6600 noise levels +10dB higher on my new 6600 vs. my older 6500

K9SO

During testing of my brand new 6600, I see a higher noise floor than I had on my 6500. With no antenna connected (unterminated), a 400Hz CW BW and 0dB gain on the preamps, I have a -110dBm noise floor on the 6600 as compared to -123dBm on the 6500 ( measured on the S meters).

The displayed noise line is lower of course but is consistently almost 10dB higher on the 6600 than the 6500 with no antenna. 

There may be SW issues as I'm running the 6600 on the latest V2 software and the 6500 on V1 for this comparison (I had not upgraded the 6500). 

I'm in a very quiet rural location and my 40m noise level is about -108dBm  with the antenna connected. In other words, I see very nearly the same noise levels with and without the external antennas connected. Nothing wrong with the antenna in TX. 

So I'm concerned that the 6600's internal noise is close to my atmospheric noise levels which would not be very good.

Does my 6600 have a problem or am I being overly critical?

Joe N3HEE

Fred. I agree that would not be good.  I am noticing issues on 160 meters with RXA input being 10 db worse when radio is using bandpass filters.  In WIDE mode the noise floor drops 10 db and siganal levels and signal to noise ratios dramatically improve.  This does not happen with ANT 1 input.  I have not checked noise floor levels with and without antenna connected.  I will check that tonight and report back.  I am about to open a ticket with my findings.  I will also produce a video of the issue.  -Joe N3HEE

Gerald-K5SDR

Fred and Joe,

Thanks for your question, which opens the opportunity for me to talk about noise.  The optimal receiver noise figure relationship to antenna noise is not well understood within the amateur community.  I will confess that I did not really understand it well before I put in the study several years ago.  I have a talk I have done at several events called, "Grokking Receiver Performance" that hits the subject extensively.

I will give you the bottom line and then I will explain some of the technical reasons.  

For optimal weak signal performance near the atmospheric (antenna) noise floor you want your receiver noise floor (sensitivity/MDS) to be 8 to 10 dB below the noise coming from the antenna.  For strong signal reception, less sensitivity is almost always better.

Here is the process to optimize receiver sensitivity for your given operating conditions:

1. Find a clear frequency on your VFO.  Bandwidth doesn't matter.
2. Disconnect the antenna - You can select an input with no connection like RXA or XVTA.
3. Note the dBm reading with no antenna
4. Connect the antenna
5. Note the reading in dBm.  This is where an accurate dBm calibrated S meter really counts.  Ours is truly 6 dB per S unit and measures the actual receiver sensitivity in the selected bandwith.
6. If the noise goes up about 8-10 dB, you have the optimal setting for the noise on your antenna given time of day and propagation.  
7. If the noise goes up significantly more than 10 dB, you have too much preamp gain, which will limit dynamic range for large signals.  Reduce preamp gain.
8. If the noise goes up much lest than 8 dB, increase preamp gain to get it close.

If you are not working weak signals near the noise floor, then less sensitivity is better because you want to use the least amount of preamp gain that allows you to hear the signals of interest.  Even attenuation can be helpful on the lower bands. This maximizes large signal dynamic range.

Now to the difference between the FLEX-6500/FLEX-6700 and the FLEX-6600.  The FLEX-6600 incorporates 7th order band pass filters on all of the HF contest bands.  The 6500/6700 incorporate 3rd order filters on those bands.  The 7th order filters provide greater than 50 dB of band to band isolation between the contest bands at the cost of higher insertion loss. This is excellent for Field Day, SO2R, and other contesting configurations. This additional loss is easily compensated with additional preamp gain if required.  

Let's take a look at the default preamp gain strategy for the 6400 and 6600 radios.  We used the "Rural" noise floor numbers for our calculations to set the default gain for each band.  Note that the 6400 filters have lower loss on the contest bands because it uses 3rd order preselector filters on all bands just as the 6500 and 6700 do.  Below you can see the ITU typical noise power in dBm in a 500 Hz bandwidth with S unit call outs in red.  



Given the Rural noise floor for each band and the receiver noise for each band we calculated a the target sensitivity to be 10 dB below the ITU typical.  We then picked the preamp or attenuation setting.  The table below shows the gain setting and the expected MDS with no antenna attached for that setting on the respective radio.  If you need more or less gain for your location, antenna and operating conditions use the procedure above to adjust accordingly.  



Hope this helps.

73,
Gerald

Joe N3HEE

Gerald.  Thanks for the detailed response and information.  I totally agree with this.  However it did not answer what we are seeing when the receiver is in WIDE with RXA antenna input connected.    Does RXA go through the bandpass filters ?  Why are we seeing better receiver performance in WIDE mode on 160 meters ?  Can you please shed some light on how signals routed from various antenna inputs. Thanks -Joe N3HEE

Gerald-K5SDR

Joe, 

When you go into WIDE mode it bypasses the band pass filter and only has a high pass filter to remove the AM broadcast band.  This is true no matter which antenna connector you use.  The receiver signal chain is the same for ANT1, ANT2, RXA, RXB, XVTA, and XVTB.  There is no harm in going to WIDE mode on 160m or any other band unless there is a strong co-located (e.g Field Day) signal on another ham band, which won't normally be the case.  Lots of strong signals actually add and subtract randomly to look like Gaussian (white) noise in the ADC.

Gerald

Joe N3HEE

Thanks Gerald.  That's good info to know.  So what we are seeing is normal ?  If so this would be good info for the user manual.  I found this by accident.  Going into WIDE with RX antennas on 160 meters makes a big improvement when working weaker DX ! 

It would be nice to be able to switch into WIDE mode without needing to zoom out on the panadapter.  In other words a switch that would turn it on or off and maintain the current resolution of the panadapter. 

Joe
N3HEE

K9SO

Thanks Gerald. I just got back in town and tried your suggestion. I was not taking into account the added loss of the new filters in the 6600 and was not introducing enough preamp gain, especially on 20m.

Surprisingly, I need between 16 and 24 dB of preamp gain to reach the "8-10dB drop" in noise floor with/without the antenna on 20m. Much more gain than what was needed with the 6500. As I said, I am in a quiet rural location. I see 20m daytime noise levels of about -120dBm (antenna) vs. -128dBm (no antenna) with the 16dB preamp setting. 

The old processor between the ears has been reprogrammed to understand this now. I appreciate your explanation. 

Fred
K9SO