AGC, Dynamic Range, 200 Hz Roofing Filters, etc

[Bill W4ZV]

 From my previous post:

"As the resolution and speed of ADC hardware improve, digital radio
receivers become less dependent on analog AGC to meet dynamic-range
requirements. State-of-the-art 24-bit IF-DSP converters, such as
those used in the Ten-Tec Orion, produce about 100 dB of dynamic
range. That means a receiver can handle signals from the noise floor
to almost 40 dB over S9 without analog AGC. Above that level, analog
AGC is still necessary to maintain the linearity of analog circuits
and to prevent overload in the ADC hardware."

                 http://www.doug-smith.net/dspdynamics.htm

         Is 100 dB AGC dynamic range enough?  Generally yes, even
though this range can be extended to 140 dB if very strong signals
activate the Analog AGC protection ahead of the DSP's Digital AGC.
Here are some real-world measurements on 160m by W8JI:

                 http://www.w8ji.com/receiving.htm

The strongest signal measured was -32 dBm above the noise floor
of -127 dBm, which resulted in a needed dynamic range of 95 dB,
well within the capability of the ~100 dB Digital AGC dynamic
range inside the DSP (without any need for Analog AGC protection,
assuming the operator properly positioned the lower limit of the
receiver using attenuation).

         Why might the actual range needed in a contest be less?  Because
the above "quiet band" noise floor of -127 dBm will be significantly
raised by key clicks and transmitted phase noise from many less than
perfect signals.  This is why it becomes somewhat academic to pursue
extremely high IMD performance at very close spacings using roofing
filters much less than about 500 Hz bandwidth.

         At spacings <500 Hz, the noise floor rises dramatically due to
transmitted phase noise and key clicks.  Let's assume a 200 Hz
roofing filter gives you 95 dB IMD performance at 200 Hz signal
spacings.  This would put an interfering signal well outside the
+/- 100 Hz pass band of a 200 Hz roofing filter.  Even for the cleanest
rig available today (Orion) in terms of keyed CW waveform and phase
noise, transmitted noise will be in the ballpark of only 30-40 dB
down at 200 Hz from the carrier (an optimistic assumption).  Thus
the 95 dB you thought you bought with your 200 Hz roofing filter is
totally obliterated by the adjacent transmitted signal, and you can
do *absolutely nothing* in your receiver to eliminate it.  The SDR-X
may hold some future promise of help but Orion or K3 cannot.

         It may make some feel good to buy such filters, but in the real
world of imperfect transmitted signals, it really doesn't do us much
good as far as improving the dynamic range we'll actually experience.
Using steady-stage (unkeyed) crystal oscillators (with exceptional
phase noise) in a laboratory setting to get exceptional IMD results
may make us feel good, but in the real world of transmitter key clicks
and synthesizer phase noise it actually makes little if any difference.

         Of course I realize some will still buy such filters, but
quite frankly it reminds me of audiophile or Citizens Band arguments
for things like gold-plated speaker wire and microphones.  :-)

                                                 73,  Bill  W4ZV

P.S.  My guess for the K3's dynamic range specs are:

BDR at 2 kHz - 143 dB (Wayne already gave us the answer!)
IMD at 2 kHz - 95 dB (equal to Orion - possibly a little better)
Digital AGC - 100 dB (similar to Orion if both use 24-bit ADCs)


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com