Re: HiQSDR Noise floor/sensitivity
|
Hello Guenter,
Many thanks for your elaborate answer. It is clear that the signal indication seems fine, close to what you observed on the DL1CC and your home made board (well done BTW!). The noise floor on the other hand is much lower than what I observed. So it seems that there is indeed something wrong with my board. I do use the low noise regulators, etc, etc. so either the parallel decoupling capacitors are broken, or perhaps too much noise from the switch mode power supply?? I will try to replace the power supply first.
Tnx, es 73!
Peter Op 17 nov. 2023 17:21 schreef guenter wilfrid richter <dl7la@...>:
oops, Peter,
(the wrong finger here too fast)
results (continued from prior post) for indicated noise levels (S-Meter setting 1 sec)
3k Bandwidth on both sets, -80 dBm IN showing displayed outputs of -79 dBm on DB1CC's board, -84 dBm on mine
frequency / MHz ???????????? DB1CC????????? DL7LA
1??????????????????????????????????? -116???????????????? -104
4??????????????????????????????????? -111??????????????? -112
10????????????????????????????????? -110??????????????? -114
30????????????????????????????????? -104??????????????? -108
50????????????????????????????????? --98???????????????? -110
taking in account the different sensitivites -79 to -84 one can see that even our homebuilt solution performs astonishing well. It consist of a double silver-clad PCB, mainly ground areas on both sides, soldered VIAS into holes and power rails with hookup wire. No comparison to Helmut's board using four layers and low noise regulators. Maybe there is noise originating in the power chain (defunct regulator or else) or spurious oscillation somwhere. Additional ferrites might help.
The noise figures due to ADC processing were discussed in an early stage but this was on the now archived platform when this group was on yahoo. There is however a primer to the topic on Analog Device's web, look for an article 'adc-input-noise.html'.
From a HAM's view these figures are rather high indeed, but explainable.
Connecting my HP-348 noise (ENR 15 dB) source e.g. will not noticeably increase display level;? only the hissing noise will just change slightly.
I might add that my homebult solution needed quite a lot of ferrite beads glued to discrete locations on both sides of the PCB to damp out parasitic currents and other spurs.
As to the 10 dB positive setting: the attenuator will stay at 0 dB, this feature, when activated, pulls pin 5 on connector X1 of Helmut's board to H, allowing an external LNA to be switched into the signal path before the input.
Good luck...
Guenter DL7LA
|
