermail/hpsdr-hpsdr.org/
> > > 
> > 
> > 
> 
> 
> 
> 

 1152112638.0