YAQ - Yet Another (K3) Question

[Bill Coleman-2]

On Jun 19, 2007, at 9:42 AM, Brian Lloyd wrote:

> Not true. The K3 converts from A:D as a step in the second IF. The  
> final filtering takes place in the K3's DSP as does the demodulation.
>
> If you then go to a soundcard you are now adding another conversion  
> and a third IF. The K3's second IF (DSP) does its conversion and  
> filtering but then does a conversion to a third IF, including a  
> conversion from digital back to analog. We tend think of this as  
> audio but it really is not. It is a third IF with no control over  
> the AGC. The final amplification, filtering, and demodulation now  
> takes place in the computer's DSP.

One way of simplifying this whole process is to move the radio into  
the computer. This is in line with a prediction I made way back in  
1995 on the CQ-Contest list -- that the radio of the future would be  
inside the computer. In 1995, processors were just appearing that had  
sufficient DSP capability to do this. 12 years later, it would take a  
relatively insignificant portion of the main CPU (or just a portion  
of a few cores, as multi-core machines are now common).

What you'd end up with for the "receiver" would be a Mixer and clean  
DDS, followed by a high-speed, wide-range A/D converter. Everything  
else would be done in the host computer. The "transmitter" would go  
the opposite way, a D/A converter followed by a mixer fed by a DDS.  
Power amplification could be external to the computer (if the  
transceiver were a card).

The interesting part of this approach is that we can re-define what  
we mean by a receiver. The detection portion of the radio need not  
resolve to the width of an audio channel. Consider a receiver that  
can decode every CW signal in a 50 kHz portion of the band.  
Simultaneously. How useful would that be?

It also would be good to sell the receiver and transmitters  
separately. That way, obtaining the two receiver, one transmitter  
configuration needed by SO2R operation could be inexpensively obtained.

Of course, to achieve the IMD and dynamic range of the K3, the mixer  
and A/D would be pretty marvelous pieces of equipment.

> Let me put it another way: one of the reasons that the Elecraft  
> receivers work so well is that they do fewer conversions and use  
> lower IF frequencies so that they can put good filtering as far  
> forward in the chain as possible. This gets rid of products that  
> could cause IMD in later stages.

It's still possible to get good IMD characteristics with an up-
conversion general-coverage receiver. There are some $10,000 radios  
on the market that do exactly this.

Elecraft has apparently mastered the art of offering high-performance  
gear at an excellent price point.

> I want a whole boatload of demodulators there in the K3's DSP with  
> access coming out to me in some convenient fashion -- like on an  
> ethernet connector.

Sounds like what you really want is something more like the 1995 pipe  
dream.

Bill Coleman, AA4LR, PP-ASEL        Mail: [hidden email]
Quote: "Not within a thousand years will man ever fly!"
             -- Wilbur Wright, 1901

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[David Woolley (E.L)]

Bill Coleman wrote:

> What you'd end up with for the "receiver" would be a Mixer and clean
> DDS, followed by a high-speed, wide-range A/D converter. Everything else

I don't understand why you write about such things as in the future.
There are several products on the amateur market, although I don't know
if any of them use a DDS VFO.

> The interesting part of this approach is that we can re-define what we
> mean by a receiver. The detection portion of the radio need not resolve
> to the width of an audio channel. Consider a receiver that can decode
> every CW signal in a 50 kHz portion of the band. Simultaneously. How
> useful would that be?

The telcos where doing direct digital conversion from analogue carrier
systems to multiple PCM channels a long time ago.  I think before your
1995 article.


--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Julian, G4ILO]

On 7/9/07, Bill Coleman <[hidden email]> wrote:
>
> One way of simplifying this whole process is to move the radio into
> the computer. This is in line with a prediction I made way back in
> 1995 on the CQ-Contest list -- that the radio of the future would be
> inside the computer.

I hope not. Clearly it's an option, and I can see the attraction for
the experimenter who wants to be able to make mods and recompile their
radio. The folks who buy FlexRadios seem to like the idea. But they
are a pretty small minority.

Computers are very versatile, but being general purpose tools they
don't often do a specific job better than dedicated hardware.

Radios don't crash, get infected by viruses or spyware or need
critical updates every couple of weeks. They are more portable and use
less power than computers - something we should all think about in
these environmentally conscious days.

Also I personally prefer to operate real radio controls than click
buttons on a computer screen.
--
Julian, G4ILO
G4ILO's Shack: www.g4ilo.com
K2 s/n: 392  K3 s/n: ???
www.Ham-Directory.com: the best ham resources on the net
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Brian Lloyd-6]

> One way of simplifying this whole process is to move the radio into  
> the computer. This is in line with a prediction I made way back in  
> 1995 on the CQ-Contest list -- that the radio of the future would  
> be inside the computer. In 1995, processors were just appearing  
> that had sufficient DSP capability to do this. 12 years later, it  
> would take a relatively insignificant portion of the main CPU (or  
> just a portion of a few cores, as multi-core machines are now common).
>
> What you'd end up with for the "receiver" would be a Mixer and  
> clean DDS, followed by a high-speed, wide-range A/D converter.  
> Everything else would be done in the host computer. The  
> "transmitter" would go the opposite way, a D/A converter followed  
> by a mixer fed by a DDS. Power amplification could be external to  
> the computer (if the transceiver were a card).
>
> The interesting part of this approach is that we can re-define what  
> we mean by a receiver. The detection portion of the radio need not  
> resolve to the width of an audio channel. Consider a receiver that  
> can decode every CW signal in a 50 kHz portion of the band.  
> Simultaneously. How useful would that be?

Well, you have just described the product offerings from Flex Radio.  
They are certainly interesting competitors to Elecraft. They are a  
completely different approach to constructing the radio. I am not  
convinced that their approach is better than Elecraft's but they are  
certainly interesting.


> It also would be good to sell the receiver and transmitters  
> separately. That way, obtaining the two receiver, one transmitter  
> configuration needed by SO2R operation could be inexpensively  
> obtained.

Or they could be multiple boards in the same chassis.

> Of course, to achieve the IMD and dynamic range of the K3, the  
> mixer and A/D would be pretty marvelous pieces of equipment.

Elecraft has exactly the same issues in the K3.

My concern over the Flex Radio SDR approach compared to Elecraft's  
approach in the K3 is that, in order to be able to receive multiple  
signals simultaneously, e.g. like we do in demodulating PK31, you  
have to accept all the noise and cruft in the wider passband. If  
there is a strong signal in there you have to pass it through to the  
A:D and hope that the A:D has sufficient dynamic range to deal with  
the difference between the desired signal and the undesired signal.  
Elecraft gets rid of the undesired signal by using tight roofing  
filters.

Basically you make you choice and accept the limitations. Elecraft  
has optimized for reception of a single signal. Want the ultimate in  
CW reception? I think that the K3 is probably the winner. Want the  
ability to demodulate several signals at once or do some new wideband  
mode? I think that you have to look at the Flex Radio offerings. But  
you give up performance in one area to get performance in the other.

There ain't no such thing as a free lunch.

>> Let me put it another way: one of the reasons that the Elecraft  
>> receivers work so well is that they do fewer conversions and use  
>> lower IF frequencies so that they can put good filtering as far  
>> forward in the chain as possible. This gets rid of products that  
>> could cause IMD in later stages.
>
> It's still possible to get good IMD characteristics with an up-
> conversion general-coverage receiver. There are some $10,000 radios  
> on the market that do exactly this.

But all of them upconvert to something like a 70MHz 1st IF. You  
aren't going to find a 200Hz roofing filter there. That means you  
aren't going to get the good close-in (1KHz spacing) IMD and BDR  
performance. So to get general coverage receiver performance you give  
up close-in BDR and IMD performance. Again TANSTAAFL.

> Elecraft has apparently mastered the art of offering high-
> performance gear at an excellent price point.

I agree. Elecraft should be receiving the order from the ARRL for our  
school's K2. I plan to let the kids (4th-8th grades) build the rig  
under my guidance. I think that the K2 will perform a lot better than  
the other rigs that they were offering us, e.g. Icom IC-706, and I  
think that the kids will understand and appreciate the radio better  
if they have a hand in building, testing, and calibrating it.  
(Besides, it will dovetail nicely with my "this is how a radio works"  
section in science class.)

>> I want a whole boatload of demodulators there in the K3's DSP with  
>> access coming out to me in some convenient fashion -- like on an  
>> ethernet connector.
>
> Sounds like what you really want is something more like the 1995  
> pipe dream.

Well, it is not a pipe-dream anymore. You can have it. It all depends  
on what parameters you want to optimize for.

>
> Bill Coleman, AA4LR, PP-ASEL        Mail: [hidden email]
> Quote: "Not within a thousand years will man ever fly!"
>             -- Wilbur Wright, 1901
>
>

73 de Brian, WB6RQN
Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Bill Coleman-2]

On Jul 9, 2007, at 2:59 AM, David Woolley wrote:

> Bill Coleman wrote:
>
>> What you'd end up with for the "receiver" would be a Mixer and  
>> clean DDS, followed by a high-speed, wide-range A/D converter.  
>> Everything else
>
> I don't understand why you write about such things as in the  
> future. There are several products on the amateur market, although  
> I don't know if any of them use a DDS VFO.

What I wrote about in 1995 was the entire receiver as a very simple  
bit of hardware, and then lots of software.

I've seen several attempts at computer/radio integration, but they  
all used far more hardware than I imagined. That's why I used the  
future tense.

> The telcos where doing direct digital conversion from analogue  
> carrier systems to multiple PCM channels a long time ago.  I think  
> before your 1995 article.

The technology has been around a while. It's just waiting to be  
applied to amateur radio....

Bill Coleman, AA4LR, PP-ASEL        Mail: [hidden email]
Quote: "Not within a thousand years will man ever fly!"
             -- Wilbur Wright, 1901

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Bill Coleman-2]

On Jul 9, 2007, at 4:44 PM, Brian Lloyd wrote:
>
> Well, you have just described the product offerings from Flex Radio.

Not exactly what I had in mind -- there's way too much hardware  
involved.

> My concern over the Flex Radio SDR approach compared to Elecraft's  
> approach in the K3 is that, in order to be able to receive multiple  
> signals simultaneously, e.g. like we do in demodulating PK31, you  
> have to accept all the noise and cruft in the wider passband. If  
> there is a strong signal in there you have to pass it through to  
> the A:D and hope that the A:D has sufficient dynamic range to deal  
> with the difference between the desired signal and the undesired  
> signal.

No. You don't "hope" the A/D has enough dynamic range. You have to  
design in that kind of dynamic range. Which means you've got to use  
an A/D converter which has lots of bits -- very expensive today.

> Elecraft gets rid of the undesired signal by using tight roofing  
> filters.

Yup, that's one approach. It also is very cost-effective for a single-
signal type receiver.

>> It's still possible to get good IMD characteristics with an up-
>> conversion general-coverage receiver. There are some $10,000  
>> radios on the market that do exactly this.
>
> But all of them upconvert to something like a 70MHz 1st IF. You  
> aren't going to find a 200Hz roofing filter there. That means you  
> aren't going to get the good close-in (1KHz spacing) IMD and BDR  
> performance. So to get general coverage receiver performance you  
> give up close-in BDR and IMD performance. Again TANSTAAFL.

Consider the Yaesu FT DX 9000. This design has a 40 MHz first IF that  
has roofing filters of 3, 6 and 15 kHz.

Similarly, the Icom IC-7800. This design has a first IF of 64.455 MHz  
and roofing filters of 3, 6 and 15 kHz.

Both of these designs have competitive IMD and BDR characteristics.  
It is possible to do this with wider roofing filters in an up-
conversion design. However, it isn't as cost-effective as the  
Elecraft design. (I can buy several fully-loaded K3s for the price of  
one FT DX 9000 or IC-7800)

> I agree. Elecraft should be receiving the order from the ARRL for  
> our school's K2. I plan to let the kids (4th-8th grades) build the  
> rig under my guidance. I think that the K2 will perform a lot  
> better than the other rigs that they were offering us, e.g. Icom  
> IC-706, and I think that the kids will understand and appreciate  
> the radio better if they have a hand in building, testing, and  
> calibrating it.

I totally agree. When I was first licensed, one of my dreams was to  
design and build a state of the art homebrew transceiver. For a  
number of reasons, I never quite did that, but with the Elecraft  
K2/100, I've come close. (I didn't design it, but I certainly built it)


Bill Coleman, AA4LR, PP-ASEL        Mail: [hidden email]
Quote: "Not within a thousand years will man ever fly!"
             -- Wilbur Wright, 1901

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com