[hpsdr] [HPSDR] Hermes now rxing

Wed May 6 07:50:07 PDT 2015

Glen,

If you set the attenuator to 20dB the scale of the display/meters will
change +20dB to compensate so that it still reports an accurate power
measurement at the Rx input. Likewise when you enable the 6m LNA on
ALEX/ANAN, a corresponding value is subtracted from the display. A
-80dBm signal applied to the Rx input will show as -80dBm regardless
of the attenuator setting. As noted earlier, this is exactly how most
modern test and measurement equipment work. Most gear also allows you
to specify a manual offset in the event that you are using an outboard
attenuator or preamp so that you dont have to think about the offset.

The MDS of Hermes is about -145dBm or somewhere around there; if your
band noise is measuring at -120dBm with no attenuation and you
increase the attenuator to 30dB you will push the band noise below the
sensitivity/noise floor of the LT2208 itself and the measured noise
floor will start going up. At that point, -120dBm noise into the
receiver will be -150dBm at the ADC. If the ADC can only measure
-145dBm, PowerSDR will show the noise level as -115dBm.

The purpose of the attenuator is to give you a tool to keep strong
signals from overloading the ADC, that is all. It's good practice to
keep the attenuator set to as high a value as you can while still
giving you enough dynamic range to pick out your desired signals over
band /thermal noise.

Just like with your Hermes Lite vs K3 experiment, the data is all
subjective until you have both calibrated your equipment with a known
signal source and are aware of how each devices metering and DSP
behaves. Elecraft has an inexpensive signal source that may be useful
for what you are doing: http://www.elecraft.com/XG3/xg3.htm This will
be an effective tool to tell you if something is wrong. You say your
20m noise is -117dBm (about S1) which in my location I would find
unusually low. Is that normal for your location? If not there may be
something wrong in the analog Rx path; perhaps a transformer on the
wrong-way round or inadequate current to the ADC driver. There's
fortunately not a whole lot there to debug but it will take the proper
equipment to do it.

73, John K5IT

On Wed, May 6, 2015 at 7:33 AM, MDS Info <info at mds-ham.com> wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
>
> Scott,
>
> This is counter-intuitive to me.  I suspect this is wrong, sorry.  So, I took my Wavetek attenuator and hooked it to the rx input on the Hermes, 20M ssb.
> As the DAT31 is right there before any amplification, the Wavetek is doing the same job.
>
> With the S-ATT set at zero, as I increase the input attenuation (Wavetek), only the signal decreases.  Of course, part of the noise floor is band noise, so some
> lowering of the ultimate noise floor was shown.
>
> The external Wavetek works exactly as I suspected the internal DAT31 should, only decrease the incoming signal (plus noise).
>
> I am concluding that there is some problem with my Hermes.
>
> When I tune to an unused portion of the band, the noise floor shows -117db.  As I increase the attenuation to 30db and beyond, the bottom reaches -120 no more.
>
> 73's
>
> Glen K4KV
>
>
>
> -----Original Message-----
> From: Hpsdr [mailto:hpsdr-bounces at lists.openhpsdr.org]On Behalf Of Scott Traurig
> Sent: Tuesday, May 05, 2015 10:45 PM
> To: Glenn Elmore
> Cc: hpsdr
> Subject: Re: [hpsdr] Hermes now rxing
>
> Perhaps it is more important to realize that the step attenuator is in front of the first receiver amplifier. Because of that the receiver noise figure is going to be increased (worsened) by the amount of step attenuation selected.
>
> 73,
>
> Scott/w-u-2-o
>
>
> On Tue, May 5, 2015 at 6:44 PM, Glenn Elmore <n6gn at sonic.net> wrote:
>>
>> ***** High Performance Software Defined Radio Discussion List *****
>>
>>
>> I would say this is both normal and the desired behavior. It's what a commercial spectrum analyzer does. The reported level is referenced to the antenna/input connector. What happens inside the receiver, other than the resolution/display bandwidth, shouldn't affect this.�� When internal attenuation or preamplification is added or removed, the receiver fixes the reporting to make those changes invisible. The exception to this is that the displayed S Meter reading is reporting the total power within the selected bandwidth. If it's mostly a carrier with high S/N then you get the carrier's level. If it is mostly/all noise then you get the noise power in that bandwidth.�� Attenuation or preamplification within the receiver just changes the sensitivity (noise figure) but doesn't/shouldn't change what is coming in or how it is reported.
>> Having the reporting done this way not only removes the receiver from the measurement but lets one calculate, for example, system noise figure by subtracting 10*log(bw) from the displayed reading on noise and comparing it with KTB in 1 Hz - -174 dBm.�� This is useful in comparing QRN levels to a "quiet" world on HF (which I never see any more!)
>>
>> Glenn n6gn
>>
>>> The noise floor with the S-ATT at zero is very low, probably around -135db
>>> on 20M right now.�� As I raise the S-ATT the noise
>>> floor rises to -110db and the signals disappear.�� I would have thought just
>>> the signals would decrease as I raised the S-ATT!?
>>>
>>> Is this normal behavior?
>>>
>> *****************************
>> _______________________________________________
>> HPSDR Discussion List
>> To post msg: hpsdr at openhpsdr.org
>> Subscription help: http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
>> HPSDR web page: http://openhpsdr.org
>> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>
>
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help: http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/

 1430923807.0