Tayloe Mixer
W3JDR
#6,230,000 (http://www.uspto.gov). This is an intriguing & simple circuit
that offers direct-to-audio conversion, very high dynamic range, inherent
high Q bandpass response, and quadriphase audio outputs...all with only a
few parts! The circuit is just crying out for someone to build it up,
evaluate it, and report on the results. Has anyone taken a look at this yet
and tried it out? If so, I'd like to hear about your experience.
Joe
W3JDR
Roy Lewallen
that the patent has been assigned to Motorola, Inc. Has Motorola given
permission for anyone to make and use the circuit? If so, where has this
permission been published?
Roy Lewallen, W7EL
Ian White, G3SEK
Amateurs have been experimenting with the Tayloe mixer for well over a
year now (for example, there have been many reports in 'Technical
Topics', RadCom).
It gives very good results in a direct-conversion SSB receiver with a
polyphase analog phase-shift network, and I'm sure some people are
already using it with DSP.
Seems like that particular genie is already well out of the bottle...
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.com/g3sek
Phil Rice
I could be accused of blowing my own trumpet on this one (the link below
;-), but here goes:
Have a look at http://ironbark.bendigo.latrobe.edu.au/~rice/ssb/ssb.html
It makes nice SSB at frequencies up to about 5MHz using cheap LS and CMOS parts.
This was originally published in February 1998 in "Amateur Radio", the
Journal of the Wireless Institute of Australia, the worlds oldest amateur
radio organisation (founded about 1910). (So if anyone tries to take me to
court for patent infringment, I can claim "prior art" ;-)
I am amazed that the US patent office would patent this. As far as I know,
commutating mixers were well known in Australia prior to 1979. I first saw
them described in course notes from the Footscray Institute of Technology
in 1979.
A serious literature search in a real library, not the web, will turn up
dozens of reputable papers on commutating mixers/modulators. I did this
years ago, thinking I had discovered something new & wonderful and was
going to collect many academic "brownie points"[1]. It turns out that I
hadn't & I didn't, but then I really enjoyed developing my own SSB exciter.
[1] Brownies - a youth organisation common in England, Australia and New
Zealand. If you have never heard of 'em, then the joke is lost on you. It's
no fun if you have to ask.
Now, I think I will get a US patent on the resistive voltage divider.
Should be easy.
73 & have fun.
de Phil VK3BHR
Remember to keep smiling.
It makes everyone wonder what mischief you have been up to.
Paul Keinanen
wrote:
>Anyone making, selling, or using a patented item is infringing.
I do not know what the exact patent rules are in the US, but in other
parts of the world you can build one for evaluating purposes, but not
sell it. For this reason, patents are often written in a way that it
is hard to duplicate, but the patent should still be able to limit
the use of derived ideas as widely as possible.
Paul OH3LWR
Roy Lewallen
http://www.uspto.gov/web/offices/pac/doc/general/whatis.htm.
Roy Lewallen, W7EL
Roy Lewallen
I'm certain that in the U.S., there is no protection until a patent has
issued. That is, anyone has been free to publish or experiment with the
circuit (unless it qualifies as a trade secret, which seems unlikely
given the publicity it's apparently gotten) up until the time it's
patented. But then it's no longer o.k. to make, use, or sell it in the
U.S. without the patent holder's permission. The fact that people
published, built, or even sold an item before a patent issues doesn't,
as far as I know, cause any dilution of the patent holder's rights. A
patent itself lets an invention "out of the bottle", since it's public
and is required to provide detailed information about the patented item.
Once a patent issues, anyone with the required skill has enough
information to build it. But they're not allowed to. That's essentially
the exchange between the government and the patent holder -- the patent
holder has to provide complete and public information about the device
in exchange for the excluse rights to it for a limited period.
Of course, a U.S. patent has no force that I know of in the U.K., and in
any case I know patent laws are quite different in other parts of the
world.
Roy Lewallen, W7EL
Richard Crowley
> I have to once again state clearly that I'm not a
> patent attorney. But I'm certain that in the U.S.,
> there is no protection until a patent has issued.
Patents are protected from the date of APPLICATION, not from the date of
PUBLICATION. "Patent Pending" is not put on millions of items every year for
nothing!
OTOH, copyright material is covered from the date of PUBLICATION.
> That is, anyone has been free to publish or experiment
> with the circuit
You can experiment all you want, but I wouldn't publish such material
without advice of a real IP lawyer.
> ... up until the time it's patented. But then it's no longer o.k.
> to make, use, or sell it in the U.S. without the patent holder's
> permission.
The prohibition is on SELLING (or actually "exploiting") the IP. The law
says nothing about hobbyists building things for themselves. Why do you
think they are published for everyone to read?
If you look up any pattent (including mine) you will see that dozens to
hundreds of subsequent patents refer to it. Do you think those subsequent
patent holders didn't experiment with concepts in other patents?
> Of course, a U.S. patent has no force that I know
> of in the U.K.
Most IP of any value, particularly if owned by or assigned to multi-national
corporations has protection in place (or pending) in most countries. And
copyrighted IP is covered in most of the world by the Berne Convention.
RC
Grant
>
> Not so in the U.S. See for example
> http://www.uspto.gov/web/offices/pac/doc/general/whatis.htm.
>
If the "you can't even build it" rule actually applies to me building
one (assuming I can come up with a working device from the material in
the patent), and then using it (for no financial gain), then it is
essentially an unenforceable and absurd law in any case.
Jim Weir
shared these priceless pearls of wisdom:
->Anyone making, selling, or using a patented item is infringing.
Roy, I respect your technical expertise and I really appreciate your willingness
to share it with us on a regular basis. However, it is my own personal opinion
that nobody in a "homebrew" newsgroup would give a hairy rat's hiney for your
legal opinions on homebrewing.
Now, as the original poster asked...has anybody played with this thing
(especially at VHF and above) and gotten decent results?
Jim
Jim Weir, VP Eng. RST Eng. WX6RST
A&P, CFI, and other good alphabet soup
AA1LL
curiosity that
a patent disclosure would stop him/her from experimenting. If anything, patents
inspire experimentation. But thet assignee of the patent does have the
right to stop you from building the circuit if he invokes it.
I looked at the patent and noticed 2 things,
1. You still need the 90 degree phase shift at i.f. (audio) which is
always a major pain;
2. The 4 paths of the SP4T switch have to be matched in amplitude
and phase to obtain good carrier and undesired sideband rejection.
To achieve 40dB would require phase accuracy of +/-0.5deg. and
amplitude matching of +/-.08dB, a little tricky in a lot of ham shacks.
The convention these days is to digitize the input by sampling
at 4x the the carrier with a single a/d, do the polarity reversals
and decimate by 2 digitally into 2 paths which are then baseband I and Q.
These I and Q streams represent cosine and sine multiplied versions
of the signal you downconverted.
see http://www.graychip.com for some help if interested.
It would be a good idea to build the circuit but not sell it.
In article <3B282B11...@eznec.com>, Roy Lewallen says...
73, Paul, AA1LL
Mason, NH
http://www.qsl.net/aa1ll
Roy Lewallen
law.
Roy Lewallen, W7EL
Roy Lewallen
haven't made any statements on that topic. I've only commented on what
the law says, for those who care. But I think Jim Weir is right that
most people on this group don't really care to know more about patent
law. So I won't comment further except to say that there are some really
incorrect statements in this posting. Anyone who is interested can get
correct information from http://uspto.gov, or when it really matters,
from a patent attorney.
Have fun!
Roy Lewallen, W7EL
Roy Lewallen
that matter my opinions on whether patent laws are reasonable. They
consisted only of statements of what the law is, in the apparently
mistaken assumption that someone would be interested.
I see that you're not interested in knowing what the law says, and
you're probably right in your implication that nobody else on this
newsgroup particularly cares either -- especially if it might
inconvenience them in doing what they want to do.
So have fun, folks. Ignorance is bliss. I won't bother you further with
the petty details.
Roy Lewallen, W7EL
John Buck
As usual, I find Roy's comments informative and useful. Roy documented
a good source for his opinion. This most certainly is a homebrew
relevant topic. Contrary to the view expressed by another poster, I do
care.
However, on this subject, I believe that Roy's interpretation is too
severe.
From the reference Roy mentioned --
"The right conferred by the patent grant is, in the language of the
statute and of the grant itself, “the right to exclude others from
making, using, offering for sale, or selling” the invention in the
United States or “importing” the invention into the United States. What
is granted is not the right to make, use, offer for sale, sell or
import, but the right to exclude others from making, using, offering for
sale, selling or importing the invention."
My personal interpretation of this is that the patent holder has the
right to exclude me from making, using or selling his patented item.
But I also believe that you or I are not scoff laws in building or
experimenting with one. If he advises me that he does not want me to
use the item in my experiments, then I should stop.
Aloha,
John KH7T
John Miles
>
> I am amazed that the US patent office would patent this.
I'm not. The US patent office, as best as anyone can tell, is manned by
a gang of howling idiots. If there's a limit to what they will accept
as a valid patent claim, nobody I know has been able to find it yet.
Remember, these are the rocket scientists who brought us the Amazon
1-click patent. :( Not to mention the gem at
http://www.delphion.com/details?pn=US05443036__.
It's rapidly becoming impossible to do any software development work in
the US without running afoul of one or more bogus patents that would
never have been granted by an examiner with two brain cells to rub
together. Let's just say that the days when Einstein walked the
hallowed halls of the USPTO are clearly over.
-- jm
------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------
Rick Haub
the phone with my brother who is a corporate lawyer. Although not a patent lawyer he
has worked with companies on patent issues and according to him the part in the patent
law that reads "make or use" only applies if you have commercial "intent". However the
law is fairly (read, how a lawyer or judge wants to interpret it) clear on
"experimental" use. As long as my intent is for experimental purposes I can build it
and use it. It's kind of like having and using a satellite descrambler. I can use one
as long as it is for experimental use only. There is a kind of fuzzy area here though.
If my experimental unit has a commercially available equivalent then I am required to
buy the commercial version. But in this case where the unit in question is a circuit
that is part of larger whole how do you determine the commercial equivalent? The
answer is. . . . . you can't without a lot of lawyers and courts. So unless the
company that has the patent is Disney and as long as your intent is experimental the
question of weather or not the law is being broken is going to remain in that fuzzy
area.
As I said my bother is not a patent lawyer and may not have the right angle on this so
I would like to hear from someone, maybe even the US patent office about this.
Rick KC0GIX
Bozidar Pasaric
> The fact that people published, built, or even
> sold an item before a patent issues doesn't,
> as far as I know, bause any dilution of the
> patent holder's rights.
A question for Roy, W7EL: Does this mean that in the USA
it does not matter who invented something, but who was
the first to claim for the patent, even if he had stolen
the idea? Bozidar, 9A2HL
Roy Lewallen
In the U.S. only the inventor can apply for a patent. But he has to
apply no later than a year after the first public disclosure (e.g.,
publication). If a patent application isn't filed within that year, then
no one can patent the item. But a patent typically takes several years
to be awarded after the application is filed. It was during the period
between the first public disclosure and the time the patent issues that
I was talking about.
In the U.S., patent applications aren't public. So no one except the
applicant even knows the details of what has been applied for, and
there's no protection until the patent actually issues. (And what
actually gets patented is often not very similar to what's in the
original application.) When the patent does issue, the fact that people
discussed or built it during the time it was pending doesn't dilute the
rights of the person to whom it's assigned.
Evidence that the patented item (that is, the thing described in the
patent's claims) was publicly described more than a year before the
patent application was made can be used to have a patent declared
invalid.
This process and these rules are very different in many, maybe even
most, other countries. I believe that in some countries, a patent can be
obtained by the first person to apply, whether or not he's the inventor.
But not in the U.S.
Roy Lewallen, W7EL
DISCLAIMER: I'm not an attorney, and this isn't legal advice. If you
need legal advice regarding patents, see a patent attorney. I'd
appreciate corrections from a knowledgeable person if anything I posted
above isn't correct.
Bozidar Pasaric
> No, not at all.
>
> In the U.S. only the inventor can apply for a patent. But he has to
> apply no later than a year after the first public disclosure (e.g.,
> publication).
Dear Roy,thanks for the info. The American system sounds quite fair. I
understood that Phil Rice, VK3BHR, had something to do with the Tayloe
Mixer. I built his "Fourth Method SSB Exciter" in 1998 after his article
in 'Amateur Radio' of Feb. 1998, and it works fine, exactly as he
described it. The VFO must be four times the transmitting frequency, and
I use a DDS 0-60 MHz with a 100 Hz step, which is 25 Hz in the
transmitter. In his article Phil describes how to use the mixer also for
an RX, but I have not tried it, yet. Instead, there was an article with
the same mixer in the Hungarian ham magazine 'Radiotechnika' 1998/5,
which I tried to make, but was not satisfied because here in Europe we
have very strong broadcasting stations (Free Europe etc.) which
penetrate through simple receivers, so we have a background concert when
listening to the amateur bands. Superheterodynes are still a must here.
Someone told me that electrical diagrams cannot be patented, only their
concrete realization, a product, but I am not sure. Best wishes,
Bozidar, 9A2HL
Jim Pennell
> No, not at all.
>
> In the U.S. only the inventor can apply for a patent. But he has to
> apply no later than a year after the first public disclosure (e.g.,
> publication).
>
Speaking of publishing, I noticed the schematic diagram for the tayloe
mixer and a nice technical discussion about how it works in QEX Mar/Apr 2001
edition Page 61.
This letter did not mention any patent, or any suggestion to not use the
information for non-commercial use. Considering that it was written by
Dan Tayloe, N7VE, I took it to indicate that using it for personal
experimentation by amateurs was allowed.
Jim Pennell
N6BIU
Gerald Youngblood
have been in communication with Dan Tayloe, N7VE, since November of 2000. I
have been developing a Software Defined Radio using Visual Basic and the PC
sound card. I first ran across Dan's design through a Google search. I
dropped him an email and he has been as helpful and open as anyone can be.
He even sent me the schematic last year and I sent him several articles from
QEX that were relevant to commutating mixers and commutating digial filters
but none were identical to his design. An article in QEX even went into
great detail about why 6 dB is the minimum noise figure for mixers that
physics predicts. He subsequently sent a letter to the editor relating to
some of those articles and published his schematic. This you will find on
page 61 of the Mar/April 2001 issue of QEX.
With regard to patent issues, Dan has given me permission to use his design
and to describe his circuit when I present my SDR design paper at the
Central States VHF Conference on June 30. After I recently learned of his
patent I dropped him an email to ask what his intentions were. He said
that there is no problem with experimental use but he said and I quote,
"Just don't try to make a few million dollars off of it. :)" I am sure that
if any of you are serious about experimenting, Dan will be happy to
communicate with you. Some of my friends are interested in using if for
weak signal work so we will be looking at VHF and above. I will drop Dan an
email and tell him to start reading this thread.
I built the circuit earlier this year first on a breadboard and then in
protoboard form. It really does work well. I do not have the test
equipment to verify the noise figure but for 1V in I get roughly 1V out of
the detector. I can also put 4V in without distortion when I use a 5V part.
I plan to layout a PCB soon so that I can get reliable performance. I
expect to be able to go to 30 MHz on receive with 74AC series parts in the
clock circuit. ECL should allow higher frequencies.
Let me clear up another misconception, this is NOT a mixer although it has
some of the characteristics. It is really a quadrature commutating digital
filter. Therefore it does not generate the sum frequency that a mixer does.
This is why it does not have the high loss/noise figure (6 dB+) that a mixer
does. I have done a lot of reasearch and I have not found anything exactly
like it. It is NOT a commutating mixer. Those have been around forever and
they generate both sum and difference frequencies. This is a very high Q
quadrature bandpass filter operating at the carrier frequency, which is in
DSP terms equal to Fs/4 (sampling rate divided by 4). In DSP when you do
downconversion at Fs/4 you do not have any loss in the mixer either because
you simply change the sign every other sample like this 0,1,0,-1,0,1...
In the real design implementation there are NO resistors (ignoring the 2.5
ohms of the paralleled multiplexor switches) until you get past the first
ultra low noise op amp. That means you have no resistor induced noise
before the gain stage. This is an improvement over what you see in the
patent.
Phil Rice, VK3BHR, refers to a design of his in an earlier post. He states
on his web page at:
http://ironbark.bendigo.latrobe.edu.au/~rice/ssb/ssb.html...
"This switching modulator should be capable of acting as a demodulator. This
would require reversing of the direction of signal flow through the circuit.
The problems expected in doing this are firstly that attenuation through the
sequence network would prevent the reception of microvolt signals. Secondly,
to obtain 40 dB of opposite sideband suppression, the signal level through
the analogue switch would have to be held below 0.05 volt peak to peak. This
would result in a poor dynamic range."
The problems Phil mentions are not an issue in Dan's invention. Also with
DSP you can easily correct for phase and amplitude imbalance in software.
If you are interested in experimenting with the Tayloe Detector (NOT mixer)
I will be happy to put you in touch with Dan and to share my expriences with
you. I would also love to communicate with anyone interested in PC based
Software Defined Radios.
I hope that this helps to clear the air on some of the questions in this
thread. If you have questions please feel free to email me at
gyoun...@seedstage.com. I don't normally read news groups but a fellow
ham referred me to this discussion because of the Tayloe discussion.
Happy experimenting and 73,
Gerald Youngblood
AC5OG
"Jim Pennell" <me-n...@killspam.com> wrote in message
news:9gd5ce$o62$1...@slb0.atl.mindspring.net...
Clifton T. Sharp Jr.
> Dear Roy,thanks for the info. The American system sounds quite fair.
It would be if it was implemented and run correctly.
Now please excuse me. I have to finish my application. I'm patenting pants.
--
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Cliff Sharp | "Speech isn't free when it comes postage-due." |
| WA9PDM | -- Jim Nitchals, founder, FREE |
+-+-+-+-+-+-+-+-+-+- http://www.spamfree.org/ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Dan Tayloe
Good defense in this note, Gerald! I don't think enough people
understand the detector well enough to understand why it is not an
ordinary commutating mixer, and how it is different from all the other
commutating mixers that have ever been done.
It sounds a lot like a regular mixer. I simply gave up debating the
point with one fellow who just didn't get it (and likewise thought **I**
just didn't get it).
I saw Phil Rice's article for the first time about a year ago. It is
not the same, but it is similar. However, as you have pointed out, his
circuit was not useful for either weak signal nor high dynamic range
receiver work, where as mine has been optimized for both.
Now on the subject of permission:
I do not own this patent. I surrendered my rights to Motorola, and I am
not empowered to speak for the company. As such what I say carries no
more weight than what Roy says.
However, I will offer my humble opinion. It seems like it all comes
down to "What is in it for me?". Motorola is in the business to make
money. Motorola will lose no money by anyone experimenting with this
detector.
On the contrary, if experimenters use this circuit, find it useful, and
publish positive results, that will greatly increase the probability
that some commercial interest will want to use this detector in a high
volume consumer application. Such an application could in turn generate
Motorola royalty revenues, thereby profiting Motorola.
People experiment, Motorola wins.
I therefore see no business reason why Motorola would discourage the
experimentation and promotion of this detector, since it seems to be in
the company's best interest, and there is no profit to be gained from
discouraging such experimentation.
I have no idea what Motorola's interest threshold is here, but selling a
few hobby kits is likely way below that threshold. However, if serious
commercial revenues were to be made on the application of this
invention, it would eventually get their attention.
- Dan Tayloe, N7VE
Jim Weir
shared these priceless pearls of wisdom:
->I see that you're not interested in knowing what the law says, and
->you're probably right in your implication that nobody else on this
->newsgroup particularly cares either
No, Roy, that's not what I said. RTFM (and variants thereof). I said I
respected your opinion on things electronic. Until you can put JD after your
name and have a few fortnights of experience in patent law, I said I didn't care
for your legal opinion. I said nothing about not caring about the law, just
didn't care for an opinion from an unqualified source.
Clifton T. Sharp Jr.
> I saw Phil Rice's article for the first time about a year ago. It is
> not the same, but it is similar. However, as you have pointed out, his
> circuit was not useful for either weak signal nor high dynamic range
> receiver work, where as mine has been optimized for both.
The question I've most wanted to ask you since I first saw this nifty
circuit: what got you started in this direction? What were you looking
for or experimenting with that eventually led you to do this design?
Dan Tayloe
"Clifton T. Sharp Jr." wrote:
>
> Dan Tayloe wrote:
> > I saw Phil Rice's article for the first time about a year ago. It is
> > not the same, but it is similar. However, as you have pointed out, his
> > circuit was not useful for either weak signal nor high dynamic range
> > receiver work, where as mine has been optimized for both.
>
> The question I've most wanted to ask you since I first saw this nifty
> circuit: what got you started in this direction? What were you looking
> for or experimenting with that eventually led you to do this design?
Actually what got me started in this direction was the availability of
the TI 74CBT3253 high speed analog mux and a 3000 mile (Phoenix to New
York) cross country drive. I had lots of time to do nothing but think
and ponder. The 74CBT3253 is a zero delay analog mux (as opposed to a
high delay digital gate) used to connect microprocessor address and data
lines to different banks of memory. It was never intended for analog
use, but it works quite well for that task.
I knew about audio switched capacitor filters using an input resistor
and four capacitors which were switched at a 4x rate. The input side to
the mux ends up exhibiting sharp bandpass behavior, based upon the R/C
time constant of the single R and multiple Cs used.
The idea of an active RF bandpass filter on a receiver front end is what
got me started. Since we now had a *really* fast dual 4:1 mux part,
perhaps I could do some active RF bandpass filtering using the switched
capacitor filter in front of a receiver mixer rather than a simple fixed
tune bandpass L/C.
The really neat thing about this is that if I combined the VFO and IF
BFO signals together at a 4x rate, this narrow active RF bandpass filter
would track the receiver tuning. This should enhance receiver
performance on congested bands such as 20m (contests) or on 40m with big
adjacent SWBC signals.
I pondered this and the fact that if I was using a really narrow rf
active filter, the signal I really wanted was at an offset of 600 to 800
Hz, and if I tried to do 4x at that offset frequency, I would likely be
injecting a really strong signal on top of what I wanted to listen to.
I pondered this aspect quite a bit.
In order to produce the necessary 4x clocking for the rf switched
capacitor filter, it would be necessary to combine the BFO and VFO
together and then produce a composite 4x signal (maybe using a PLL or a
bunch of exactly synchronized DDSs... a bit complex but manageable) to
drive the filter.
In order to simply test the idea, I decided upon a scheme to use a
direct conversion receiver as the test bed. Then it would be simple
since the 4x signal used for the RF switched capacitor filter could be
divided by four and used for the direct conversion mixer VFO.
Much simpler! A direct conversion receiver would not be optimum, but it
would allow me to test the idea before trying the more complex 4x
clocking trick on a superhet receiver.
It was then that I started thinking (I was still driving) about what the
signals looked like on the switched capacitor filter at zero beat to an
incoming RF signal, which would be four equidistant points on a
sinewave, stationary. Then I thought about what they looked like when
not exactly on frequency: each of the four outputs would drift slowly
tracing a sinewave at the difference frequency, each formed of RF
samples taken 1/4 cycle apart, each output then being 90 degrees apart
from each other.
I suddenly realized that the switched capacitor topology could be used
as a detector (no separate direct conversion mixer required), that it
naturally produced all four phases of output, had bandpass
characteristics, and that the input RF signal was being "integrated" on
to each of the four capacitors through the input filter R, so that only
difference frequency appeared and not the sum. Lastly, I realized that
the new detector seemed to capture the peak RF voltage, and thus have a
low detection loss (under 0.9 db conversion loss).
Since I have read about the R2 phasing receiver, I began to work on
receiver designs based upon that approach (two branch I and Q audio
combined using 90 degree phasing shifting to given single sided
reception). I have been working on incrementally making improvements to
the basic design for several years now. I am on my 4th generation
design, and there are still ideas to try out and improvements to made.
I am trying out new modifications this weekend.
I never returned to pursue the RF active front end receiver filtering
aspect. There is only so much time in the world to experiment! It might
not work the first time, but problems are simply engineering
opportunities.
- Dan Tayloe, N7VE; Phoenix, Az; Az ScQRPions
Ralph Lane
details of your improvements to the Tayloe mixer. The more I read about this
mixer, the I am convinced that it is the way to go for future receivers.
Ralph Lane
Gerald Youngblood
I agree with you that this is the way to vastly improve the performance of
receivers of the future. The improvements over the patent are really Dan's.
The main thing that is different from the patent is that no resistors are
required on the op amp inputs and the op amps are used in true differential
mode. The RC integration is provided by the 50 ohms of the antenna and the
sampling capacitor which is selected for the bandwidth of the filter
(usually .33 to .47 uF). Since the R is duty cycled at 25% the impedance
seen by the input of the op amp is actually 200 ohms. So the first stage
audio gain is Rf/200. The bandpass filter bandwidth is BW=1/Pi*200*C. I am
using a Linear Technology LT1115 Ultra Low Noise Op Amp in my circuit that
at 200 ohms input yields only a couple of dB noise factor.
You should feel free to contact Dan directly. I am sure he will be glad to
tell you more about the design.
"Ralph Lane" <crl...@erols.com> wrote in message
news:3B2D8A23...@erols.com...
Duncan Clark
writes
>My personal interpretation of this is that the patent holder has the
>right to exclude me from making, using or selling his patented item.
Not quite.
You are allowed to use whatever is in patent to carry out research on
the area that is covered. You cannot sell what you do/manufacture but
you could apply for a patent on any improvements you make. Anyone
wanting to use your patent to produce the item would invariably have to
also take out a license from the original patent. If they do not grant
one then no product can be sold.
You are not barred from manufacturing anything in a patent in order to
test that the item patented actually does behave as claimed in the
patent. This is very important in patent law! A patent has to contain
all the 'instructions' for one to duplicate the item(s) being patented.
If it is not possible to do this from the instructions or said patented
product does not work as claimed then it usually leads to a patent
challenge and has been known to lead to the throwing out of the patent
completely - via the usual court systems.
My knowledge is primarily biotechnology but the same general
observations apply to all patent law. I will be seeing a patent attorney
in the next week or two and will conform this with them.
Duncan
G4ELJ
--
Support bacteria. They are the only culture some people have.
Duncan Clark
Duncan Clark
writes
>In the U.S., patent applications aren't public.
Now changed. Full patent app. is now published 12 months after date of
filing, I think from last October (?) bringing the US law into line with
EU.
Duncan
Duncan Clark
writes
>On the contrary, if experimenters use this circuit, find it useful, and
>publish positive results, that will greatly increase the probability
>that some commercial interest will want to use this detector in a high
>volume consumer application.
There has been an excellent series of articles published in the UK in
the RSGB's Radcom magazine in it's Technical Topics section, March,
April, May and June 2001. G3OGW has built a 4 path polyphase D-C
receiver using your mixer, based around an FST3253 fast bus switch. The
results look very impressive and so simple! Even before I saw this
thread (rather late I admit) I was considering trying G3OGW's
experimental design just to see how it sounded in practice.
Duncan
G4ELJ
Ps I'm not sure if this info. is available on the RSGB web site or not.
Bozidar Pasaric
For the whole of my life (I am 69 now) I had to read and understand
analogue electric diagrams which I can do very quickly - almost at a
glance. Then, digital technology has come and is pushing out the
analogous one. Now, I cannot read those new diagrams as easily as
before. My brain is helplessly analogous. When reading those digital
boxes I usually begin thinking of a lamp switch, which is either OFF
or ON, but when the things become more complicated I lose the thread.
Now I am trying to understand how the Tayloe mixer works. Dan's
description of his stream of thoughts when driving over the continent
has helped me a lot, and Gerald Youngblood's additional explanation
was also useful. Because I am a visual type, I have made the following
picture in my mind:
We all know those situations when we watch a film or TV and when a
carriage passes by, but the wheels turn very slowly forwards or even
backwards. Obviously, what we see is the difference frequency between
the shooting speed of the camera (25 pictures per second) and the
frequency of the spokes in the wheel. First, the process is digital,
because the picture is either on, or off; and so are the spokes. Second,
there is no sum of frequencies, only the difference, so it is not a
picture of a mixer. Third, we can see this difference frequency
directly, without any additional device, like in the Tayloe Mixer. If
the spoke frequency is higher than the camera frequency, the wheel turns
slowly forwards (USB); and if the spoke frequency is lower than the
camera frequency it turns slowly backwards (LSB). Anyhow, we see the
result directly, pretty in the same way as in Tayloe Mixer when we
get the SSB directly from his detector (or digital filter).
Now, will someone with a deeper insight into Tayloe Mixer tell me if
my comparison and explanation are correct? Thanks.
Bozidar, 9A2HL
WB8RCR
about the thought process that brought him to the mixer. That might help.
Also, there were several threads on QRP-L in the fall of '98 around the
mixer between Dan, Paul Harden and Dave Fifield. Great stuff reading the
discussion between these folks.
The QRP-L archives are at http://qrp.lehigh.edu/lists/qrp-l/
didleydadidah
"Bozidar Pasaric" <bpas...@mac.com> wrote in message
news:1evch2r.1fxab551fc06aiN%bpas...@mac.com...
Clifton T. Sharp Jr.
> How does the Tayloe Mixer work?
Read <http://www.amrad.org/pipermail/tacos/1998/000464.html>, which is an
explanation by Dan himself, then re-read it, then ask questions. :)
Terry Ritter
On Thu, 21 Jun 2001 09:10:19 +0100, in
<1evch2r.1fxab551fc06aiN%bpas...@mac.com>, in
rec.radio.amateur.homebrew bpas...@mac.com (Bozidar Pasaric) wrote:
>[...]
>We all know those situations when we watch a film or TV and when a
>carriage passes by, but the wheels turn very slowly forwards or even
>backwards. Obviously, what we see is the difference frequency between
>the shooting speed of the camera (25 pictures per second) and the
>frequency of the spokes in the wheel. First, the process is digital,
>because the picture is either on, or off; and so are the spokes. Second,
>there is no sum of frequencies, only the difference, so it is not a
>picture of a mixer.
Well, that is an interesting analogy. It basically seems to be a
Nyquist sampling problem, with the sampling rate being too slow to
resolve the motion of the spokes. But if we can't see the spokes at
their full rotational rate, we surely can't expect to see the sum of
sampling plus that rate. So, in my view, the experiment does not
testify that no sum frequency occurs.
Switching mixers are old (tube) technology, and bus-switch mixers work
pretty much the same way: They switch a linear signal on and off.
Balanced mixers hide the mixer frequency by switching two linear
signals which have opposite phase (or by switching a transformer
output back and forth in phase). This is much easier now that we have
small, low-cost, high-speed and matched components in an IC.
Of course, in reality, most digital logic is composed of "linear"
(analog) devices used over restricted voltage regions. As one might
expect, a large signal into the FET which is a bus switch does affect
the conduction of that switch to some extent, which is distortion.
In my view, any switching mixer *must* create both sum and difference
frequencies. (Indeed, a switching mixer *also* will create sum and
difference frequencies, with reducing amplitudes, at square wave
mixing multiples of 3x, 5x, 7x and so on.) But removing the major sum
frequency when converting to baseband has never been a problem: We
just use a simple R-C filter, or even amplify just audio. We can and
sometimes do listen to AM signals with a zero-beat BFO. Nobody is
concerned about the double IF frequency produced by the product
detector.
The real problem has always been that this is not a "single signal"
system, in that signals slightly above the mixing frequency convert to
audio (baseband) just like signals slightly below the mixing
frequency. So it works for AM or DSB, but not so well for SSB or CW.
Now, if we have a pair of mixers in quadrature, with appropriate
fixed-phase quadrature signal delays, it is possible to reject one
mixing sideband, to some extent (up to 30dB or, with another stage,
perhaps 50dB). This is also well-known old technology (and we really
want 100+dB). Unfortunately, the necessary fixed-phase delay at audio
has always been both a problem and a limiting factor in performance.
But I don't see that as having been fixed, or, indeed, affected at
all.
>Third, we can see this difference frequency
>directly, without any additional device, like in the Tayloe Mixer. If
>the spoke frequency is higher than the camera frequency, the wheel turns
>slowly forwards (USB); and if the spoke frequency is lower than the
>camera frequency it turns slowly backwards (LSB). Anyhow, we see the
>result directly,
> pretty in the same way as in Tayloe Mixer when we
>get the SSB directly from his detector (or digital filter).
That sounds like the real problem, displayed: We can "see" spokes
equally well both at slightly above, and also slightly below, the
mixing frequency. In fact, if the shot lasts long enough, we can go
through a couple of these, as the wheels speed up. I think I have
actually seen cowboy movies where the stagecoach uses larger and
smaller wheels, where one can see the spokes of different wheels
rotate in opposite directions simultaneously.
>Now, will someone with a deeper insight into Tayloe Mixer tell me if
>my comparison and explanation are correct? Thanks.
>Bozidar, 9A2HL
Perhaps we can encourage a deeper discussion.
Terry KF5MH
Gerald Youngblood
Your analogy sounds like a very good description to me but Dan would be
better at confirming it. Another way of thinking of it is that when the
carrier is at zero beat the in-phase side of the detector will integrate 0
deg and 180 degree segments of a cosine wave which adds to zero through the
differential inputs of the op amp. When you offset the carrier either way
it will no longer add to zero but will be a wave which corresponds to the
difference frequency. If you think about the sum frequency which is the
same as 2X the carrier frequency you will have two complete cycles of which
also add to zero. In other words with perfect components even harmonics add
to zero in the output of the detector.
For those who do think digital, this looks a lot like a comb filter where it
responds to odd harmonics of the carrier frequency. By the way you only
need a roofing filter with this detector to get rid of the harmonics. My
measured output from the detector looks like this with no input filter:
7.1591 MHz 0 dB
7.16385 MHz -3 dB
7.16743 MHz -6 dB
7.21065 MHz -20 dB
7.61212 MHz -50 dB
If I did my math correctly that is a 9.5 KHz bandpass and a Q of 753. I can
reduce the bandpass simply by changing the sampling capacitor value.
73 to all,
Gerald, AC5OG
"Bozidar Pasaric" <bpas...@mac.com> wrote in message
news:1evch2r.1fxab551fc06aiN%bpas...@mac.com...
Dan Tayloe
tell from the apparent direction of the wheel whether the it was faster
(moving forward) or slower (moving backwards) than the motionless state,
which is exactly in sync.
The same effect is used to determine if a signal is LSB (slower than
zero beat) or USB (faster than zero beat). Looking at just the 0 degree
("I" for "In Phase") and 90 degree ("Q" for "Quadrature") components,
"I" leads "Q" by 90 degrees on one side of the carrier, and "Q" leads
"I" on the other side.
Thinking about the wagon wheel analogy to USB/LSB, I think I leading Q
is USB (wheel moving forward, frequency higher than zero beat), and Q
leading I is LSB (wheel moving backward, frequency lower than zero
beat).
What the phase shifting network does is delay one side by an extra 90
degrees more than the other side. Suppose we shift "Q" by an extra 90
degrees compared to I.
On USB, I leads Q by 90 degrees, now adding phase shift, I leads Q by
180 degrees.
On LSB, where Q leads I by 90 degrees, add the phase shift and now I and
Q are both in phase.
If you sum the two sides together after phase shifting, LSB signals will
add from I and Q, but USB signals will cancel (180 degree signals are
negatives of each other).
Alternatively, if you use an op-amp to differentially sum the two
together, the 180 degree apart USB signals will now add together, and
the same phase LSB signals will now cancel.
Take your pick. USB or LSB depending on how they are put together.
On the other hand, if you used a Tayloe detector, just running the
detector clocking backwards will also switch the sidebands, which may be
simpler.
- Dan Tayloe, N7VE; Phoenix, Az; Az ScQRPions
Clifton T. Sharp Jr.
> On the other hand, if you used a Tayloe detector, just running the
> detector clocking backwards will also switch the sidebands, which may be
> simpler.
It just has to feel really great to say "Tayloe detector", knowing the
name is really deserved. I know it's way different from how it feels when
I say "Sharp television". :)
Gerald Youngblood
To get good second harmonic suppression you have to have good clock
symmetry. I did not in the test that I did earlier. I plan to correct that
in my future layout.
With poor symmetry I got:
14.318 MHz -26 dB (This should be down >40 dB with a good clock)
21.4782 MHz -13 dB
28.6369 MHz -31 dB
50.111 MHz -25 dB
What you need with the Tayloe Detector is a simple low pass filter rather
than a band pass filter. The detector itself is a narrow band pass filter
you do have to get rid of the harmonics.
73,
Gerald, AC5OG
-----Original Message-----
From: Crabt...@aol.com [mailto:Crabt...@aol.com]
Sent: Friday, June 22, 2001 10:33 AM
To: gyoun...@austin.rr.com
Subject: Re. Tayloe Mixer
Hello Gerald
Thanks for the excellent post. I like the description of it being a comb
filter - slowly it is starting to make sense to me. The implications of it
being sensitive to odd harmonics are that front-end filters will be
necessary, possibly a little wider than octave filters if they have steep
roll-offs. I was begining to think that the Tayloe Mixer with no front end
filtering might be a little too good to be true.
Would it be possible for you to do some measurements around the second and
third harmonics of the fundamental frequency, ie around 14.3 and 21.45 MHz,
and then post them? This would be helpful to me and no doubt others to
understand this mixer.
Thanks and 73
John KC0GGH
On 2001/06/21 you posted:
Ralph Lane
output was dc to audio output with the bandwidth set by the .47 microfarad caps.
The harmonics of the local osc. should be of little concern. Am I missing
something?
Ralph
Dan Tayloe
yourself simply by sketching a few pictures.
At the fundamental, draw one full cycle of a sinewave and break it up
into four pieces.
You will be able to see the third and four areas (180 degrees and 270
degrees) are simply the inverse of the first two (0 degrees and 90
degrees).
Remember, 0 will be differentially summed with 180 to get a single 0
degree ("I") signal, whereas 90 will be differentially summed with 270
degrees to get a single 90 degree ("Q") signal.
Remember also that these RF quarter section pulses are being
**averaged** onto a detector cap via the system impedance, each of those
four areas ends up with the average (integral) of the RF pulse over that
quarter section area.
Now draw two cycles of sinewaves and break them equally into four
pieces. You will notice that 0 and 180 now look the same and 90 and 270
degrees also look the same. When the averages of these four areas are
differentially summed together (0&180, 90&270), they will cancel. This
will happen will all even harmonics. The degree of cancellation depends
on the balance of the circuit. At one time I measured this at 40 db
down. I do not remember what circuit configuration that was.
Now draw three cycles of sinewaves and break them into four pieces.
Each of the four pieces will have 3/4 of a sinewave in it. If the 0
degree piece averages to be a bit on the positive side, the 180 piece
will average a bit on the negative side and thus there will be some net
detection. The same happens on the 90 and 270 degree sections. There
will be some net detection. I think I measured something like 12 db
down at the third harmonic.
On higher order odd harmonics, each 1/4 section will see a longer and
longer sinewave pattern, most of which averages to zero, but there will
always be a leftover portion that will be non zero, and will
differentially add with its companion detection output. You could
actually do the math (take the integral of the sinewave sequence and
calculate the average over that 1/4 section) and determine how far down
each harmonic should be from the peak.
If you do the integration for the straight fundamental case, you should
be able to derive the 0.9x detection, which is also 0.9 db of loss
(integrate over +/- 1/8 cycle either side of the sinewave peak).
- Dan Tayloe, N7VE; Phoenix, Az; Az ScQRPions
W3JDR
I think the reference to 3rd harmonic responses means that the circuit will
theoretically detect signals at odd harmonics of the LO sampling frequency.
In practice, there are also responses to signals at all harmonics of the
sampling frequency. The magnitude of these responses is a function of the
order of the harmonic, and heavily favors the odd harmonics. This just
points out the need for an RF input filter, but this filter can be somewhat
'sloppy' compared to image-reject filters required for heterodyne receivers.
What's really cool about Dan's circuit is that everything for a phasing type
DC receiver, up to the audio phase shift network, can be integrated on one
chip. Historically, the audio phase shift circuit has been the most
troublesome part of a phasing receiver, but modern DSP technology makes it
possible to implement that part with extremely high precision and very low
component complexity. With DSP, software can be implemented to automatically
compensate for phase and amplitude imbalances in the hardware.
I think Dan's technique, combined with DSP I/Q audio processing, will make
it possible to build very inexpensive & high-performance multi-band radios.
I'd like to hear about some experimentation along these lines.
Joe
W3JDR
"Ralph Lane" <crl...@erols.com> wrote in message
news:3B356163...@erols.com...
Ralph Lane
harmonic radiation from the mixer as well. I would think that the I/Q and DSP
portions of the new cellphone chips could be used to complete the design of the
receiver you suggested.
Ralph
Gerald Youngblood
The design you describe below with the Tayloe Detector directly driving a PC
sound card for DSP is exactly what I am building. I am using Visual Basic,
DirectX8 and the Intel Signal Processing Library. I have it working now
using FFT fast convolution filtering. I don't have the I & Q balancing yet
but I know generally how to do it. You can't beat this design for
simplicity, low cost, and performance from what I have seen so far.
73,
Gerald
"W3JDR" <j...@usa.com> wrote in message
news:LflZ6.12725$h27.5...@typhoon1.gnilink.net...
W3JDR
Sounds like you're doing exactly what I've been hoping someone would do. I'm
surprised that no one has used a PC sound card before to do the I-Q
processing.
I travel a lot, and I've often thought it would be nice to have an extremely
small, low power, multi-band radio I can just stuff into a corner of my bag.
Since my laptop goes wherever I go, it would be a natural to use it as the
DSP processor, audio amp, and front panel for a full-featured rig.
Please keep me and the group informed....this is ground-breaking stuff.
Joe
W3JDR
"Gerald Youngblood" <gyoun...@austin.rr.com> wrote in message
news:3DxZ6.85507$lq1.21...@typhoon.austin.rr.com...
Gerald Youngblood
I will be happy to keep the group informed. I will be presenting a paper at
the Central States VHF Conference on July 28 in Dallas. I also have gotten
the go ahead to write a four part series for QEX from Doug Smith. I hope to
use the material I am presenting for Central States and expand on it for the
series.
My current design uses the Tayloe detector on a small Radio Shack
protoboard. It has a 74CBT3253, 74AC74, 74AC04, and two LT1115s. I drive
the clock with a HP8640B signal generator. The outputs of the two LT1115
op-amps directly drive the sound card as I and Q. The PC does everything
else. You won't believe the performance you get. And its all in Visual
Basic which most mere mortals should be able to program in.
By the way right now I don't even have a front end low pass filter. On 40 M
it works pretty good without it even though I don't recommend that. I have
a lot more work to do but it is already working extremely well.
73,
Gerald
"W3JDR" <j...@usa.com> wrote in message
news:4_EZ6.25056$1T5.1...@typhoon2.gnilink.net...
Jerko Golubovic
> On Thu, 21 Jun 2001 09:10:19 +0100, in
> <1evch2r.1fxab551fc06aiN%bpas...@mac.com>, in
> rec.radio.amateur.homebrew bpas...@mac.com (Bozidar Pasaric) wrote:
>
> >[...]
> >We all know those situations when we watch a film or TV and when a
> >carriage passes by, but the wheels turn very slowly forwards or even
> >backwards. Obviously, what we see is the difference frequency between
> >the shooting speed of the camera (25 pictures per second) and the
> >frequency of the spokes in the wheel. First, the process is digital,
> >because the picture is either on, or off; and so are the spokes. Second,
> >there is no sum of frequencies, only the difference, so it is not a
> >picture of a mixer.
>
> Well, that is an interesting analogy. It basically seems to be a
> Nyquist sampling problem, with the sampling rate being too slow to
> resolve the motion of the spokes. But if we can't see the spokes at
> their full rotational rate, we surely can't expect to see the sum of
> sampling plus that rate. So, in my view, the experiment does not
> testify that no sum frequency occurs.
Same with this - you cannot hear that.
> In my view, any switching mixer *must* create both sum and difference
> frequencies. (Indeed, a switching mixer *also* will create sum and
It removes one of them with phase network after.
Speaking about phase networks - I need to shift Q by 90 degrees in range
300 Hz to 3000 Hz and get orthogonal axes on XY plot. All particular
frequencies are sine waves. When you do shift sine wave by 90 degrees
you get cosine wave. cos(x) is first derivation of sin(x). Classic
derivator won't work here - that is why we use networks.
Somebody mentioned that he used PC to do that. OK, sample I on left and
Q on right and then you derivate right and sum that two ones properly.
Is it so simple or I am wrong in something?
If it is, I can do so with PIC 16F874.
> Perhaps we can encourage a deeper discussion.
I will be glad to participate.
73 de 9A6JGJ!
Jerko Golubovic
> Roy Lewallen wrote:
> >
> > Not so in the U.S. See for example
> > http://www.uspto.gov/web/offices/pac/doc/general/whatis.htm.
> >
>
> If the "you can't even build it" rule actually applies to me building
> one (assuming I can come up with a working device from the material in
> the patent), and then using it (for no financial gain), then it is
> essentially an unenforceable and absurd law in any case.
Patents are public documents. Purpose of patent is that author or owner
gains money out of possible commercial production.
If you can not see it and try it how you can eventually produce it.
Selling kits would be illegal but just making I think not...
Regards,
Jerko Golubovic
9A6JGJ
Jerko Golubovic
> I think Dan's technique, combined with DSP I/Q audio processing, will make
> it possible to build very inexpensive & high-performance multi-band radios.
> I'd like to hear about some experimentation along these lines.
DSP-10 radio uses DSP and undersampling IF by using EZ-KIT DSP.
73 de 9A6JGJ!
Jerko Golubovic
> > Let me clear up another misconception, this is NOT a mixer although it has
> > some of the characteristics. It is really a quadrature commutating digital
> > filter. Therefore it does not generate the sum frequency that a mixer does.
> > This is why it does not have the high loss/noise figure (6 dB+) that a mixer
> > does. I have done a lot of reasearch and I have not found anything exactly
> > like it. It is NOT a commutating mixer. Those have been around forever and
> > they generate both sum and difference frequencies. This is a very high Q
> > quadrature bandpass filter operating at the carrier frequency, which is in
> > DSP terms equal to Fs/4 (sampling rate divided by 4). In DSP when you do
> > downconversion at Fs/4 you do not have any loss in the mixer either because
> > you simply change the sign every other sample like this 0,1,0,-1,0,1...
I would rather call it translator than mixer. It translates baseband
from HF to 0 Hz.
From what I can see this is just fine example of raping Shannon-Nyquist
theorem. This is what happens whan you get aliasing situation (bad
aliasing filter for example). Result sampled signal is garbage.
Same thing happeed to me in mentioned software when I put insufficient
number of points (iterations per second) in parameters - result was not
good at all and envelope of signal was floating in harmonic way with
frequency being equal to difference. But even then I did not realized
what Dan realized although from today point of wiew it is so obvious to
me.
There are number of ways to do that. Up to now doing so was considered
bad. Now it turned to be useful :) .
Regards,
Jerko Golubovic
9A6JGJ
Gerald Youngblood
First of all make sure that you understand how a phasing receiver works in
analog. Next you should read The Scientist and Engineer's Guide to Digital
Signal Processing which can be found for free at www.DSPguide.com. Next you
should read an article by Richard Lyons called Quadrature Signals: Complex,
But Not Complicated, which can be found at
http://www.dspguru.com/info/tutor/quadsig.htm on the www.dspguru.com site.
This should get you started. If you want to go further you should read
Understanding Digital Signal Processing also by Richard Lyons, which you can
buy from Amazon. All three of these references are excellent but I would
recommend reading in the order that I have given them. Don't let the math
scare you because using DSP in practice is actually easier than the math
implies. I don't really understand it all but I can use it in practice.
73,
Gerald
-----Original Message-----
From: Jerko Golubovic [mailto:385912...@vip.hr]
Sent: Wednesday, June 27, 2001 4:19 PM
To: Gerald Youngblood
Subject: DSP od I and Q
Hello!
I am reading homebrew newsgroup regarding Tayloe mixer and I am
interested in doing I and Q with DSP. Could you please point me on
appropriate resources regarding this topic.
For first it would help me to know how exactly you need to deal with I
and Q using DSP.
I never used any DSP techniques, but I am willing to learn them.
73 de 9A6JGJ!
Gerald Youngblood
Gerald Youngblood
See comments below:
"Jerko Golubovic" <385912...@vip.hr> wrote in message
news:1evomfr.ksm3mx148akt6N%385912...@vip.hr...
> Terry Ritter <rit...@io.com> wrote:
>
> > On Thu, 21 Jun 2001 09:10:19 +0100, in
> > <1evch2r.1fxab551fc06aiN%bpas...@mac.com>, in
> > rec.radio.amateur.homebrew bpas...@mac.com (Bozidar Pasaric) wrote:
> >
> > >[...]
> > In my view, any switching mixer *must* create both sum and difference
> > frequencies. (Indeed, a switching mixer *also* will create sum and
>
> It removes one of them with phase network after.
> Speaking about phase networks - I need to shift Q by 90 degrees in range
> 300 Hz to 3000 Hz and get orthogonal axes on XY plot. All particular
> frequencies are sine waves. When you do shift sine wave by 90 degrees
> you get cosine wave. cos(x) is first derivation of sin(x). Classic
> derivator won't work here - that is why we use networks.
> Somebody mentioned that he used PC to do that. OK, sample I on left and
> Q on right and then you derivate right and sum that two ones properly.
> Is it so simple or I am wrong in something?
You are right, you sample I on the left and Q on the right. I then do a
Complex Fast Fourier Transform on the input and do the filtering and
sideband selection in the frequency domain. Since the Tayloe Detector gives
I & Q directly all you need to do is to use low noise amplifiers to drive
the sound card. Very simple.
>
> If it is, I can do so with PIC 16F874.
Unlikely. You need more horsepower than a PIC would have. Todays PCs with
sound cards are much easier to deal with and are extremely fast.
Sverre Holm
> > Somebody mentioned that he used PC to do that. OK, sample I on left and
> > Q on right and then you derivate right and sum that two ones properly.
> > Is it so simple or I am wrong in something?
>
> You are right, you sample I on the left and Q on the right. I then do a
> Complex Fast Fourier Transform on the input and do the filtering and
> sideband selection in the frequency domain. Since the Tayloe Detector gives
> I & Q directly all you need to do is to use low noise amplifiers to drive
> the sound card. Very simple.
>
The filtering in the frequency domain is one approximation to a
Hilbert transform, which is really what is required here.
Sverre
______________________________________________________________________
73 de LA3ZA
http://www.qsl.net/la3za
Clifton T. Sharp Jr.
> Next you
> should read an article by Richard Lyons called Quadrature Signals: Complex,
> But Not Complicated, which can be found at
> http://www.dspguru.com/info/tutor/quadsig.htm on the www.dspguru.com site.
It sure complicated things for me. I always thought quadrature signals
were 90 degrees different in phase.
Gerald Youngblood
The Q signal is shifted 90 degrees in phase counter clockwise from the I
signal.
Gerald
"Clifton T. Sharp Jr." <cli...@clifto.com> wrote in message
news:3B3C996A...@clifto.com...
Clifton T. Sharp Jr.
> "Clifton T. Sharp Jr." <cli...@clifto.com> wrote...
> > Gerald Youngblood wrote:
> > > Next you
> > > should read an article by Richard Lyons called Quadrature Signals:
> Complex,
> > > But Not Complicated, which can be found at
> > > http://www.dspguru.com/info/tutor/quadsig.htm on the www.dspguru.com
> site.
> >
> > It sure complicated things for me. I always thought quadrature signals
> > were 90 degrees different in phase.
>
> signal.
Yup. But what the author is calling "quadrature signals" actually differ
by varying amounts depending on phase angle:
"Let's now call our two [exp(j*2*pi*Fo*t)] and [exp(-j*2*pi*Fo*t)] complex
expressions _quadrature signals_."
The way he diagrams the latter expression on the preceding page, I can't
even conceive of a real-world signal that would do what he describes.
He also says that "If you have a single complex number, represented by a
point on the complex plane, multiplying that number by j OR [exp(j*pi/2)]
will result in a new complex number that's rotated 90 degrees counterclockwise
on the complex plane." Now, *that* is quadrature, the way I learned it.
But the two expressions [exp(j*2*pi*Fo*t)] and [exp(-j*2*pi*Fo*t)] that
he calls quadrature do NOT differ by a factor of j, but by j^2.
So basically, neither his diagram nor his math appears (to me) to be
describing quadrature signals, while he's explicitly stating they are.
And I still have no concept of a real signal described by a phasor
rotating clockwise.
--
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Clifton T. Sharp Jr.
> He also says that "If you have a single complex number, represented by a
> point on the complex plane, multiplying that number by j OR [exp(j*pi/2)]
> will result in a new complex number that's rotated 90 degrees counterclockwise
> on the complex plane." Now, *that* is quadrature, the way I learned it.
> But the two expressions [exp(j*2*pi*Fo*t)] and [exp(-j*2*pi*Fo*t)] that
> he calls quadrature do NOT differ by a factor of j, but by j^2.
Oops. I knew I should come back and look at this. That's silly. But they
still aren't one multiplied by j (or exp(j*pi/2)) to give the other, so
it seems safe to say they aren't a constant 90 degrees different in phase.
smf...@gmail.com
http://www.b-kainka.de/SSBtrx.pdf
Esp. look at the output of the 6kHz RC-generator (terminals H,I,J,K) fed into a quadrature mixer/detector on the left hand side of the schematic.