Soft Rock exciter update
tfox at knology.net
Sun Oct 1 22:08:06 CDT 2006
I agree that AMRAD could play an important role here. In fact, I was trying
to figure out how to do this testing myself, but it would be much better if
the club can do it. I am getting spread a little thin.
George and I disagreed about the digitization resolution and sampling
frequency at tacos last week. I don't agree that only 16 or 17 bits are
usable. That may have been true a few years ago, but technology has
improved since then. I think the noise floor is better than that now,
especially with outboard cards. If I remember, 16 bits is only a 96db noise
floor (and 6db per bit after that), something that many sound cards can
exceed today. I have seen some estimates that 20 bits are viable today.
Proposed testing would be one way to resolve the situation. We also
discussed the need for higher sampling rates, which turns out to be good for
us hams, if it also provides a fairly flat response. I think higher samplig
rates are more beneficial and achievable than George did (correct me if I am
mis-stating your positions George).
Another thing to remember is not just check the A/D and input side. As SDRs
are moving into the transmit arena, we need to verify the D/A and output
side as well. I am reminded of this just today, as I get the Soft Rock
exciter prototype on the air.
BTW, here is a web site of someone who has done some testing already. I
think we need to verify the methodology, then take on a bunch of cards.
If we can create a lab that can truly test the various sound card
parameters, and provide that to the amateurs, that would be a great service.
About twice a week there are questions just on the Soft Rock Yahoo site,
asking about good sound cards. we need to look at internal cards, external
cards, and PCMCIA or USB or firewire cards for laptops.
Count me in on helping with this.
----- Original Message -----
From: "Frank Gentges" <fgentges at mindspring.com>
To: "Terry Fox" <tfox at knology.net>
Cc: <tacos at amrad.org>; "Melvin G. Seyle, Jr." <melvin.seyle at comcast.net>;
<dsp at hifidelity.com>
Sent: Sunday, October 01, 2006 9:29 PM
Subject: Re: Soft Rock exciter update
> I had a really great chat with George Le master this morning at the CARA
> Hamfest this morning. We gravitated quickly to the sound card problem.
> I have been looking to buy a better sound card for my SoftRock receiver.
> Sound cards are typically characterized in terms of audible sound and
> not what we need to interface with an SDR. Our conversation concluded
> that we at AMRAD may be in a good position to re characterize sound
> cards for this new and unique application. In addition he brought me up
> to date on the skills of AMRAD member Steve Dove, W3EEE and his
> knowledge and skills in audio engineering. With a little effort, we may
> have really great expertise on the problem within the group and can zero
> in on what sound cards are able to do. They should be:
> 1. cheap
> 2. best choice in performance for SDR duty.
> One key issue is that the sound cards are rated for "A" weighting
> wherein the card is measured from 20 Hz to 20 kHz with a variant of the
> Fletcher Munson hearing curve and with anything above 20 kHz being
> filtered out for measurement. When used for a 96 kHz sampling rate this
> is not a good overall measure for the SDR application. What we need is
> a broad sampling of the many sound cards to see if any do well and if
> not, how to modify one or to design a new one that does.
> While many sound cards now perform a 24 bit digitization, overall
> dynamic range is more like 17 bits with the rest of the bits being
> buried in card noise. We may find a 16 bit card turns out to be the
> best for SDR. Apparently, many sound cards have excess noise above 20
> kHz which will make them less useful in SDR applications. This subject
> is potentially deeper than reading a few reviews from the computer
> Thanks to our esteemed Vice President for his insight into the problem.
> Hopefully, AMRAD members can rise to the challenge.
> I will be announcing the availability of some dumpster grade computers
> in the next message. Combined with a SoftRock receiver and good sound
> card the amateur can have a great SDR on the cheap. I would like to
> call it the "AMRAD Cheap Thrill for All-You-All". As many of you are
> aware, I don't do nested acronyms so we will just forget about using
> Then we need to catch up with you on the exciter side of the
> technology. Similar issues will likely arise on the D/A side of the
> sound card and the software-human interface.
> I am copying this to the Tacos list in hope of stimulating some
> discussion and hopefully work and effort on the problem of sound cards.
> Steve, W3EEE may be able to bring light and knowledge onto the problem.
> Frank K0BRA
> Terry Fox wrote:
> > Hey there all-you-all. The latest here is the soft rock exciter
> > have is working fine on the bench. I am measuring about 1/2W out on
> > 7.060MHz and thereabouts. Putting the signal through a 40db pad and
> > homebrew spectrum analyzer, I am measuring the second harmonic down 46db
> > from the main signal.
> > Now, I am having to learn the PowerSDR software, so I can make my first
> > contact with the SoftRock transceiver.
> > Terry
> > _______________________________________________
> > Tacos mailing list
> > Tacos at amrad.org
> > http://www.amrad.org/mailman/listinfo/tacos
More information about the Tacos