From: 		Robert McGwier
Subject: 	[Discuss-gnuradio] Star-10 Transceiver article in QEX
Date: 		Sun, 11 Nov 2007 19:44:26 -0500
User-agent: 	Thunderbird 2.0.0.6 (Windows/20070728)

Dear QEX Editor:

Cornell Drentea, designer of Dentron amplifiers, and one of the many who claim to have invented DDS referenced PLL's (he has as good a case as any) has shown a beautiful example of serious professional engineering in his Star-10 article and he is to be congratulated on a brilliant design and beautiful craftsmanship in building it. It is a very clever and beautifully done traditional design.

This notwithstanding, I have to say that I am particularly disappointed that the editors of QEX allowed the tone of the article to go unchecked. I have comments on the "commentary" and the pseudoscience in the article.

As the ARRL Software Defined Working Group Chairman, I found much of his commentary both insulting and just plain wrong both personally and in my position as a member of this working group. Let me go through my objections in detail. Drentea refers to several projects as "so called software defined radio" projects. The so called software defined radio projects to which he refers are now in use in the Department of Defense, Department of Justice, Department of Transportation, and many radio astronomy sites and many laboratories in many countries. Virginia Tech, the premiere university in the United States for Mobile and Wireless communications engineering education has one of the strongest, if not the strongest, software defined radio program in the country and it is but one. This and many other universities use GnuRadio, DttSP, and more which are the objects of Drentea's scorn. The SOFTWARE defines them as software defined radios and they meet every definition of a software defined radio in the recent FCC rules on SDR and cognitive radio in their application to the SDR-1000, Flex 5000, Softrock, Universal Software Radio Peripheral, HPSDR, uwSDR, and more.

In a clear reference to Flex Radio's SDR-1000, Flex 5000A, Softrock, Norcal 2030 and other radios based upon the Quadrature Sampling Detector or Tayloe detector (balanced or unbalanced), Drentea denies that the measurements made by the ARRL labs in the review of the SDR-1000 and the reviews made of the Flex 5000A by Rob Sherwood (published recently in Passport to HF and soon on his web site) constitute credible authorities on the characterization of radios. Having never seen any of Drentea's measurements of his own radio, I cannot attest to the credibility of his measurements but I do have a comment. As a person who works professionally to do software radio for the U.S. government, I do not know of a lab that can measure 150 dB of IMD dynamic range. The required purity of the oscillators alone involved in both the Star-10 and the test equipment is beyond imagining. The PRODUCT of their noises must be so low as to require something like temperatures that are physically impossible to get and have the oscillators continue to function! The editorial board of QEX should not allow such ridiculous claims to go into print.

Finally, let me state as emphatically as possible. Neither the QSD or the Tayloe mixers are direct conversion receivers of the type Drentea refers to in his article. He simply does not understand what they are. Having analyzed the QSD in the SDR-1000 with the mathematics of Laplace Transforms to find both its transitory and steady state response, I can assure you, he simply does not know what he is talking about and has made a fool of himself.

To say that I am disappointed is to really understate the situation. I am livid beyond almost all repair.

Bob
N4HY

--
AMSAT Director and VP Engineering. Member: ARRL, AMSAT-DL,
TAPR, Packrats, NJQRP, QRP ARCI, QCWA, FRC. ARRL SDR WG Chair


[Bob later posted an apology that read as if it were written by Drentea's and/or the ARRL's lawyer.]

=====================

KR9D
11-14-2007, 09:57 PM
I read with considerable interest Cornell Drentea's article on the Star-10 transceiver in latest issue of QEX.

But I found myself getting more and more confused.

Ostensibly, the point of the article is a demonstration of how "today's technically inclined ham is fully capable of developing full coverage, high performance transceivers that can compete in performance and features with their professional counterparts." This is the conclusion of a brief introductory homily complaining that hams have become mere appliance operators, etc.

I just don't get how the article demonstrates that point. The author is "an accomplished RF technologist, an engineer and a scientist with over 40 years of hands-on experience in the aerospace, telecommunications and electronics industry. He has been involved in the design and development of complex RF, radar, guidance, and communications systems at frequencies of up to 100 GHz." Is that what he means by "technically inclined ham"? That doesn't seem a realistic expectation to me, so I figured that wasn't really the point of the article. After all, those of us not in the biz won't have the benefit of those 80,000 hours of professional development. It's a lot to expect of even a technically inclined ham.

Then, I thought the point might have been to describe a new product that would be available. But no--there is no apparent intention to market the radio. As the article states, "The Star-10 was not intended as a commercial product. Its duplication is not economically feasible."

Perhaps he is bestowing his professional expertise to provide a head start so that "technically inclined" hams can take it the rest of the way. But to achieve the results he claimed, the author boasted of "10,000 lines of perfect code" (written by a professional programmer--not by him--and not available in any way to interested readers), and the computer-aided design of specialty circuit boards (also not done by the author). And the test lab (not the author's) required to integrate it all and make it work is pretty daunting by any standards. And components, such as "samples of the high performance professional grade aerospace PLL chip used in the [frequency reference unit]" are not even available on the market. The samples were donated by a ham who is also a professional at the British company Plessey (RIP), who may not have even been aware that they slipped out the door. Then there's the custom IF filters designed by INRAD and Alpha Components, which you also cannot buy unless you pay to have them custom made all over again.

So, you can't buy it, and you can't build it. Designing something like it is economically infeasible, and requires the cooperation of high-end professionals and companies. Plans, PCB layouts, and proper schematics are not available, and neither are the custom parts and software made especially for it.

The article is to be continued in the next issue. I hope the point of it becomes more apparent then. But it sure seems to me that the point is the author demonstrating that he is smarter than anyone else--just ask him.

I must be wrong, though. Being new, everyone will tell me I don't get it. Fine. Educate me.

Rick "who has never seen 10,000 lines of perfect code" Denney

=====================

KR9D
11-15-2007, 03:40 PM

I don't mind an article about something that isn't reproducible in practical terms by the reader. But what put me off was the homily at the beginning that suggested it should be by average, technically inclined hams. I found that statement utterly unsupported by the article. And I was also confused by the author's writing, which in many places read like sales literature, for something that is not for sale.

Had the author started out saying, "Here's a description of a design process I have undertaken over many years to produce what I think is a true state-of-the-art transceiver. You can't buy it, and I don't pretend that anyone can reasonably make one, but the process itself is instructive," I would have had far less complaint. But even then, I would have expected more elucidation of the process and less bragging about the result.

I review research papers for several academic journals in my line of work, and I would never tolerate a paper such as this. I would think a peer-reviewed journal process would be appropriate, to a point, for something like QEX, but I'm prepared to be corrected.

Every article ought to provide value to the interested reader--that is an implied warranty, if you will. The point of this article, however, seems to have been, "if you are as smart as me, and have as many resources available, you could do this, too, but I'm not going to tell you what 'this' really is."

Rick "not challenging the author's credentials or results" Denney
KR9D

=====================

KR9D
11-15-2007, 03:44 PM

Please don't misunderstand. I read that article prepared to be entertained and instructed, even though I have no desire to duplicate that effort or even buy the product. But for an article that concerns something that can't be produced or purchased reasonably, it ought to at least spend more time on the "how" and less time on the "what".

It's like a review, written by its creator, of an unobtainable product. Most unsatisfying to read.

Rick "who expects some QEX to be esoteric, but still instructive" Denney

=====================