[Elecraft] K3 & ADAT ADT-200A by HB9CBU

[Nico Palermo]

Bill W4ZV wrote:

> Correct.  I believe the Perseus only has a 14-bit ADC so it's more limited
> in dynamic range than the 24-bit ADCs commonly used by other SDR rigs.

You probably are unaware that an ADC dynamic range in a given bandwidth
does not depend just on its ENOB (effective number of bits) but also on the
sampling frequency.

> Its dynamic range may be OK for IMD but today's SDRs cannot approach the
BDR of
> rigs like the K3 until even higher resolution ADCs become available.

Sorry to tell you a bad new: you are wrong!
Direct sampling receivers have a TRUE dynamic range, at least at small
spacings,
which is much higher than that of rigs which use IF sampling. This has
nothing to
do with ADC themselves, but with the receiver LOs. Modern SDRs use crystal
grade
LOs and digital tuners. IF sampling receivers don't and they ALWAYS suffer
the phase
noise of their synthesizers.

> I believe the best BDR I've seen for any SDR is the Flex 5000, which has
> just over 120 dB...the K3 is better by about 20 dB.

If we follow the definition of BDR as made by ARRL tests you are right.
Unfortunately this definition does not give an exact idea of the true
dynamic
range of a receiver.
I can demonstrate, and I think that Phil can do it as well with his QS1R,
that if you feed a so called SDR with a + 7 dBm (S 9+80) carrier interferer,

you can comfortably listen to a - 104 dBm (S 4) signal which is 2 kHz apart
the interferer.
Demonstrate that you can do the same with a receiver which uses a 24-bit ADC
at the
IF level and I will publicly apologize with you. In the case you can't I
will be glad to accept
your apologies.

"Conventional" radios begins to behave a little better than direct sampling
platforms
when carrier interferers are at least 100 kHz apart.
But please don't say that the close spacing blocking dynamic range of
"analog" platforms
is 20 dB better than that of good RF sampling radios. This is a really
amazing assertion.

>...which is why the hybrid approach of a narrow roofing filter before the
>ADC works so well in Orion and the K3.  The narrow roofing filters limit
>extremely strong signals outside the passband which would otherwise kick in
>ADC protection AGC.  I don't think there's a way around the fundamental
>limitation of ADC resolution in the direct conversion designs.

This could be true if your receiver LO phase noise were negligible.
Unfortuntately it isn't.

>Phil, on the QS-1R specifications webpage, I see BDR is specified at 125
>dB...very similar to the Flex 5000's 123 dB.  I haven't seen any
independent
>measurements of the QS-1R yet, but hope one will be forthcoming by ARRL.

If they will deserve QS1R the same treatment Perseus had on QST, I wish Phil
a very
good luck.

>Until SDR receivers are in the same >140 dB league of the K3, I'm not sure
>the contest community will bite this apple.  When higher resolution ADCs
>eventually become available at a reasonable price, I'm sure the situation
>will change.

You are still wrong. The situation is already changed but you are not aware
of it
yet. You are confusing a figure which is the Blocking Gain Compression with
the
true dynamic range of a receiver.

> P.S.  How many hits for "1000 " or "5000 " in the contest soapbox so far?
> ZERO.  (Which says something about SDRs in use by real contesters).

This reasoning is a little bit amazing, I think.
Probably in the fourties you had no television set but this did not
prevented a television
set to enter the home of everyone and yours too.
Don't confuse what people knows with what people don't because they haven't
seen yet.
It takes time for news to be widely accepted, but sooner or later they are.
(Doesn't matter if pioneers died meanwhile, life is this way).

73
Nico Palermo, IV3NWV
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qth.net/pipermail/elecraft/attachments/20081202/5b5196d1/attachment.htm