gareth wrote:

What is interesting is the simplicity of the approach (ignoring the hidden
cost of the ubiquitous PC), but it ignores the multiple harmonics that
come out of DDS chips.

Up to about 600MHz I think the DDSen are pure enough, but above that the
VNWA deliberately uses mixer products of some kind. As to how, without
losing all accuracy, I afraid is beyond my understanding. And down to
the genius of the inventor.



So, if you were to working at, say, 3MHz, how would youknow that you were
not actually analysing responses at 6, 9, 12, 15, etc MHz?

The multiple spurious products around are many more than simply
harmonics. But in some way the phase of the desired product is
tracked(?).




http://sdr-kits.net/DG8SAQ/VNWA/Baier_VNWA2_QEX.pdf

But it does raise one question, and that is, who is fooling who about
its purchase price of ooo £450, for there cannot be more than about £40
in its component costs?

That is simply not true. In small quantities they would cost a great
deal more than 450GBP, and the VNWA is not produced in very large
quantities. Consider the costs of quite a complicated PCB in relatively
small quantities, the USB chip is quite expensive to pay for so-called
intellectual property, and requires individual licensing. The DDSs are
not cheap, neither is the hardware.


--

Roger Hayter