Discussion:
[Elecraft] Elecraft technology >> K2+
Bill Tippett
2003-12-27 16:12:01 UTC
Permalink
In one way, the K2 goes back to even earlier, single-band superhet
design. That's why its dynamic range figures are so good. Most rig
manufacturers aren't likely to move back to this simpler design, largely
because they don't want to give up general coverage receive.
I hope your last statement is wrong. I believe the K2 and Orion's
Main RX
are so superior that they may cause some manufacturers to rethink the need for
general coverage. Elecraft's K2 design has the basics of an excellent contest
radio, but it needs attention in two main areas to be truly competitive:

1. Ergonomics...for someone whose hand span exceeds the entire front panel
width by 1.5" and with finger size to match....for someone who typically spends
hours in front of the radio in a contest and wants a minimum of menu changes
(e.g. single-touch access to different filter BW's, etc)...panel readability
for someone whose eyesight is not what it once was.

2. Replacing the programmable crystal filter with fixed BW 8-pole filters
and <2:1 shape factors (quite achievable at the K2's IF). The current filter
has a ~3:1 shape factor which may be one reason the K2's performance degrades
for IMD and BDR at signal spacings under 5 kHz. From 5 kHz to 1 kHz signal
spacings, IMD goes from 88 to 67 dB and BDR goes from 124 dB TO 116 dB. That
data is from the swept graphs on pages 19 and 20 of ARRL's Expanded Test
Report
done in 2000...pending update of ETR data for the new K2/100 review.

http://www2.arrl.org/members-only/prodrev/pdf/lab/k2.pdf

With the addition of a K2+ model for performance-oriented contesters
and DXers in the <$2k range, I believe Elecraft could dramatically expand their
market while maintaining their original target market for the K2. Someone is
going to do this eventually, and it might as well be Elecraft since the K2 is
already very close to what is needed IMHO.

73, Bill W4ZV




--- StripMime Report -- processed MIME parts ---
multipart/mixed
text/plain (text body -- kept)
The reason this message is shown is because the post was in HTML
or had an attachment. Attachments are not allowed. To learn how
to post in Plain-Text go to: http://www.expita.com/nomime.html ---
N***@aol.com
2003-12-27 17:07:00 UTC
Permalink
In a message dated 12/27/03 10:08:42 AM Eastern Standard Time,
Post by Bill Tippett
In one way, the K2 goes back to even earlier, single-band superhet
design. That's why its dynamic range figures are so good. Most rig
manufacturers aren't likely to move back to this simpler design, largely
because they don't want to give up general coverage receive.
Just noticed a typo in that paragraph - should be "single conversion" not
"single band"
Post by Bill Tippett
I hope your last statement is wrong. I believe the K2 and Orion's
Main RX
are so superior that they may cause some manufacturers to rethink the need
for
general coverage.
I hope so too but it's not me you have to convince ;-) Note that the basic
engineering of the upconversion-to-VHF-first-IF scheme is amortized (by Ikensu)
over many models - some of them not even ham rigs.

Elecraft's K2 design has the basics of an excellent contest
Post by Bill Tippett
1. Ergonomics...for someone whose hand span exceeds the entire front panel
width by 1.5" and with finger size to match....for someone who typically
spends
hours in front of the radio in a contest and wants a minimum of menu changes
(e.g. single-touch access to different filter BW's, etc)...panel readability
for someone whose eyesight is not what it once was.
Agreed! They don't call me "Big Paws" for nothin'! In fact, I have actually
thought about putting K2 innards in a bigger box so that much bigger knobs
could be used on the controls, and they could be spaced farther apart. Such a
change is not beyond the scope of an experienced homebrewer. All that would be
needed is a replacement front panel board that would be ribbon-cabled to the
control board.

The display is another story, though, because it's a unique part and I doubt
there is a larger equivalent. So now we're talking major mods to both hardware
and firmware.
Post by Bill Tippett
2. Replacing the programmable crystal filter with fixed BW 8-pole filters
and <2:1 shape factors (quite achievable at the K2's IF). The current
filter
has a ~3:1 shape factor which may be one reason the K2's performance
degrades
for IMD and BDR at signal spacings under 5 kHz. From 5 kHz to 1 kHz signal
spacings, IMD goes from 88 to 67 dB and BDR goes from 124 dB TO 116 dB.
That
data is from the swept graphs on pages 19 and 20 of ARRL's Expanded Test
Report
done in 2000...pending update of ETR data for the new K2/100 review.
http://www2.arrl.org/members-only/prodrev/pdf/lab/k2.pdf
Doing that would require major redo of the RF board, plus additional IF gain
to make up for the additional loss in the filters. I'd suggest relay (not
diode) switching, too. And perhaps at least a 4 crystal second filter.

I'd also suggest:

3) Space in the cabinet for all possible options, and a different heatsink
arrangement for the PA transistors.

Some folks would like passband tuning, but that's even more of a change to
the RF board.

Call it "K2BIG"...


The problem with all these changes is that the cost will rise to the point
where it becomes more of a midline transceiver and less of the lowcost
highperformance take-anywhere alternative rig that has been the Elecraft paradigm since
the beginning. Heck, I find it truly amazing that an under-$600-base-price
kit rig can even be considered in the same discussion with an Orion or FT-1K.
Post by Bill Tippett
With the addition of a K2+ model for performance-oriented
contesters
and DXers in the <$2k range, I believe Elecraft could dramatically expand
their
market while maintaining their original target market for the K2. Someone
is
going to do this eventually, and it might as well be Elecraft since the K2
is
already very close to what is needed IMHO.
Well, we can ask....

73 de Jim, N2EY



--- StripMime Report -- processed MIME parts ---
multipart/alternative
text/plain (text body -- kept)
text/html
The reason this message is shown is because the post was in HTML
or had an attachment. Attachments are not allowed. To learn how
to post in Plain-Text go to: http://www.expita.com/nomime.html ---
George, W5YR
2003-12-27 17:46:00 UTC
Permalink
In the interim, you can use an external receiver as a downconverter, as I
recently posted re the Icom PRO2. This provides the IF DSP capabilities that
the K2 lacks, notably effective notch filters and IF filters with very low
shape factors.

73, George W5YR
***@att.net


----- Original Message -----
From: "Bill Tippett" <***@alum.mit.edu>
To: <***@mailman.qth.net>
Sent: Saturday, December 27, 2003 9:06 AM
Subject: [Elecraft] Elecraft technology >> K2+
Post by Bill Tippett
In one way, the K2 goes back to even earlier, single-band superhet
design. That's why its dynamic range figures are so good. Most rig
manufacturers aren't likely to move back to this simpler design, largely
because they don't want to give up general coverage receive.
I hope your last statement is wrong. I believe the K2 and Orion's
Main RX
are so superior that they may cause some manufacturers to rethink the need for
general coverage. Elecraft's K2 design has the basics of an excellent contest
1. Ergonomics...for someone whose hand span exceeds the entire front panel
width by 1.5" and with finger size to match....for someone who typically spends
hours in front of the radio in a contest and wants a minimum of menu changes
(e.g. single-touch access to different filter BW's, etc)...panel readability
for someone whose eyesight is not what it once was.
2. Replacing the programmable crystal filter with fixed BW 8-pole filters
and <2:1 shape factors (quite achievable at the K2's IF). The current filter
has a ~3:1 shape factor which may be one reason the K2's performance degrades
for IMD and BDR at signal spacings under 5 kHz. From 5 kHz to 1 kHz signal
spacings, IMD goes from 88 to 67 dB and BDR goes from 124 dB TO 116 dB.
That
Post by Bill Tippett
data is from the swept graphs on pages 19 and 20 of ARRL's Expanded Test
Report
done in 2000...pending update of ETR data for the new K2/100 review.
http://www2.arrl.org/members-only/prodrev/pdf/lab/k2.pdf
With the addition of a K2+ model for performance-oriented contesters
and DXers in the <$2k range, I believe Elecraft could dramatically expand their
market while maintaining their original target market for the K2. Someone is
going to do this eventually, and it might as well be Elecraft since the K2 is
already very close to what is needed IMHO.
73, Bill W4ZV
--- StripMime Report -- processed MIME parts ---
multipart/mixed
text/plain (text body -- kept)
The reason this message is shown is because the post was in HTML
or had an attachment. Attachments are not allowed. To learn how
to post in Plain-Text go to: http://www.expita.com/nomime.html ---
_______________________________________________
You must be subscribed to post to the list.
http://mailman.qth.net/mailman/listinfo/elecraft
Post by Bill Tippett
Elecraft Web Page: http://www.elecraft.com
Also see: http://www.elecraft.com/elecraft_list_guidelines.htm
Bill Tippett
2003-12-27 18:28:00 UTC
Permalink
Post by George, W5YR
In the interim, you can use an external receiver as a downconverter, as I
recently posted re the Icom PRO2. This provides the IF DSP capabilities that
the K2 lacks, notably effective notch filters and IF filters with very low
shape factors.
George, that would definitely not pass my ergonomics test! Not
one but two radios in tandem to keep track of when I'm already sleep-
deprived in a contest? No thanks. The K2 really doesn't need additional
IF DSP capability IMHO. It simply needs slightly better filtering on the
front end and a more user-friendly ergonomic design.

73, Bill W4ZV
Bill Coleman
2003-12-28 23:09:01 UTC
Permalink
Post by Bill Tippett
In one way, the K2 goes back to even earlier, single-band superhet
design. That's why its dynamic range figures are so good. Most rig
manufacturers aren't likely to move back to this simpler design, largely
because they don't want to give up general coverage receive.
I hope your last statement is wrong. I believe the K2 and Orion's
Main RX
are so superior that they may cause some manufacturers to rethink the need
for
general coverage.
I don't see the Japanese manufacturers giving up general coverage
receiption. If anything, we've seen the opposite trend. Wide receivers
have also weakened the performance of VHF and UHF receivers, making them
more subsceptable to intermod.

The Omni V showed that amatuer-only can give superior performance many
years ago, yet there has been no change.
Post by Bill Tippett
Elecraft's K2 design has the basics of an excellent
contest
1. Ergonomics...for someone whose hand span exceeds the entire front panel
width by 1.5" and with finger size to match....for someone who typically
spends
hours in front of the radio in a contest and wants a minimum of menu changes
(e.g. single-touch access to different filter BW's, etc)...panel readability
for someone whose eyesight is not what it once was.
Realistically, there are only a few things that the K2 needs to improve
in its front panel:

* current filter indication on display
* one-button access to the Rx antenna port. (toggle)
* DSP indication on display and easy-access controls
Post by Bill Tippett
2. Replacing the programmable crystal filter with fixed BW 8-pole filters
and <2:1 shape factors (quite achievable at the K2's IF). The current filter
has a ~3:1 shape factor which may be one reason the K2's performance degrades
for IMD and BDR at signal spacings under 5 kHz.
In my experience, the K2 CW filters, while they don't have the best shape
factor, perform admirably because of the other fine characteristics of
the receiver, particularly the lack of IMD compared to a number of
transceivers with wideband PLL noise.

The SSB filter could definitely be improved, although it isn't bad, even
for contest use. The DSP cleans it up quite a bit.
Post by Bill Tippett
With the addition of a K2+ model for performance-oriented contesters
and DXers in the <$2k range, I believe Elecraft could dramatically expand
their
market while maintaining their original target market for the K2. Someone is
going to do this eventually, and it might as well be Elecraft since the K2 is
already very close to what is needed IMHO.
Once can't increase the front panel of the K2 much without radically
changing the construction technique used. It might work to put in one
more column of buttons and a slightly wider display, or a few low-power
LEDs.

One difficulty with a contest-oriented K2+ is that the costs of the
changes would have to be bourne only by K2+ sales, not over the K2 kit in
general. The K2 is currently a great compromise between the QRP,
battery-powered, backpacking utility radio and a overall HF radio.

Bill Coleman, AA4LR, PP-ASEL Mail: ***@arrl.net
Quote: "Not within a thousand years will man ever fly!"
-- Wilbur Wright, 1901
Mike McCoy
2003-12-29 14:57:01 UTC
Permalink
Post by Bill Coleman
Realistically, there are only a few things that the K2 needs to
* current filter indication on display
* one-button access to the Rx antenna port. (toggle)
* DSP indication on display and easy-access controls

Although its power requirements preclude any portable operation that =
last point is the primary reason I've decided not to replace my Timewave =
599ZX with a KDSP2.

I'm not a big fan of 'toggling' thru filters ala AFIL/XFIL ('oops, that =
last selection was the best choice'). I guess I prefer the 599ZX being =
external and adjusting the 599ZX's 'infinite' variability using front =
panel knobs/switches and having the settings displayed on the Timewaves =
LCD and/or LED's. Plus it fits perfectly on top of the K2 almost as if =
it were made for it (pretty much same width/depth as the K2).

Not to disparage the KDSP2 though... I am absolutely in awe of Mr. =
Johnsons efforts as well as Elecraft offering it as an official option =
(no 'NIH' evident at Elecraft!). I'm sure one day I will *add* the KDSP2 =
for the sake of portability.

Mike, K5PU



--- StripMime Report -- processed MIME parts ---
multipart/alternative
text/plain (text body -- kept)
text/html
The reason this message is shown is because the post was in HTML
or had an attachment. Attachments are not allowed. To learn how
to post in Plain-Text go to: http://www.expita.com/nomime.html ---
Vic Rosenthal
2003-12-29 17:51:59 UTC
Permalink
Post by Mike McCoy
I'm not a big fan of 'toggling' thru filters ala AFIL/XFIL ('oops,
that last selection was the best choice'). I guess I prefer the 599ZX
being external and adjusting the 599ZX's 'infinite' variability using
front panel knobs/switches and having the settings displayed on the
Timewaves LCD and/or LED's.
A couple of reasons why I prefer the Elecraft DSP to my Timewave 59+
(not as fancy as the 599ZX):

1) Sidetone is is injected in the K2 after the DSP, so there's no
trouble with delay or with the sidetone being attenuated by the DSP
filter. I know the Timewave units can be switched out on transmit, but
it isn't practical with high-speed QSK.

2) The denoiser in the Elecraft DSP works better than the one in my 59+.

Having said that, I agree with Mike's comments. Perhaps the K2+ or K3
needs a bigger panel and some old fashioned switches instead of buttons.
I also agree with the previous poster's comments about the need for a
better IF filter. I have a TS850s with a pair of 400 Hz INRAD filters
and I miss the IF filter performance in the K2 (although I prefer the K2
overall and the 850 is now my standby rig).
--
73,
Vic, K2VCO
Fresno CA
http://www.qsl.net/k2vco
George, W5YR
2003-12-30 03:26:00 UTC
Permalink
Vic, connect a 22 pF cap to the Noise Blanker input and bring out the other
lead with RG-174 to one of the Xverter hole into which you have installed a
BNC connector. Run a shielded cable from that BNC to your 850 antenna
connector. Set the 850 to around 4915 kHz and tune around until you pick up
the K2 signal. Get it tuned properly and you will have all the features of
the 850 to use with the K2 front end. With my Icom PRO2, I get a Spectrum
Scope display of the K2 signal.

A better tap for downconverting to an external receiver is the output of the
AGC amp through a 0.005 uF cap in series with a 47K resistor to RG-174 to a
BNC installed in the other Xverter hole. The output here is around 154 kHz.

With either tap, tune the external receiver until it produces the same sound
as the K2; i.e., CW tones are the same, SSB voice sounds the same, etc.

Depending upon the band and mode, the external receiver may be tuned to the
same mode as the K2 and it may also have to be tuned to the "other" mode,
depending upon your external receiver. For example, on some bands and modes,
CW on the K2 is CW-R on the other radio. Or USB on the K2 may be received as
LSB on the other radio.

I find that this is an excellent addition to the K2 for in-house operation.
I have all the DSP IF tools of the PRO2 available, including excellent notch
filters and 50 Hz IF filters if needed.

These taps have no measurable effect on the K2 operation and performance as
far as I can determine. The 47K resistor in the AGC tap point was chosen to
make the PRO2 S-meter show about the same reading as the K2 meter.

If you try any of this, I would interested in hearing the results. O guess
we can call this the "K2+" . . .

73, George W5YR
***@att.net


----- Original Message -----
From: "Vic Rosenthal" <***@rakefet.com>
To: "Elecraft" <***@mailman.qth.net>
Sent: Monday, December 29, 2003 10:47 AM
Subject: Re: [Elecraft] Elecraft technology >> K2+
Post by Vic Rosenthal
Post by Mike McCoy
I'm not a big fan of 'toggling' thru filters ala AFIL/XFIL ('oops,
that last selection was the best choice'). I guess I prefer the 599ZX
being external and adjusting the 599ZX's 'infinite' variability using
front panel knobs/switches and having the settings displayed on the
Timewaves LCD and/or LED's.
A couple of reasons why I prefer the Elecraft DSP to my Timewave 59+
1) Sidetone is is injected in the K2 after the DSP, so there's no
trouble with delay or with the sidetone being attenuated by the DSP
filter. I know the Timewave units can be switched out on transmit, but
it isn't practical with high-speed QSK.
2) The denoiser in the Elecraft DSP works better than the one in my 59+.
Having said that, I agree with Mike's comments. Perhaps the K2+ or K3
needs a bigger panel and some old fashioned switches instead of buttons.
I also agree with the previous poster's comments about the need for a
better IF filter. I have a TS850s with a pair of 400 Hz INRAD filters
and I miss the IF filter performance in the K2 (although I prefer the K2
overall and the 850 is now my standby rig).
--
73,
Vic, K2VCO
Fresno CA
http://www.qsl.net/k2vco
Floyd Sense" (Floyd Sense)
2003-12-29 18:05:05 UTC
Permalink
Since you brought up the 599zx, I'd like to ask question regarding the
KDSP2. As past owner of three different DSP units, including the 599zx, I
know that not all DSPs are created equal. While the 599zx had a lot of nice
features, I found that the auto-notch function made voices very hard to
understand. My current DSP is an MFJ-874B, whose auto-notch function uses a
different approach resulting in very little distortion to voices, but it
takes a fraction of a second for an interfering beat note to be cancelled.
My question is: does the KDSP2 auto-notch function degrade voice readability
or does it function more like the MFJ?

73, Floyd - K8AC


----- Original Message -----
From: "Mike McCoy" <***@austin.rr.com>
<<snip>>
Although its power requirements preclude any portable operation that last
point is the primary reason I've decided not to replace my Timewave 599ZX
with a KDSP2.

I'm not a big fan of 'toggling' thru filters ala AFIL/XFIL ('oops, that last
selection was the best choice'). I guess I prefer the 599ZX being external
and adjusting the 599ZX's 'infinite' variability using front panel
knobs/switches and having the settings displayed on the Timewaves LCD and/or
LED's. Plus it fits perfectly on top of the K2 almost as if it were made for
it (pretty much same width/depth as the K2).

Not to disparage the KDSP2 though... I am absolutely in awe of Mr. Johnsons
efforts as well as Elecraft offering it as an official option (no 'NIH'
evident at Elecraft!). I'm sure one day I will *add* the KDSP2 for the sake
of portability.

Mike, K5PU
Charles Greene
2003-12-29 22:34:00 UTC
Permalink
Mike and all,

I use my Timewave 599ZX with K2 492 which has no KDSP2 but which has the
KAF2. K2/100 3571 has the KDSP2. I have mixed emotions about which is
better and which is more convenient. Actually, I have found the KDSP2 is
not as good as the Timewave in noise reduction, but once you have the
filters set up for SSB or CW, the KDSP2 is more convenient, in my opinion,
and of course is more portable.
Post by Bill Coleman
Post by Bill Coleman
Realistically, there are only a few things that the K2 needs to
* current filter indication on display
* one-button access to the Rx antenna port. (toggle)
* DSP indication on display and easy-access controls
Although its power requirements preclude any portable operation that last
point is the primary reason I've decided not to replace my Timewave 599ZX
with a KDSP2.
I'm not a big fan of 'toggling' thru filters ala AFIL/XFIL ('oops, that
last selection was the best choice'). I guess I prefer the 599ZX being
external and adjusting the 599ZX's 'infinite' variability using front
panel knobs/switches and having the settings displayed on the Timewaves
LCD and/or LED's. Plus it fits perfectly on top of the K2 almost as if it
were made for it (pretty much same width/depth as the K2).
Not to disparage the KDSP2 though... I am absolutely in awe of Mr.
Johnsons efforts as well as Elecraft offering it as an official option (no
'NIH' evident at Elecraft!). I'm sure one day I will *add* the KDSP2 for
the sake of portability.
Mike, K5PU
73, Chas, W1CG
K2 462b, 3571
Phil Wheeler
2003-12-30 00:18:11 UTC
Permalink
Funny how tastes differ in this DSP arena. At one time I owned a 599ZX
and an NIR-12 which I A-B tested extensively. Kept the NIR-12 but no
longer use it much: Very satisfied with the KDSP2.

Phil
Post by Charles Greene
Mike and all,
I use my Timewave 599ZX with K2 492 which has no KDSP2 but which has
the KAF2. K2/100 3571 has the KDSP2. I have mixed emotions about
which is better and which is more convenient. Actually, I have found
the KDSP2 is not as good as the Timewave in noise reduction, but once
you have the filters set up for SSB or CW, the KDSP2 is more
convenient, in my opinion, and of course is more portable.
Post by Bill Coleman
Post by Bill Coleman
Realistically, there are only a few things that the K2 needs to
* current filter indication on display
* one-button access to the Rx antenna port. (toggle)
* DSP indication on display and easy-access controls
Although its power requirements preclude any portable operation that
last point is the primary reason I've decided not to replace my
Timewave 599ZX with a KDSP2.
I'm not a big fan of 'toggling' thru filters ala AFIL/XFIL ('oops,
that last selection was the best choice'). I guess I prefer the 599ZX
being external and adjusting the 599ZX's 'infinite' variability using
front panel knobs/switches and having the settings displayed on the
Timewaves LCD and/or LED's. Plus it fits perfectly on top of the K2
almost as if it were made for it (pretty much same width/depth as the
K2).
George, W5YR
2003-12-30 03:43:00 UTC
Permalink
Phil and all,

I also have the NIR-12 and find that as a DSP unit, it does a superb job.
Since it has dual 40 MHz DSP chips, it is able to perform all functions
concurrently: filtering, notching and nose reduction. Plus having an audio
amp to feed a speaker.

The NIR-12 is another item to watch for on Ebay and at hamfests. They were
expensive when they were available. I paid around $300 for mine as I recall,
but it was at the time the best investment I had made in equipment for the
results it produced. It also fits neatly on top of the K2 and its solid
black color is in contrast to the grey of the K2.

73, George W5YR
***@att.net


----- Original Message -----
From: "Phil Wheeler" <***@socal.rr.com>
To: "Charles Greene" <***@qsl.net>
Cc: "Mike McCoy" <***@austin.rr.com>; <***@arrl.net>; "Elecraft"
<***@mailman.qth.net>
Sent: Monday, December 29, 2003 5:12 PM
Subject: Re: [Elecraft] Elecraft technology >> K2+
Post by Phil Wheeler
Funny how tastes differ in this DSP arena. At one time I owned a 599ZX
and an NIR-12 which I A-B tested extensively. Kept the NIR-12 but no
longer use it much: Very satisfied with the KDSP2.
Phil
Mike McCoy
2003-12-30 06:43:00 UTC
Permalink
Post by George, W5YR
Phil and all,
I also have the NIR-12 and find that as a DSP unit, it does a superb job.
Since it has dual 40 MHz DSP chips, it is able to perform all functions
concurrently: filtering, notching and nose reduction. Plus having an audio
amp to feed a speaker.
As Phil said earlier, 'Funny how tastes differ in this DSP arena'. I
considered the NIR-12 as well but this is what convinced me to opt for the
Timewave :

http://www.timewave.com/compare.html

But to each his own. That Charles (W1CG) compares the KDSP2 favorably to the
Timewave is quite a recommendation and really makes me think twice about not
having one in my K2. I'm thinking (rationalizing? ;) I should have the best
of both worlds... the Timewave for shack and the KDSP2 for field ;)

Mike, K5PU
Phil Wheeler
2003-12-30 07:05:00 UTC
Permalink
The Timewave definitely has more bells and whistles -- and I expected to
prefer it. It did not turn out that way in actual use.

Phil
Post by Mike McCoy
Post by George, W5YR
Phil and all,
I also have the NIR-12 and find that as a DSP unit, it does a superb job.
Since it has dual 40 MHz DSP chips, it is able to perform all functions
concurrently: filtering, notching and nose reduction. Plus having an audio
amp to feed a speaker.
As Phil said earlier, 'Funny how tastes differ in this DSP arena'. I
considered the NIR-12 as well but this is what convinced me to opt for the
http://www.timewave.com/compare.html
--- StripMime Report -- processed MIME parts ---
multipart/alternative
text/plain (text body -- kept)
text/html
The reason this message is shown is because the post was in HTML
or had an attachment. Attachments are not allowed. To learn how
to post in Plain-Text go to: http://www.expita.com/nomime.html ---
George, W5YR
2003-12-30 08:45:01 UTC
Permalink
Never had a chance to listen to the Timewave units, Mike, but I don't doubt
that they are good units. The thing that attracted me to the NIR-12 is that
at the time it was the only unit that could concurrently filter, notch and
noise reduce. And do each job very, very well. But, I understand that the
Timewave has some useful functions built in . . .

73, George W5YR
***@att.net


----- Original Message -----
From: "Mike McCoy" <***@austin.rr.com>
To: "George, W5YR" <***@att.net>; "Phil Wheeler" <***@socal.rr.com>;
"Charles Greene" <***@qsl.net>
Cc: <***@arrl.net>; "Elecraft" <***@mailman.qth.net>
Sent: Monday, December 29, 2003 11:38 PM
Subject: Re: [Elecraft] Elecraft technology >> K2+
Post by Mike McCoy
Post by George, W5YR
Phil and all,
I also have the NIR-12 and find that as a DSP unit, it does a superb job.
Since it has dual 40 MHz DSP chips, it is able to perform all functions
concurrently: filtering, notching and nose reduction. Plus having an audio
amp to feed a speaker.
As Phil said earlier, 'Funny how tastes differ in this DSP arena'. I
considered the NIR-12 as well but this is what convinced me to opt for the
http://www.timewave.com/compare.html
But to each his own. That Charles (W1CG) compares the KDSP2 favorably to the
Timewave is quite a recommendation and really makes me think twice about not
having one in my K2. I'm thinking (rationalizing? ;) I should have the best
of both worlds... the Timewave for shack and the KDSP2 for field ;)
Mike, K5PU
_______________________________________________
You must be subscribed to post to the list.
http://mailman.qth.net/mailman/listinfo/elecraft
Post by Mike McCoy
Elecraft Web Page: http://www.elecraft.com
Also see: http://www.elecraft.com/elecraft_list_guidelines.htm
Charles Greene
2003-12-30 15:35:00 UTC
Permalink
Phil and All,

I have a IC-706MKIIG and it has the best NR function of any. It's all that
it does, but it beats the TimeWave units and the KDSP2 hands down. I don't
have any other devices that have the NR function to compare to except the
Omni-VI +, which is hard to tell when it is engaged. I will say this
about the KDSP2. It has a lot of optional settings which I tried, but they
don't seem to make any difference. So it is probable that I don't have my
KDSP2 optimized for NR.

Say, did any of you guys look at the audio spectrum of the DSP NR audio
using Spectrogram? It's pretty weird, and I need someone to explain what
is actually going on.
Post by Phil Wheeler
Funny how tastes differ in this DSP arena. At one time I owned a 599ZX
and an NIR-12 which I A-B tested extensively. Kept the NIR-12 but no
longer use it much: Very satisfied with the KDSP2.
Phil
Post by Charles Greene
Mike and all,
I use my Timewave 599ZX with K2 492 which has no KDSP2 but which has the
KAF2. K2/100 3571 has the KDSP2. I have mixed emotions about which is
better and which is more convenient. Actually, I have found the KDSP2 is
not as good as the Timewave in noise reduction, but once you have the
filters set up for SSB or CW, the KDSP2 is more convenient, in my
opinion, and of course is more portable.
Post by Bill Coleman
Post by Bill Coleman
Realistically, there are only a few things that the K2 needs to
* current filter indication on display
* one-button access to the Rx antenna port. (toggle)
* DSP indication on display and easy-access controls
Although its power requirements preclude any portable operation that
last point is the primary reason I've decided not to replace my Timewave
599ZX with a KDSP2.
I'm not a big fan of 'toggling' thru filters ala AFIL/XFIL ('oops, that
last selection was the best choice'). I guess I prefer the 599ZX being
external and adjusting the 599ZX's 'infinite' variability using front
panel knobs/switches and having the settings displayed on the Timewaves
LCD and/or LED's. Plus it fits perfectly on top of the K2 almost as if
it were made for it (pretty much same width/depth as the K2).
73, Chas, W1CG
K2 462b, 3571
Phil Wheeler
2003-12-30 17:52:00 UTC
Permalink
Post by Charles Greene
Phil and All,
I have a IC-706MKIIG and it has the best NR function of any.
My 706MkII (no G) works well. But I think the NR in my FT-757 is a tad
more effective. I have a bias against Icom because both Yaesu and
Kenwood service are a 30 minute drive for me (and Elecraft service is
right here!). The 706 series was a tight fit for my new car so I went
with the smaller 757 -- to get 440 without having two radios as I did in
the older car (where I used the 706MkII only for HF).
Post by Charles Greene
Say, did any of you guys look at the audio spectrum of the DSP NR
audio using Spectrogram? It's pretty weird, and I need someone to
explain what is actually going on.
I have not. I use Spectrogram only rarely since I cannot leave the
laptop set up all the time in the shack without occupying my "build
space". Likely Lyle is the one to ask this question.

73, Phil
Charles Greene
2003-12-30 20:14:00 UTC
Permalink
Phil,

You don't mean 757? I have Yaesu FT-757GX sitting right hear beside
me. It is a 1970's radio, .5-30 MHz, no DSP. I use the 706MKIIG for HF
through UHF.
Post by Phil Wheeler
Post by Charles Greene
Phil and All,
I have a IC-706MKIIG and it has the best NR function of any.
My 706MkII (no G) works well. But I think the NR in my FT-757 is a tad
more effective. I have a bias against Icom because both Yaesu and Kenwood
service are a 30 minute drive for me (and Elecraft service is right
here!). The 706 series was a tight fit for my new car so I went with the
smaller 757 -- to get 440 without having two radios as I did in the older
car (where I used the 706MkII only for HF).
Post by Charles Greene
Say, did any of you guys look at the audio spectrum of the DSP NR audio
using Spectrogram? It's pretty weird, and I need someone to explain what
is actually going on.
I have not. I use Spectrogram only rarely since I cannot leave the laptop
set up all the time in the shack without occupying my "build
space". Likely Lyle is the one to ask this question.
73, Phil
73, Chas, W1CG
K2 462b, 3571
Phil Wheeler
2003-12-30 20:51:00 UTC
Permalink
Guess it is an FT-857 :-)
Post by Charles Greene
Phil,
You don't mean 757? I have Yaesu FT-757GX sitting right hear beside
me. It is a 1970's radio, .5-30 MHz, no DSP. I use the 706MKIIG for
HF through UHF.
Post by Phil Wheeler
Post by Charles Greene
Phil and All,
I have a IC-706MKIIG and it has the best NR function of any.
My 706MkII (no G) works well. But I think the NR in my FT-757 is a
tad more effective. I have a bias against Icom because both Yaesu
and Kenwood service are a 30 minute drive for me (and Elecraft
service is right here!). The 706 series was a tight fit for my new
car so I went with the smaller 757 -- to get 440 without having two
radios as I did in the older car (where I used the 706MkII only for HF).
Post by Charles Greene
Say, did any of you guys look at the audio spectrum of the DSP NR
audio using Spectrogram? It's pretty weird, and I need someone to
explain what is actually going on.
I have not. I use Spectrogram only rarely since I cannot leave the
laptop set up all the time in the shack without occupying my "build
space". Likely Lyle is the one to ask this question.
73, Phil
73, Chas, W1CG
K2 462b, 3571
--- StripMime Report -- processed MIME parts ---
multipart/alternative
text/plain (text body -- kept)
text/html
The reason this message is shown is because the post was in HTML
or had an attachment. Attachments are not allowed. To learn how
to post in Plain-Text go to: http://www.expita.com/nomime.html ---
Ron D'Eau Claire
2003-12-29 18:08:00 UTC
Permalink
One of the nicest things about the little KX1 (other than it's size) is =
the
front panel "knob" to change filter bandwidth. It's VERY nice to simply =
turn
a pot shaft to go from maximum to minimum bandwidth.=20

Ron AC7AC

-----Original Message-----
From: elecraft-***@mailman.qth.net =
[mailto:elecraft-***@mailman.qth.net]
On Behalf Of Mike McCoy


...I'm not a big fan of 'toggling' thru filters ala AFIL/XFIL ('oops, =
that
last selection was the best choice'). ...

Mike, K5PU
Lyle Johnson
2003-12-29 18:18:03 UTC
Permalink
Hello Floyd!
Post by Floyd Sense" (Floyd Sense)
My question is: does the KDSP2 auto-notch function degrade voice
readability or does it function more like the MFJ?
Any time you process signals, you alter them and add artifacts or other
forms of distortion. The KDSP2 autonotch uses the LMS algorithm, and it
will sound similar to other units employing this type of algorithm. The
distortion is usually much less objectionable than the interference that is
being reduced.

It turns out that people have very different reactions to various forms of
these artifacts, so the only real way to know is to hear one yourself and
judge. It is very much a subjective call, not an objective, measurable one.

73,

Lyle KK7P
Floyd Sense" (Floyd Sense)
2003-12-30 00:10:59 UTC
Permalink
Yes - I understand that, that's why I asked if anyone had compared the ones
I mentioned. I have no idea which approach is used in the 599zx or the
MFJ-784B, but they are clearly different, and the difference is not subtle.
As the MFJ proves, it's not impossible to notch one or more tones without
introducing artifacts that make the desired signal difficult to listen to.
Perhaps someone can offer their opinion on the distortion that the KDSP2
adds in this case. It's not likely that I'll invest the price of the KDSP2
only to find it's worse than the MFJ!

73 and Happy New Year, Floyd - K8AC

----- Original Message -----
From: "Lyle Johnson" <***@fidalgo.net>
To: "Floyd Sense" <***@sprintmail.com>
Cc: "Elecraft" <***@mailman.qth.net>
Sent: Monday, December 29, 2003 12:14 PM
Subject: RE: [Elecraft] Elecraft technology >> K2+
Post by Lyle Johnson
Hello Floyd!
Post by Floyd Sense" (Floyd Sense)
My question is: does the KDSP2 auto-notch function degrade voice
readability or does it function more like the MFJ?
Any time you process signals, you alter them and add artifacts or other
forms of distortion. The KDSP2 autonotch uses the LMS algorithm, and it
will sound similar to other units employing this type of algorithm. The
distortion is usually much less objectionable than the interference that is
being reduced.
It turns out that people have very different reactions to various forms of
these artifacts, so the only real way to know is to hear one yourself and
judge. It is very much a subjective call, not an objective, measurable one.
73,
Lyle KK7P
Jim Brown" (Jim Brown)
2003-12-29 19:29:00 UTC
Permalink
Post by Floyd Sense" (Floyd Sense)
My question is: does the KDSP2 auto-notch function degrade voice readability
or does it function more like the MFJ?
Although I own a 599ZX that is hooked up to my Omni V, I have not used
its autonotch function.

I like the KDSP2 a lot, and find that its autonotch is quite useable.
It does take a short time to lock in, and if you can leave it engaged
as you tune across the band, you will hear it unlocking on one carrier
and locking on another. I do not find that it seriously degrades
speech.

There is, by the way, a far better way to take out a carrier than a
notch, that has virtually NO effect on speech. That is to lock an
oscillator to the carrier, track its amplitude, and precisely match its
phase, then subtract it from the carrier. In other words, it nulls the
carrier rather than filtering it. The distortion of speech that you
hear with a notch is due to the extreme phase shift created by the
filter extending out both ways from the filter skirts. In general, the
deeper and sharper the notch the greater the phase distortion.

This nulling technique is well known in pro audio, but I don't know if
it has found its way into ham DSP. It certainly should -- it has the
potential to provide superior results if it can be pulled off well,
because it does not add phase distortion to the audio.

Jim Brown K9YC
Don Allen
2003-12-29 20:36:00 UTC
Permalink
Post by Floyd Sense" (Floyd Sense)
My question is: does the KDSP2 auto-notch function degrade voice readability
or does it function more like the MFJ?
I've owned some of the better DSP audio filters including the Timewave
599zx, JPS NIR-10, and dual-DSP JPS NIR-12. All have fast acting and
deep auto-notch functions. However, the best auto-notch functions I've
ever used are analog, and available in the Datong (Dr. David Tong in the
U.K.) FL1 and FL3. Both the FL1 and FL3 can be used in the auto-notch
mode or manual notch, and the notch depths (to my ear) are as deep or
deeper than any of the aforementioned DSP units. More importantly,
neither the FL1 or FL3 seem to distort the voice while using the notch
function to the level a DSP-based notch does. And, the auto-notch is
very fast in operation. Both of these filters also include the normal
bandpass op amp-based filters. These units were available from around
1977 through 1984 I believe. Just a thought, as not everything has to
be DSP-based to work well . . .

Don W9CW
George, W5YR
2003-12-30 03:38:00 UTC
Permalink
Full agreement with Don re the FL1 and FL3. The FL3 is especially phenomenal
in its operation in all modes. For an analog filter, it produces almost
unbelievablely good results in filtering and notching.

These are no longer being made and are collector's items. Due to the
diligence of Dr. Megacycle at a hamfest, I am the proud owner of a virtually
brand new FL3 complete with all documentation, including the original bill
of sale. It fits precisely on top of the K2 and is close enough in color to
at least be in the family. I owe Jim a debt of gratitude for spotting this
one for me.

Be on the lookout on Ebay and every hamfest you attend for a Datong audio
filter. I bought my FL1 from a dealer because it was DOA out of the box -
think I paid him $10 for it! They sold back in the 70's for $300 or so! The
problem was a defective chip whose replacement I obtained at Radio Shack for
a buck . .

One last comment: the FL1 is a good audio filter but the FL3 .is an order of
magnitude better in all respects.

73, George W5YR
***@att.net


----- Original Message -----
From: "Don Allen" <***@intergate.com>
To: "Jim Brown" <***@rcn.com>; <***@mailman.qth.net>
Sent: Monday, December 29, 2003 1:31 PM
Subject: Re: [Elecraft] Elecraft technology >> K2+
Post by Don Allen
Post by Floyd Sense" (Floyd Sense)
My question is: does the KDSP2 auto-notch function degrade voice readability
or does it function more like the MFJ?
I've owned some of the better DSP audio filters including the Timewave
599zx, JPS NIR-10, and dual-DSP JPS NIR-12. All have fast acting and
deep auto-notch functions. However, the best auto-notch functions I've
ever used are analog, and available in the Datong (Dr. David Tong in the
U.K.) FL1 and FL3. Both the FL1 and FL3 can be used in the auto-notch
mode or manual notch, and the notch depths (to my ear) are as deep or
deeper than any of the aforementioned DSP units. More importantly,
neither the FL1 or FL3 seem to distort the voice while using the notch
function to the level a DSP-based notch does. And, the auto-notch is
very fast in operation. Both of these filters also include the normal
bandpass op amp-based filters. These units were available from around
1977 through 1984 I believe. Just a thought, as not everything has to
be DSP-based to work well . . .
Don W9CW
Lyle Johnson
2003-12-30 16:18:01 UTC
Permalink
Hello Charles!
Post by Charles Greene
Say, did any of you guys look at the audio spectrum of the DSP NR audio
using Spectrogram? It's pretty weird, and I need someone to explain what
is actually going on.
The KDSP2 uses an adaptive LMS algorithm. This is DSP-speak to say that it
is adjusting an audio filter. When it sees what it thinks is noise, it
tires to suppress it, and because noise is spread out over the whole audio
spectrum, you will see what looks like a "comb" effect of a number of
filters. Maybe "lumpy" is a better description of the resulting passband
when there is only noise.

As signals appear, the filter elements adjust to try and pass the signal
content, so you'll see the response peak up on the signals. Since voice is
a complex waveform, the filter adjusts itself to pass the part of the voice
response that it thinks is a signal, and to continue to try and suppress the
frequencies that it thinks are noise. The resulting audio characteristic is
somewhat "hollow" sounding.

73,

Lyle KK7P
Loading...