Elecraft

CW rise time setting

Classic

List

Threaded

22 messages Options

Richard Stutsman

CW rise time setting

I for one would like to have some control over the rise/fall times.

You want the cleanest (narrowest) of CW signals when operating on a crowded
band or in a contest - unless you're a rare DX station. Most of my
operations are 22wpm rag chews on very uncrowded bands. We're often the
only discernible signals on an entire CW band. And conditions are often
noisy with deep QSB. Why not allow those of us operating in those
circumstances to shorten the rise times a bit, which makes it easier to
copy in noisy conditions, when any close-in clicks or thumps aren't going
to bother anybody? Copying a 'soft' weak CW signal is like trying to read a
'crisp' signal that's an entire S-unit weaker, IMO.

My TS-590sg let's me do that, and I love it!

Will the K4 sound as good as a Drake T4C? (Just askin'.)

Rick N6IET

> N4ZR wrote:
> >
> > Will the K4 have options for setting CW rise and fall times...?
> Hi Pete,
> Probably not. We've always been very careful to ensure our rigs have an
> exceptional clean, click-free CW signal. (Third-party testing bears this
> out.) Minimizing the bandwidth requires a very specific rise/fall time and
> a hand-crafted sigmoidal shaping function in DSP.
> Colonel Sanders closely guards his fried chick recipe, Mrs. Fields won't
> reveal what's in her chocolate-chip cookies, and only selected firmware
> monks--sworn to secrecy--are privy to Elecraft's keying envelope
> coefficients :)
> 73,
> Wayne
> N6KR

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

David Gilbert-2

Re: CW rise time setting

You don't need to generate clicks to have a crisp CW tone. Elecraft uses a
pretty much optimally shaped waveform (some version of a cosine function)
and if I remember correctly the rise time is only about 2.5 msec, although
I could be wrong about that last part. And while you may be careful not to
use short rise/fall times when the band is active, in the past there have
been folks on the contesting reflector who openly admitted they purposely
generate clicks by shortening the rise/fall times to give themselves elbow
room. I will always appreciate that Elecraft doesn't give those miscreants
the means to pollute the band.

73,
Dave AB7E

On Mon, Jun 8, 2020 at 9:23 AM Richard Stutsman <[hidden email]> wrote:

> I for one would like to have some control over the rise/fall times.
>
> You want the cleanest (narrowest) of CW signals when operating on a crowded
> band or in a contest - unless you're a rare DX station. Most of my
> operations are 22wpm rag chews on very uncrowded bands. We're often the
> only discernible signals on an entire CW band. And conditions are often
> noisy with deep QSB. Why not allow those of us operating in those
> circumstances to shorten the rise times a bit, which makes it easier to
> copy in noisy conditions, when any close-in clicks or thumps aren't going
> to bother anybody? Copying a 'soft' weak CW signal is like trying to read a
> 'crisp' signal that's an entire S-unit weaker, IMO.
>
> My TS-590sg let's me do that, and I love it!
>
> Will the K4 sound as good as a Drake T4C? (Just askin'.)
>
> Rick N6IET
>
> > N4ZR wrote:
> > >
> > > Will the K4 have options for setting CW rise and fall times...?
> > Hi Pete,
> > Probably not. We've always been very careful to ensure our rigs have an
> > exceptional clean, click-free CW signal. (Third-party testing bears this
> > out.) Minimizing the bandwidth requires a very specific rise/fall time
> and
> > a hand-crafted sigmoidal shaping function in DSP.
> > Colonel Sanders closely guards his fried chick recipe, Mrs. Fields won't
> > reveal what's in her chocolate-chip cookies, and only selected firmware
> > monks--sworn to secrecy--are privy to Elecraft's keying envelope
> > coefficients :)
> > 73,
> > Wayne
> > N6KR
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

Richard Stutsman

Re: CW rise time setting

Yes, I would regard a 2.5 msec rise time (using appropriate sigmoid
shaping) to probably be ideal, in which case I would have no desire to
modify or shorten it. I'll bet it will sound even better than a Drake or a
Ten-Tec!

I hereby withdraw my previous request/opinion.

Rick N6IET

On Mon, Jun 8, 2020 at 10:27 AM David Gilbert <[hidden email]> wrote:

> You don't need to generate clicks to have a crisp CW tone. Elecraft uses
> a pretty much optimally shaped waveform (some version of a cosine function)
> and if I remember correctly the rise time is only about 2.5 msec, although
> I could be wrong about that last part. And while you may be careful not to
> use short rise/fall times when the band is active, in the past there have
> been folks on the contesting reflector who openly admitted they purposely
> generate clicks by shortening the rise/fall times to give themselves elbow
> room. I will always appreciate that Elecraft doesn't give those miscreants
> the means to pollute the band.
>
> 73,
> Dave AB7E
>
> On Mon, Jun 8, 2020 at 9:23 AM Richard Stutsman <[hidden email]> wrote:
>
>> I for one would like to have some control over the rise/fall times.
>>
>> You want the cleanest (narrowest) of CW signals when operating on a
>> crowded
>> band or in a contest - unless you're a rare DX station. Most of my
>> operations are 22wpm rag chews on very uncrowded bands. We're often the
>> only discernible signals on an entire CW band. And conditions are often
>> noisy with deep QSB. Why not allow those of us operating in those
>> circumstances to shorten the rise times a bit, which makes it easier to
>> copy in noisy conditions, when any close-in clicks or thumps aren't going
>> to bother anybody? Copying a 'soft' weak CW signal is like trying to read
>> a
>> 'crisp' signal that's an entire S-unit weaker, IMO.
>>
>> My TS-590sg let's me do that, and I love it!
>>
>> Will the K4 sound as good as a Drake T4C? (Just askin'.)
>>
>> Rick N6IET
>>
>> > N4ZR wrote:
>> > >
>> > > Will the K4 have options for setting CW rise and fall times...?
>> > Hi Pete,
>> > Probably not. We've always been very careful to ensure our rigs have an
>> > exceptional clean, click-free CW signal. (Third-party testing bears this
>> > out.) Minimizing the bandwidth requires a very specific rise/fall time
>> and
>> > a hand-crafted sigmoidal shaping function in DSP.
>> > Colonel Sanders closely guards his fried chick recipe, Mrs. Fields won't
>> > reveal what's in her chocolate-chip cookies, and only selected firmware
>> > monks--sworn to secrecy--are privy to Elecraft's keying envelope
>> > coefficients :)
>> > 73,
>> > Wayne
>> > N6KR
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
>>
>

Jim Brown-10

Re: CW rise time setting

In reply to this post by Richard Stutsman

On 6/8/2020 9:21 AM, Richard Stutsman wrote:
> I for one would like to have some control over the rise/fall times.
>
> You want the cleanest (narrowest) of CW signals when operating on a crowded
> band or in a contest - unless you're a rare DX station. Most of my
> operations are 22wpm rag chews on very uncrowded bands. We're often the
> only discernible signals on an entire CW band. And conditions are often
> noisy with deep QSB. Why not allow those of us operating in those
> circumstances to shorten the rise times a bit, which makes it easier to
> copy in noisy conditions,

Because that's not how it works. Keying in all Elecraft rigs since the
K3 has been carefully shaped for optimum copy AND minimum bandwidth.
What Elecraft does is FAR superior to those rigs with adjustable rise time.

> My TS-590sg let's me do that, and I love it!

You should ALWAYS run it at the longest rise time setting.

>Will the K4 sound as good as a Drake T4C? (Just askin'.)

Although I worked for Drake doing final test of their first TR3s, I
can't say that I know what a T4C sounds like. Do you?

73, Jim K9YC
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Edward H Russell

Re: CW rise time setting

In reply to this post by Richard Stutsman

I'm used to CW rise times in th 4-7ms range. Is this 2.5ms arrived at by a
different metric? Is the curve added by sigmoidal shaping somehow excluded
from the measurement?

Ed / w2rf

-----Original Message-----
From: [hidden email] <[hidden email]> On
Behalf Of Richard Stutsman
Sent: Monday, June 8, 2020 1:31 PM
To: David Gilbert <[hidden email]>
Cc: Elecraft Reflector <[hidden email]>
Subject: Re: [Elecraft] CW rise time setting

Yes, I would regard a 2.5 msec rise time (using appropriate sigmoid

shaping) to probably be ideal, in which case I would have no desire to
modify or shorten it. I'll bet it will sound even better than a Drake or a
Ten-Tec!

I hereby withdraw my previous request/opinion.

Rick N6IET

On Mon, Jun 8, 2020 at 10:27 AM David Gilbert < <mailto:[hidden email]>
[hidden email]> wrote:

> You don't need to generate clicks to have a crisp CW tone. Elecraft

> uses a pretty much optimally shaped waveform (some version of a cosine

> function) and if I remember correctly the rise time is only about 2.5

> msec, although I could be wrong about that last part. And while you

> may be careful not to use short rise/fall times when the band is

> active, in the past there have been folks on the contesting reflector

> who openly admitted they purposely generate clicks by shortening the

> rise/fall times to give themselves elbow room. I will always

> appreciate that Elecraft doesn't give those miscreants the means to
pollute the band.

>

> 73,

> Dave AB7E

>

> On Mon, Jun 8, 2020 at 9:23 AM Richard Stutsman <
<mailto:[hidden email]> [hidden email]> wrote:

>

>> I for one would like to have some control over the rise/fall times.

>>

>> You want the cleanest (narrowest) of CW signals when operating on a

>> crowded band or in a contest - unless you're a rare DX station. Most

>> of my operations are 22wpm rag chews on very uncrowded bands. We're

>> often the only discernible signals on an entire CW band. And

>> conditions are often noisy with deep QSB. Why not allow those of us

>> operating in those circumstances to shorten the rise times a bit,

>> which makes it easier to copy in noisy conditions, when any close-in

>> clicks or thumps aren't going to bother anybody? Copying a 'soft'

>> weak CW signal is like trying to read a 'crisp' signal that's an

>> entire S-unit weaker, IMO.

>>

>> My TS-590sg let's me do that, and I love it!

>>

>> Will the K4 sound as good as a Drake T4C? (Just askin'.)

>>

>> Rick N6IET

>>

>> > N4ZR wrote:

>> > >

>> > > Will the K4 have options for setting CW rise and fall times...?

>> > Hi Pete,

>> > Probably not. We've always been very careful to ensure our rigs

>> > have an exceptional clean, click-free CW signal. (Third-party

>> > testing bears this

>> > out.) Minimizing the bandwidth requires a very specific rise/fall

>> > time

>> and

>> > a hand-crafted sigmoidal shaping function in DSP.

>> > Colonel Sanders closely guards his fried chick recipe, Mrs. Fields

>> > won't reveal what's in her chocolate-chip cookies, and only

>> > selected firmware monks--sworn to secrecy--are privy to Elecraft's

>> > keying envelope coefficients :) 73, Wayne N6KR

>> ______________________________________________________________

>> Elecraft mailing list

>> Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
http://mailman.qth.net/mailman/listinfo/elecraft

>> Help: <http://mailman.qth.net/mmfaq.htm>
http://mailman.qth.net/mmfaq.htm

>> Post: <mailto:[hidden email]> mailto:[hidden email]

>>

>> This list hosted by: <http://www.qsl.net> http://www.qsl.net Please help
support this

>> email list: <http://www.qsl.net/donate.html>
http://www.qsl.net/donate.html Message delivered to

>> <mailto:[hidden email]> [hidden email]

>>

>

______________________________________________________________

Elecraft mailing list

Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
http://mailman.qth.net/mailman/listinfo/elecraft

Help: <http://mailman.qth.net/mmfaq.htm> http://mailman.qth.net/mmfaq.htm

Post: <mailto:[hidden email]> mailto:[hidden email]

This list hosted by: <http://www.qsl.net> http://www.qsl.net

Please help support this email list: <http://www.qsl.net/donate.html>
http://www.qsl.net/donate.html Message delivered to <mailto:[hidden email]>
[hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Alan Bloom

Re: CW rise time setting

In the "good old days" key shaping was done simply by adding a capacitor
to the key line or equivalent. That results in an exponential rise
and/or fall time, which is not optimum, so the time constant had to be
set pretty slow to avoid key clicks. Typically 5-10 ms. 10 ms results
in "mushy" keying, especially at high CW speeds.

Raised-cosine key shaping is close to optimum and is easy to implement
with a DSP. It allows faster rise/fall times without key clicks. I
assume Elecraft is using something like that.

Alan N1AL

On 2020-06-08 12:51, E.H. Russell wrote:

> I'm used to CW rise times in th 4-7ms range. Is this 2.5ms arrived at by a
> different metric? Is the curve added by sigmoidal shaping somehow excluded
> from the measurement?
>
> Ed / w2rf
>
> -----Original Message-----
> From: [hidden email] <[hidden email]> On
> Behalf Of Richard Stutsman
> Sent: Monday, June 8, 2020 1:31 PM
> To: David Gilbert <[hidden email]>
> Cc: Elecraft Reflector <[hidden email]>
> Subject: Re: [Elecraft] CW rise time setting
>
> Yes, I would regard a 2.5 msec rise time (using appropriate sigmoid
>
> shaping) to probably be ideal, in which case I would have no desire to
> modify or shorten it. I'll bet it will sound even better than a Drake or a
> Ten-Tec!
>
> I hereby withdraw my previous request/opinion.
>
> Rick N6IET
>
> On Mon, Jun 8, 2020 at 10:27 AM David Gilbert < <mailto:[hidden email]>
> [hidden email]> wrote:
>
>> You don't need to generate clicks to have a crisp CW tone. Elecraft
>
>> uses a pretty much optimally shaped waveform (some version of a cosine
>
>> function) and if I remember correctly the rise time is only about 2.5
>
>> msec, although I could be wrong about that last part. And while you
>
>> may be careful not to use short rise/fall times when the band is
>
>> active, in the past there have been folks on the contesting reflector
>
>> who openly admitted they purposely generate clicks by shortening the
>
>> rise/fall times to give themselves elbow room. I will always
>
>> appreciate that Elecraft doesn't give those miscreants the means to
> pollute the band.
>
>> 73,
>
>> Dave AB7E
>
>> On Mon, Jun 8, 2020 at 9:23 AM Richard Stutsman <
> <mailto:[hidden email]> [hidden email]> wrote:
>
> I for one would like to have some control over the rise/fall times.

>

>> You want the cleanest (narrowest) of CW signals when operating on a

>> crowded band or in a contest - unless you're a rare DX station. Most

>> of my operations are 22wpm rag chews on very uncrowded bands. We're

>> often the only discernible signals on an entire CW band. And

>> conditions are often noisy with deep QSB. Why not allow those of us

>> operating in those circumstances to shorten the rise times a bit,

>> which makes it easier to copy in noisy conditions, when any close-in

>> clicks or thumps aren't going to bother anybody? Copying a 'soft'

>> weak CW signal is like trying to read a 'crisp' signal that's an

>> entire S-unit weaker, IMO.

>

>> My TS-590sg let's me do that, and I love it!

>

>> Will the K4 sound as good as a Drake T4C? (Just askin'.)

>

>> Rick N6IET

>

> N4ZR wrote:

>

> Will the K4 have options for setting CW rise and fall times...?

> Hi Pete,

> Probably not. We've always been very careful to ensure our rigs

> have an exceptional clean, click-free CW signal. (Third-party

> testing bears this

> out.) Minimizing the bandwidth requires a very specific rise/fall

> time

>> and

> a hand-crafted sigmoidal shaping function in DSP.

> Colonel Sanders closely guards his fried chick recipe, Mrs. Fields

> won't reveal what's in her chocolate-chip cookies, and only

> selected firmware monks--sworn to secrecy--are privy to Elecraft's

> keying envelope coefficients :) 73, Wayne N6KR

>> ______________________________________________________________

>> Elecraft mailing list

>> Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
http://mailman.qth.net/mailman/listinfo/elecraft

>> Help: <http://mailman.qth.net/mmfaq.htm>
http://mailman.qth.net/mmfaq.htm

>> Post: <mailto:[hidden email]> mailto:[hidden email]

>

>> This list hosted by: <http://www.qsl.net> http://www.qsl.net Please help
support this

>> email list: <http://www.qsl.net/donate.html>
http://www.qsl.net/donate.html Message delivered to

>> <mailto:[hidden email]> [hidden email]

>

______________________________________________________________

Elecraft mailing list

Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
http://mailman.qth.net/mailman/listinfo/elecraft

Help: <http://mailman.qth.net/mmfaq.htm>
http://mailman.qth.net/mmfaq.htm

Post: <mailto:[hidden email]> mailto:[hidden email]

This list hosted by: <http://www.qsl.net> http://www.qsl.net

Please help support this email list: <http://www.qsl.net/donate.html>
http://www.qsl.net/donate.html Message delivered to
<mailto:[hidden email]>
[hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Edward H Russell

Re: CW rise time setting

I suppose shaping of the curve corners removes harmonics introduced by the abrupt transitions, allowing an accelerated ramp inbetween. But does this really reduce the total rise time to 2.5ms? It seems that the softening process must take some time. Wish I had a K3 here to scope against other radios. Anything published out there?

Ed / w2rf

From: Alan Bloom <[hidden email]>
Sent: Monday, June 8, 2020 3:15 PM
To: E.H. Russell <[hidden email]>
Cc: 'Elecraft Reflector' <[hidden email]>
Subject: Re: [Elecraft] CW rise time setting

In the "good old days" key shaping was done simply by adding a capacitor to the key line or equivalent. That results in an exponential rise and/or fall time, which is not optimum, so the time constant had to be set pretty slow to avoid key clicks. Typically 5-10 ms. 10 ms results in "mushy" keying, especially at high CW speeds.

Raised-cosine key shaping is close to optimum and is easy to implement with a DSP. It allows faster rise/fall times without key clicks. I assume Elecraft is using something like that.

Alan N1AL

On 2020-06-08 12:51, E.H. Russell wrote:

I'm used to CW rise times in th 4-7ms range. Is this 2.5ms arrived at by a
different metric? Is the curve added by sigmoidal shaping somehow excluded
from the measurement?

Ed / w2rf

-----Original Message-----
From: [hidden email] <mailto:[hidden email]> <[hidden email] <mailto:[hidden email]> > On
Behalf Of Richard Stutsman
Sent: Monday, June 8, 2020 1:31 PM
To: David Gilbert <[hidden email] <mailto:[hidden email]> >
Cc: Elecraft Reflector <[hidden email] <mailto:[hidden email]> >
Subject: Re: [Elecraft] CW rise time setting

Yes, I would regard a 2.5 msec rise time (using appropriate sigmoid

shaping) to probably be ideal, in which case I would have no desire to
modify or shorten it. I'll bet it will sound even better than a Drake or a
Ten-Tec!

I hereby withdraw my previous request/opinion.

Rick N6IET

On Mon, Jun 8, 2020 at 10:27 AM David Gilbert < <mailto:[hidden email] <mailto:[hidden email]> >
[hidden email] <mailto:[hidden email]> > wrote:

You don't need to generate clicks to have a crisp CW tone. Elecraft

uses a pretty much optimally shaped waveform (some version of a cosine

function) and if I remember correctly the rise time is only about 2.5

msec, although I could be wrong about that last part. And while you

may be careful not to use short rise/fall times when the band is

active, in the past there have been folks on the contesting reflector

who openly admitted they purposely generate clicks by shortening the

rise/fall times to give themselves elbow room. I will always

appreciate that Elecraft doesn't give those miscreants the means to

pollute the band.

73,

Dave AB7E

On Mon, Jun 8, 2020 at 9:23 AM Richard Stutsman <

<mailto:[hidden email] <mailto:[hidden email]> > [hidden email] <mailto:[hidden email]> > wrote:

I for one would like to have some control over the rise/fall times.

You want the cleanest (narrowest) of CW signals when operating on a

crowded band or in a contest - unless you're a rare DX station. Most

of my operations are 22wpm rag chews on very uncrowded bands. We're

often the only discernible signals on an entire CW band. And

conditions are often noisy with deep QSB. Why not allow those of us

operating in those circumstances to shorten the rise times a bit,

which makes it easier to copy in noisy conditions, when any close-in

clicks or thumps aren't going to bother anybody? Copying a 'soft'

weak CW signal is like trying to read a 'crisp' signal that's an

entire S-unit weaker, IMO.

My TS-590sg let's me do that, and I love it!

Will the K4 sound as good as a Drake T4C? (Just askin'.)

Rick N6IET

N4ZR wrote:

Will the K4 have options for setting CW rise and fall times...?

Hi Pete,

Probably not. We've always been very careful to ensure our rigs

have an exceptional clean, click-free CW signal. (Third-party

testing bears this

out.) Minimizing the bandwidth requires a very specific rise/fall

time

and

a hand-crafted sigmoidal shaping function in DSP.

Colonel Sanders closely guards his fried chick recipe, Mrs. Fields

won't reveal what's in her chocolate-chip cookies, and only

selected firmware monks--sworn to secrecy--are privy to Elecraft's

keying envelope coefficients :) 73, Wayne N6KR

______________________________________________________________

Elecraft mailing list

Home: <http://mailman.qth.net/mailman/listinfo/elecraft>

http://mailman.qth.net/mailman/listinfo/elecraft

Help: <http://mailman.qth.net/mmfaq.htm>

http://mailman.qth.net/mmfaq.htm

Post: <mailto:[hidden email] <mailto:[hidden email]> > mailto:[hidden email] <mailto:[hidden email]>

This list hosted by: <http://www.qsl.net> http://www.qsl.net Please help

support this

email list: <http://www.qsl.net/donate.html>

http://www.qsl.net/donate.html Message delivered to

<mailto:[hidden email] <mailto:[hidden email]> > [hidden email] <mailto:[hidden email]>

______________________________________________________________

Elecraft mailing list

Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
http://mailman.qth.net/mailman/listinfo/elecraft

Help: <http://mailman.qth.net/mmfaq.htm> http://mailman.qth.net/mmfaq.htm

Post: <mailto:[hidden email] <mailto:[hidden email]> > mailto:[hidden email] <mailto:[hidden email]>

This list hosted by: <http://www.qsl.net> http://www.qsl.net

Please help support this email list: <http://www.qsl.net/donate.html>
http://www.qsl.net/donate.html Message delivered to <mailto:[hidden email] <mailto:[hidden email]> >
[hidden email] <mailto:[hidden email]>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email] <mailto:[hidden email]>

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email] <mailto:[hidden email]>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

wayne burdick

Re: CW rise time setting

Administrator

It’s closer to 4 ms.

Wayne
N6KR

----
elecraft.com

> On Jun 8, 2020, at 12:32 PM, E.H. Russell <[hidden email]> wrote:
>
> I suppose shaping of the curve corners removes harmonics introduced by the abrupt transitions, allowing an accelerated ramp inbetween. But does this really reduce the total rise time to 2.5ms? It seems that the softening process must take some time. Wish I had a K3 here to scope against other radios. Anything published out there?
>
>
>
> Ed / w2rf
>
>
>
>
>
>
>
> From: Alan Bloom <[hidden email]>
> Sent: Monday, June 8, 2020 3:15 PM
> To: E.H. Russell <[hidden email]>
> Cc: 'Elecraft Reflector' <[hidden email]>
> Subject: Re: [Elecraft] CW rise time setting
>
>
>
> In the "good old days" key shaping was done simply by adding a capacitor to the key line or equivalent. That results in an exponential rise and/or fall time, which is not optimum, so the time constant had to be set pretty slow to avoid key clicks. Typically 5-10 ms. 10 ms results in "mushy" keying, especially at high CW speeds.
>
>
>
> Raised-cosine key shaping is close to optimum and is easy to implement with a DSP. It allows faster rise/fall times without key clicks. I assume Elecraft is using something like that.
>
>
>
> Alan N1AL
>
>
>
>
>
>
>
>
>
>
>
> On 2020-06-08 12:51, E.H. Russell wrote:
>
> I'm used to CW rise times in th 4-7ms range. Is this 2.5ms arrived at by a
> different metric? Is the curve added by sigmoidal shaping somehow excluded
> from the measurement?
>
>
>
> Ed / w2rf
>
>
>
>
>
>
>
> -----Original Message-----
> From: [hidden email] <mailto:[hidden email]> <[hidden email] <mailto:[hidden email]> > On
> Behalf Of Richard Stutsman
> Sent: Monday, June 8, 2020 1:31 PM
> To: David Gilbert <[hidden email] <mailto:[hidden email]> >
> Cc: Elecraft Reflector <[hidden email] <mailto:[hidden email]> >
> Subject: Re: [Elecraft] CW rise time setting
>
>
>
> Yes, I would regard a 2.5 msec rise time (using appropriate sigmoid
>
> shaping) to probably be ideal, in which case I would have no desire to
> modify or shorten it. I'll bet it will sound even better than a Drake or a
> Ten-Tec!
>
>
>
> I hereby withdraw my previous request/opinion.
>
>
>
> Rick N6IET
>
>
>
> On Mon, Jun 8, 2020 at 10:27 AM David Gilbert < <mailto:[hidden email] <mailto:[hidden email]> >
> [hidden email] <mailto:[hidden email]> > wrote:
>
>
>
>
>
>
> You don't need to generate clicks to have a crisp CW tone. Elecraft
>
>
>
>
>
> uses a pretty much optimally shaped waveform (some version of a cosine
>
>
>
>
>
> function) and if I remember correctly the rise time is only about 2.5
>
>
>
>
>
> msec, although I could be wrong about that last part. And while you
>
>
>
>
>
> may be careful not to use short rise/fall times when the band is
>
>
>
>
>
> active, in the past there have been folks on the contesting reflector
>
>
>
>
>
> who openly admitted they purposely generate clicks by shortening the
>
>
>
>
>
> rise/fall times to give themselves elbow room. I will always
>
>
>
>
>
> appreciate that Elecraft doesn't give those miscreants the means to
>
> pollute the band.
>
>
>
>
>
> 73,
>
>
>
>
>
> Dave AB7E
>
>
>
>
>
>
> On Mon, Jun 8, 2020 at 9:23 AM Richard Stutsman <
>
> <mailto:[hidden email] <mailto:[hidden email]> > [hidden email] <mailto:[hidden email]> > wrote:
>
>
>
>
>
> I for one would like to have some control over the rise/fall times.
>
>
>
>
>
>
>
>
>
> You want the cleanest (narrowest) of CW signals when operating on a
>
>
>
>
>
> crowded band or in a contest - unless you're a rare DX station. Most
>
>
>
>
>
> of my operations are 22wpm rag chews on very uncrowded bands. We're
>
>
>
>
>
> often the only discernible signals on an entire CW band. And
>
>
>
>
>
> conditions are often noisy with deep QSB. Why not allow those of us
>
>
>
>
>
> operating in those circumstances to shorten the rise times a bit,
>
>
>
>
>
> which makes it easier to copy in noisy conditions, when any close-in
>
>
>
>
>
> clicks or thumps aren't going to bother anybody? Copying a 'soft'
>
>
>
>
>
> weak CW signal is like trying to read a 'crisp' signal that's an
>
>
>
>
>
> entire S-unit weaker, IMO.
>
>
>
>
>
>
>
>
>
> My TS-590sg let's me do that, and I love it!
>
>
>
>
>
>
>
>
>
> Will the K4 sound as good as a Drake T4C? (Just askin'.)
>
>
>
>
>
>
>
>
>
> Rick N6IET
>
>
>
>
>
>
>
>
>
> N4ZR wrote:
>
>
>
>
>
>
>
>
>
> Will the K4 have options for setting CW rise and fall times...?
>
>
>
>
>
> Hi Pete,
>
>
>
>
>
> Probably not. We've always been very careful to ensure our rigs
>
>
>
>
>
> have an exceptional clean, click-free CW signal. (Third-party
>
>
>
>
>
> testing bears this
>
>
>
>
>
> out.) Minimizing the bandwidth requires a very specific rise/fall
>
>
>
>
>
> time
>
>
>
>
>
> and
>
>
>
>
>
> a hand-crafted sigmoidal shaping function in DSP.
>
>
>
>
>
> Colonel Sanders closely guards his fried chick recipe, Mrs. Fields
>
>
>
>
>
> won't reveal what's in her chocolate-chip cookies, and only
>
>
>
>
>
> selected firmware monks--sworn to secrecy--are privy to Elecraft's
>
>
>
>
>
> keying envelope coefficients :) 73, Wayne N6KR
>
>
>
>
>
> ______________________________________________________________
>
>
>
>
>
> Elecraft mailing list
>
>
>
>
>
> Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
>
> http://mailman.qth.net/mailman/listinfo/elecraft
>
>
>
>
> Help: <http://mailman.qth.net/mmfaq.htm>
>
> http://mailman.qth.net/mmfaq.htm
>
>
>
>
> Post: <mailto:[hidden email] <mailto:[hidden email]> > mailto:[hidden email] <mailto:[hidden email]>
>
>
>
>
>
>
>
>
>
> This list hosted by: <http://www.qsl.net> http://www.qsl.net Please help
>
> support this
>
>
>
>
> email list: <http://www.qsl.net/donate.html>
>
> http://www.qsl.net/donate.html Message delivered to
>
>
>
>
> <mailto:[hidden email] <mailto:[hidden email]> > [hidden email] <mailto:[hidden email]>
>
>
>
>
>
>
>
> ______________________________________________________________
>
> Elecraft mailing list
>
> Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
> http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: <http://mailman.qth.net/mmfaq.htm> http://mailman.qth.net/mmfaq.htm
>
> Post: <mailto:[hidden email] <mailto:[hidden email]> > mailto:[hidden email] <mailto:[hidden email]>
>
>
>
> This list hosted by: <http://www.qsl.net> http://www.qsl.net
>
> Please help support this email list: <http://www.qsl.net/donate.html>
> http://www.qsl.net/donate.html Message delivered to <mailto:[hidden email] <mailto:[hidden email]> >
> [hidden email] <mailto:[hidden email]>
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email] <mailto:[hidden email]>
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email] <mailto:[hidden email]>
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]

Drew AF2Z

Re: CW rise time setting

A little off topic but is the sidetone also shaped or is it just a plain
old sine wave?

Actually, I wouldn't mind having the option to select a sawtooth or
squarewave for the sidetone. I'm not sure why but it seems to make
sending on a manual key better, prompting the reflexes to be a little
crisper maybe...

73,
Drew
AF2Z

On 06/08/20 14:38, Wayne Burdick wrote:

> It’s closer to 4 ms.
>
> Wayne
> N6KR
>
> ----
> elecraft.com
>
>> On Jun 8, 2020, at 12:32 PM, E.H. Russell <[hidden email]> wrote:
>>
>> I suppose shaping of the curve corners removes harmonics introduced by the abrupt transitions, allowing an accelerated ramp inbetween. But does this really reduce the total rise time to 2.5ms? It seems that the softening process must take some time. Wish I had a K3 here to scope against other radios. Anything published out there?
>>
>>
>>
>> Ed / w2rf
>>
>>
>>
>>
>>
>>
>>
>> From: Alan Bloom <[hidden email]>
>> Sent: Monday, June 8, 2020 3:15 PM
>> To: E.H. Russell <[hidden email]>
>> Cc: 'Elecraft Reflector' <[hidden email]>
>> Subject: Re: [Elecraft] CW rise time setting
>>
>>
>>
>> In the "good old days" key shaping was done simply by adding a capacitor to the key line or equivalent. That results in an exponential rise and/or fall time, which is not optimum, so the time constant had to be set pretty slow to avoid key clicks. Typically 5-10 ms. 10 ms results in "mushy" keying, especially at high CW speeds.
>>
>>
>>
>> Raised-cosine key shaping is close to optimum and is easy to implement with a DSP. It allows faster rise/fall times without key clicks. I assume Elecraft is using something like that.
>>
>>
>>
>> Alan N1AL
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> On 2020-06-08 12:51, E.H. Russell wrote:
>>
>> I'm used to CW rise times in th 4-7ms range. Is this 2.5ms arrived at by a
>> different metric? Is the curve added by sigmoidal shaping somehow excluded
>> from the measurement?
>>
>>
>>
>> Ed / w2rf
>>
>>
>>
>>
>>
>>
>>
>> -----Original Message-----
>> From: [hidden email] <mailto:[hidden email]> <[hidden email] <mailto:[hidden email]> > On
>> Behalf Of Richard Stutsman
>> Sent: Monday, June 8, 2020 1:31 PM
>> To: David Gilbert <[hidden email] <mailto:[hidden email]> >
>> Cc: Elecraft Reflector <[hidden email] <mailto:[hidden email]> >
>> Subject: Re: [Elecraft] CW rise time setting
>>
>>
>>
>> Yes, I would regard a 2.5 msec rise time (using appropriate sigmoid
>>
>> shaping) to probably be ideal, in which case I would have no desire to
>> modify or shorten it. I'll bet it will sound even better than a Drake or a
>> Ten-Tec!
>>
>>
>>
>> I hereby withdraw my previous request/opinion.
>>
>>
>>
>> Rick N6IET
>>
>>
>>
>> On Mon, Jun 8, 2020 at 10:27 AM David Gilbert < <mailto:[hidden email] <mailto:[hidden email]> >
>> [hidden email] <mailto:[hidden email]> > wrote:
>>
>>
>>
>>
>>
>>
>> You don't need to generate clicks to have a crisp CW tone. Elecraft
>>
>>
>>
>>
>>
>> uses a pretty much optimally shaped waveform (some version of a cosine
>>
>>
>>
>>
>>
>> function) and if I remember correctly the rise time is only about 2.5
>>
>>
>>
>>
>>
>> msec, although I could be wrong about that last part. And while you
>>
>>
>>
>>
>>
>> may be careful not to use short rise/fall times when the band is
>>
>>
>>
>>
>>
>> active, in the past there have been folks on the contesting reflector
>>
>>
>>
>>
>>
>> who openly admitted they purposely generate clicks by shortening the
>>
>>
>>
>>
>>
>> rise/fall times to give themselves elbow room. I will always
>>
>>
>>
>>
>>
>> appreciate that Elecraft doesn't give those miscreants the means to
>>
>> pollute the band.
>>
>>
>>
>>
>>
>> 73,
>>
>>
>>
>>
>>
>> Dave AB7E
>>
>>
>>
>>
>>
>>
>> On Mon, Jun 8, 2020 at 9:23 AM Richard Stutsman <
>>
>> <mailto:[hidden email] <mailto:[hidden email]> > [hidden email] <mailto:[hidden email]> > wrote:
>>
>>
>>
>>
>>
>> I for one would like to have some control over the rise/fall times.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> You want the cleanest (narrowest) of CW signals when operating on a
>>
>>
>>
>>
>>
>> crowded band or in a contest - unless you're a rare DX station. Most
>>
>>
>>
>>
>>
>> of my operations are 22wpm rag chews on very uncrowded bands. We're
>>
>>
>>
>>
>>
>> often the only discernible signals on an entire CW band. And
>>
>>
>>
>>
>>
>> conditions are often noisy with deep QSB. Why not allow those of us
>>
>>
>>
>>
>>
>> operating in those circumstances to shorten the rise times a bit,
>>
>>
>>
>>
>>
>> which makes it easier to copy in noisy conditions, when any close-in
>>
>>
>>
>>
>>
>> clicks or thumps aren't going to bother anybody? Copying a 'soft'
>>
>>
>>
>>
>>
>> weak CW signal is like trying to read a 'crisp' signal that's an
>>
>>
>>
>>
>>
>> entire S-unit weaker, IMO.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> My TS-590sg let's me do that, and I love it!
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Will the K4 sound as good as a Drake T4C? (Just askin'.)
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Rick N6IET
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> N4ZR wrote:
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> Will the K4 have options for setting CW rise and fall times...?
>>
>>
>>
>>
>>
>> Hi Pete,
>>
>>
>>
>>
>>
>> Probably not. We've always been very careful to ensure our rigs
>>
>>
>>
>>
>>
>> have an exceptional clean, click-free CW signal. (Third-party
>>
>>
>>
>>
>>
>> testing bears this
>>
>>
>>
>>
>>
>> out.) Minimizing the bandwidth requires a very specific rise/fall
>>
>>
>>
>>
>>
>> time
>>
>>
>>
>>
>>
>> and
>>
>>
>>
>>
>>
>> a hand-crafted sigmoidal shaping function in DSP.
>>
>>
>>
>>
>>
>> Colonel Sanders closely guards his fried chick recipe, Mrs. Fields
>>
>>
>>
>>
>>
>> won't reveal what's in her chocolate-chip cookies, and only
>>
>>
>>
>>
>>
>> selected firmware monks--sworn to secrecy--are privy to Elecraft's
>>
>>
>>
>>
>>
>> keying envelope coefficients :) 73, Wayne N6KR
>>
>>
>>
>>
>>
>> ______________________________________________________________
>>
>>
>>
>>
>>
>> Elecraft mailing list
>>
>>
>>
>>
>>
>> Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
>>
>> http://mailman.qth.net/mailman/listinfo/elecraft
>>
>>
>>
>>
>> Help: <http://mailman.qth.net/mmfaq.htm>
>>
>> http://mailman.qth.net/mmfaq.htm
>>
>>
>>
>>
>> Post: <mailto:[hidden email] <mailto:[hidden email]> > mailto:[hidden email] <mailto:[hidden email]>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> This list hosted by: <http://www.qsl.net> http://www.qsl.net Please help
>>
>> support this
>>
>>
>>
>>
>> email list: <http://www.qsl.net/donate.html>
>>
>> http://www.qsl.net/donate.html Message delivered to
>>
>>
>>
>>
>> <mailto:[hidden email] <mailto:[hidden email]> > [hidden email] <mailto:[hidden email]>
>>
>>
>>
>>
>>
>>
>>
>> ______________________________________________________________
>>
>> Elecraft mailing list
>>
>> Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
>> http://mailman.qth.net/mailman/listinfo/elecraft
>>
>> Help: <http://mailman.qth.net/mmfaq.htm> http://mailman.qth.net/mmfaq.htm
>>
>> Post: <mailto:[hidden email] <mailto:[hidden email]> > mailto:[hidden email] <mailto:[hidden email]>
>>
>>
>>
>> This list hosted by: <http://www.qsl.net> http://www.qsl.net
>>
>> Please help support this email list: <http://www.qsl.net/donate.html>
>> http://www.qsl.net/donate.html Message delivered to <mailto:[hidden email] <mailto:[hidden email]> >
>> [hidden email] <mailto:[hidden email]>
>>
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email] <mailto:[hidden email]>
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email] <mailto:[hidden email]>
>>
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

wayne burdick

Re: CW rise time setting

Administrator

Drew,

We apply exactly the same shape to the sidetone, mute on/off, and mark/space transitions in FSK and PSK modes.

Wayne
N6KR

> On Jun 8, 2020, at 6:54 PM, Drew AF2Z <[hidden email]> wrote:
>
> A little off topic but is the sidetone also shaped or is it just a plain old sine wave?
>
> Actually, I wouldn't mind having the option to select a sawtooth or squarewave for the sidetone. I'm not sure why but it seems to make sending on a manual key better, prompting the reflexes to be a little crisper maybe...
>
> 73,
> Drew
> AF2Z
>
>
>
> On 06/08/20 14:38, Wayne Burdick wrote:
>> It’s closer to 4 ms.
>> Wayne
>> N6KR
>> ----
>> elecraft.com
>>> On Jun 8, 2020, at 12:32 PM, E.H. Russell <[hidden email]> wrote:
>>>
>>> I suppose shaping of the curve corners removes harmonics introduced by the abrupt transitions, allowing an accelerated ramp inbetween. But does this really reduce the total rise time to 2.5ms? It seems that the softening process must take some time. Wish I had a K3 here to scope against other radios. Anything published out there?
>>>
>>>
>>>
>>> Ed / w2rf
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> From: Alan Bloom <[hidden email]>
>>> Sent: Monday, June 8, 2020 3:15 PM
>>> To: E.H. Russell <[hidden email]>
>>> Cc: 'Elecraft Reflector' <[hidden email]>
>>> Subject: Re: [Elecraft] CW rise time setting
>>>
>>>
>>>
>>> In the "good old days" key shaping was done simply by adding a capacitor to the key line or equivalent. That results in an exponential rise and/or fall time, which is not optimum, so the time constant had to be set pretty slow to avoid key clicks. Typically 5-10 ms. 10 ms results in "mushy" keying, especially at high CW speeds.
>>>
>>>
>>>
>>> Raised-cosine key shaping is close to optimum and is easy to implement with a DSP. It allows faster rise/fall times without key clicks. I assume Elecraft is using something like that.
>>>
>>>
>>>
>>> Alan N1AL
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> On 2020-06-08 12:51, E.H. Russell wrote:
>>>
>>> I'm used to CW rise times in th 4-7ms range. Is this 2.5ms arrived at by a
>>> different metric? Is the curve added by sigmoidal shaping somehow excluded
>>> from the measurement?
>>>
>>>
>>>
>>> Ed / w2rf
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: [hidden email] <mailto:[hidden email]> <[hidden email] <mailto:[hidden email]> > On
>>> Behalf Of Richard Stutsman
>>> Sent: Monday, June 8, 2020 1:31 PM
>>> To: David Gilbert <[hidden email] <mailto:[hidden email]> >
>>> Cc: Elecraft Reflector <[hidden email] <mailto:[hidden email]> >
>>> Subject: Re: [Elecraft] CW rise time setting
>>>
>>>
>>>
>>> Yes, I would regard a 2.5 msec rise time (using appropriate sigmoid
>>>
>>> shaping) to probably be ideal, in which case I would have no desire to
>>> modify or shorten it. I'll bet it will sound even better than a Drake or a
>>> Ten-Tec!
>>>
>>>
>>>
>>> I hereby withdraw my previous request/opinion.
>>>
>>>
>>>
>>> Rick N6IET
>>>
>>>
>>>
>>> On Mon, Jun 8, 2020 at 10:27 AM David Gilbert < <mailto:[hidden email] <mailto:[hidden email]> >
>>> [hidden email] <mailto:[hidden email]> > wrote:
>>>
>>>
>>>
>>>
>>>
>>>
>>> You don't need to generate clicks to have a crisp CW tone. Elecraft
>>>
>>>
>>>
>>>
>>>
>>> uses a pretty much optimally shaped waveform (some version of a cosine
>>>
>>>
>>>
>>>
>>>
>>> function) and if I remember correctly the rise time is only about 2.5
>>>
>>>
>>>
>>>
>>>
>>> msec, although I could be wrong about that last part. And while you
>>>
>>>
>>>
>>>
>>>
>>> may be careful not to use short rise/fall times when the band is
>>>
>>>
>>>
>>>
>>>
>>> active, in the past there have been folks on the contesting reflector
>>>
>>>
>>>
>>>
>>>
>>> who openly admitted they purposely generate clicks by shortening the
>>>
>>>
>>>
>>>
>>>
>>> rise/fall times to give themselves elbow room. I will always
>>>
>>>
>>>
>>>
>>>
>>> appreciate that Elecraft doesn't give those miscreants the means to
>>>
>>> pollute the band.
>>>
>>>
>>>
>>>
>>>
>>> 73,
>>>
>>>
>>>
>>>
>>>
>>> Dave AB7E
>>>
>>>
>>>
>>>
>>>
>>>
>>> On Mon, Jun 8, 2020 at 9:23 AM Richard Stutsman <
>>>
>>> <mailto:[hidden email] <mailto:[hidden email]> > [hidden email] <mailto:[hidden email]> > wrote:
>>>
>>>
>>>
>>>
>>>
>>> I for one would like to have some control over the rise/fall times.
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> You want the cleanest (narrowest) of CW signals when operating on a
>>>
>>>
>>>
>>>
>>>
>>> crowded band or in a contest - unless you're a rare DX station. Most
>>>
>>>
>>>
>>>
>>>
>>> of my operations are 22wpm rag chews on very uncrowded bands. We're
>>>
>>>
>>>
>>>
>>>
>>> often the only discernible signals on an entire CW band. And
>>>
>>>
>>>
>>>
>>>
>>> conditions are often noisy with deep QSB. Why not allow those of us
>>>
>>>
>>>
>>>
>>>
>>> operating in those circumstances to shorten the rise times a bit,
>>>
>>>
>>>
>>>
>>>
>>> which makes it easier to copy in noisy conditions, when any close-in
>>>
>>>
>>>
>>>
>>>
>>> clicks or thumps aren't going to bother anybody? Copying a 'soft'
>>>
>>>
>>>
>>>
>>>
>>> weak CW signal is like trying to read a 'crisp' signal that's an
>>>
>>>
>>>
>>>
>>>
>>> entire S-unit weaker, IMO.
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> My TS-590sg let's me do that, and I love it!
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> Will the K4 sound as good as a Drake T4C? (Just askin'.)
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> Rick N6IET
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> N4ZR wrote:
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> Will the K4 have options for setting CW rise and fall times...?
>>>
>>>
>>>
>>>
>>>
>>> Hi Pete,
>>>
>>>
>>>
>>>
>>>
>>> Probably not. We've always been very careful to ensure our rigs
>>>
>>>
>>>
>>>
>>>
>>> have an exceptional clean, click-free CW signal. (Third-party
>>>
>>>
>>>
>>>
>>>
>>> testing bears this
>>>
>>>
>>>
>>>
>>>
>>> out.) Minimizing the bandwidth requires a very specific rise/fall
>>>
>>>
>>>
>>>
>>>
>>> time
>>>
>>>
>>>
>>>
>>>
>>> and
>>>
>>>
>>>
>>>
>>>
>>> a hand-crafted sigmoidal shaping function in DSP.
>>>
>>>
>>>
>>>
>>>
>>> Colonel Sanders closely guards his fried chick recipe, Mrs. Fields
>>>
>>>
>>>
>>>
>>>
>>> won't reveal what's in her chocolate-chip cookies, and only
>>>
>>>
>>>
>>>
>>>
>>> selected firmware monks--sworn to secrecy--are privy to Elecraft's
>>>
>>>
>>>
>>>
>>>
>>> keying envelope coefficients :) 73, Wayne N6KR
>>>
>>>
>>>
>>>
>>>
>>> ______________________________________________________________
>>>
>>>
>>>
>>>
>>>
>>> Elecraft mailing list
>>>
>>>
>>>
>>>
>>>
>>> Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
>>>
>>> http://mailman.qth.net/mailman/listinfo/elecraft
>>>
>>>
>>>
>>>
>>> Help: <http://mailman.qth.net/mmfaq.htm>
>>>
>>> http://mailman.qth.net/mmfaq.htm
>>>
>>>
>>>
>>>
>>> Post: <mailto:[hidden email] <mailto:[hidden email]> > mailto:[hidden email] <mailto:[hidden email]>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> This list hosted by: <http://www.qsl.net> http://www.qsl.net Please help
>>>
>>> support this
>>>
>>>
>>>
>>>
>>> email list: <http://www.qsl.net/donate.html>
>>>
>>> http://www.qsl.net/donate.html Message delivered to
>>>
>>>
>>>
>>>
>>> <mailto:[hidden email] <mailto:[hidden email]> > [hidden email] <mailto:[hidden email]>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> ______________________________________________________________
>>>
>>> Elecraft mailing list
>>>
>>> Home: <http://mailman.qth.net/mailman/listinfo/elecraft>
>>> http://mailman.qth.net/mailman/listinfo/elecraft
>>>
>>> Help: <http://mailman.qth.net/mmfaq.htm> http://mailman.qth.net/mmfaq.htm
>>>
>>> Post: <mailto:[hidden email] <mailto:[hidden email]> > mailto:[hidden email] <mailto:[hidden email]>
>>>
>>>
>>>
>>> This list hosted by: <http://www.qsl.net> http://www.qsl.net
>>>
>>> Please help support this email list: <http://www.qsl.net/donate.html>
>>> http://www.qsl.net/donate.html Message delivered to <mailto:[hidden email] <mailto:[hidden email]> >
>>> [hidden email] <mailto:[hidden email]>
>>>
>>> ______________________________________________________________
>>> Elecraft mailing list
>>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>>> Help: http://mailman.qth.net/mmfaq.htm
>>> Post: mailto:[hidden email] <mailto:[hidden email]>
>>>
>>> This list hosted by: http://www.qsl.net
>>> Please help support this email list: http://www.qsl.net/donate.html
>>> Message delivered to [hidden email] <mailto:[hidden email]>
>>>
>>> ______________________________________________________________
>>> Elecraft mailing list
>>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>>> Help: http://mailman.qth.net/mmfaq.htm
>>> Post: mailto:[hidden email]
>>>
>>> This list hosted by: http://www.qsl.net
>>> Please help support this email list: http://www.qsl.net/donate.html
>>> Message delivered to [hidden email]
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]

alorona

Re: CW rise time setting

In reply to this post by Richard Stutsman

I used to think that the rise and fall times of the CW pulse didn't really matter much to the sideband levels; I believed that it was more a function of the waveshaping, especially at the corners of the pulse.

But I just ran a quick simulation of a pulse train going through both a raised cosine and then a sigmoid filter (because those two have been mentioned in this thread) and the rise time definitely does affect the pulse sidebands.

The reduction in the sideband levels (what some folks here called 'clicks'... not sure that's a good name for this) varies depending on where you measure it, but in general the sidebands will drop anywhere from 0 to 12 dB-- sometimes less, sometimes more-- when you go from 2 msec to 8 msec rise/fall times. For instance, arbitrarily choosing an offset of 500 Hz from the carrier, the sideband drops by 11 dB for the longer rise time. That turns out to be a fairly typical value. And by the way, in general the sigmoid does a better job than a raised cosine.

Arbitrarily defining the occupied bandwidth as the -60 dBc points of the spectrum, and using the sigmoid function with an exponent of -1, the bandwidth of the rise time = 8 msec pulse is 420 Hz versus 640 Hz for the pulse with 2 msec rise time. It's not an enormous difference, but it is something.

Anyway, there's another data point for the discussion.

R,

Al W6LX
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

k6dgw

Re: CW rise time setting

Which sigmoid function did you model, Al?

73,
Fred ["Skip"] K6DGW
Sparks NV DM09dn
Washoe County

On 6/9/2020 11:07 AM, Al Lorona wrote:

> I used to think that the rise and fall times of the CW pulse didn't really matter much to the sideband levels; I believed that it was more a function of the waveshaping, especially at the corners of the pulse.
>
> But I just ran a quick simulation of a pulse train going through both a raised cosine and then a sigmoid filter (because those two have been mentioned in this thread) and the rise time definitely does affect the pulse sidebands.
>
> The reduction in the sideband levels (what some folks here called 'clicks'... not sure that's a good name for this) varies depending on where you measure it, but in general the sidebands will drop anywhere from 0 to 12 dB-- sometimes less, sometimes more-- when you go from 2 msec to 8 msec rise/fall times. For instance, arbitrarily choosing an offset of 500 Hz from the carrier, the sideband drops by 11 dB for the longer rise time. That turns out to be a fairly typical value. And by the way, in general the sigmoid does a better job than a raised cosine.
>
> Arbitrarily defining the occupied bandwidth as the -60 dBc points of the spectrum, and using the sigmoid function with an exponent of -1, the bandwidth of the rise time = 8 msec pulse is 420 Hz versus 640 Hz for the pulse with 2 msec rise time. It's not an enormous difference, but it is something.
>
> Anyway, there's another data point for the discussion.
>
> R,
>
> Al W6LX
>

Jim Brown-10

Re: CW rise time setting

In reply to this post by alorona

On 6/9/2020 11:07 AM, Al Lorona wrote:
> The reduction in the sideband levels (what some folks here called 'clicks'... not sure that's a good name for this) varies depending on where you measure it, but in general the sidebands will drop anywhere from 0 to 12 dB-- sometimes less, sometimes more-- when you go from 2 msec to 8 msec rise/fall times. For instance, arbitrarily choosing an offset of 500 Hz from the carrier, the sideband drops by 11 dB for the longer rise time. That turns out to be a fairly typical value. And by the way, in general the sigmoid does a better job than a raised cosine.

I've not looked at the math, but I've measured more than a half dozen
radios, most with variable time constant shaping, and Elecraft with
their fixed sigmoidal shaping. The data is here.

http://k9yc.com/P3_Spectrum_Measurements.pdf

A K3 at 25W (driving a KPA500 to full power) is 50 dB down 230 Hz either
side of the signal, 60 dB down at 305 Hz. At 40W driving a legal limit
tube amp (Ten Tec Titan) sidebands at the power amp output are 50 dB
down at 235 Hz, 60 dB down at 335 Hz.

A neighbor's FT1000 Mark V Field was 50 dB down at 665 Hz.

Another neighbor's FTDX5000 set for 6 msec was 50 dB down at 410 Hz, 60
dB down at 1.05 kHz before the firmware update. It improved to -50 dB at
310 Hz and -60 at 535 Hz after the update.

73, Jim K9YC

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

alorona

Re: CW rise time setting

In reply to this post by k6dgw

I used:

1/(1+exp(-x))

where x = the pulse train. I then modulated a 7 MHz carrier with the result (although the results don't care what the RF frequency is, of course).

Al W6LX

>>>Which sigmoid function did you model, Al?

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Edward H Russell

Re: CW rise time setting

In reply to this post by Jim Brown-10

Jim,

Interesting results, especially the dramatic difference in the FTDX5000 after the firmware update. The K3 looks great. Hopefully the K4 will be at least as good.

Ed / w2rf

-----Original Message-----
From: [hidden email] <[hidden email]> On Behalf Of Jim Brown
Sent: Tuesday, June 9, 2020 3:46 PM
To: [hidden email]
Subject: Re: [Elecraft] CW rise time setting

On 6/9/2020 11:07 AM, Al Lorona wrote:

> The reduction in the sideband levels (what some folks here called 'clicks'... not sure that's a good name for this) varies depending on where you measure it, but in general the sidebands will drop anywhere from 0 to 12 dB-- sometimes less, sometimes more-- when you go from 2 msec to 8 msec rise/fall times. For instance, arbitrarily choosing an offset of 500 Hz from the carrier, the sideband drops by 11 dB for the longer rise time. That turns out to be a fairly typical value. And by the way, in general the sigmoid does a better job than a raised cosine.

I've not looked at the math, but I've measured more than a half dozen radios, most with variable time constant shaping, and Elecraft with their fixed sigmoidal shaping. The data is here.

<http://k9yc.com/P3_Spectrum_Measurements.pdf> http://k9yc.com/P3_Spectrum_Measurements.pdf

A K3 at 25W (driving a KPA500 to full power) is 50 dB down 230 Hz either side of the signal, 60 dB down at 305 Hz. At 40W driving a legal limit tube amp (Ten Tec Titan) sidebands at the power amp output are 50 dB down at 235 Hz, 60 dB down at 335 Hz.

A neighbor's FT1000 Mark V Field was 50 dB down at 665 Hz.

Another neighbor's FTDX5000 set for 6 msec was 50 dB down at 410 Hz, 60 dB down at 1.05 kHz before the firmware update. It improved to -50 dB at

310 Hz and -60 at 535 Hz after the update.

73, Jim K9YC

______________________________________________________________

Elecraft mailing list

Home: <http://mailman.qth.net/mailman/listinfo/elecraft> http://mailman.qth.net/mailman/listinfo/elecraft

Help: <http://mailman.qth.net/mmfaq.htm> http://mailman.qth.net/mmfaq.htm

Post: <mailto:[hidden email]> mailto:[hidden email]

This list hosted by: <http://www.qsl.net> http://www.qsl.net

Please help support this email list: <http://www.qsl.net/donate.html> http://www.qsl.net/donate.html Message delivered to <mailto:[hidden email]> [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Bob Wilson, N6TV

Re: CW rise time setting

In reply to this post by Edward H Russell

On Mon, Jun 8, 2020 at 12:31 PM E.H. Russell <[hidden email]> wrote:

> I suppose shaping of the curve corners removes harmonics introduced by the
> abrupt transitions, allowing an accelerated ramp inbetween. But does this
> really reduce the total rise time to 2.5ms? It seems that the softening
> process must take some time. Wish I had a K3 here to scope against other
> radios. Anything published out there?
>

Ed,

Per your request, I am publishing this scope screen capture which plots the
CW rise time in my K3 with the KSYN3A synthesizer upgrade. It's about 4 ms
from 0 RF to full RF (2 ms per horizontal division):

https://www.kkn.net/~n6tv/N6TV_K3_Ser_1494_FW_05.64_TX_DLY_8_CW_QRQ_OFF.png

The vertical markers are there to illustrate that *CONFIG:TX DLY nor 008*
provides only about *6 ms* of RF delay after "KEY OUT" goes to ground, not
8 ms, and there is jitter in that delay as well (not shown). If CW QRQ
mode is enabled, the delay drops to about 4.6 ms and the TX DLY setting is
completely ignored.

This was discussed here two years ago. See this post for suggestions on
how to avoid hot-switching a non-Elecraft amplifier driven by a K3 or K3S:

http://elecraft.365791.n2.nabble.com/Default-K3-transmit-delay-may-be-too-short-for-slow-QRO-amplifiers-td7641779.html

73,
Bob, N6TV
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Edward H Russell

Re: CW rise time setting

Bob,

Thanks for the scope shot and info. I compared the waveform to the QST K3S review, which seems to show a little over 10ms before RF appears. Is this because they used different settings?

Will be interesting to see how the new radio CW looks in time and frequency domains. Also how the turnaround latency is managed.

Tks,

73 Ed w2rf

From: Bob Wilson, N6TV <[hidden email]>
Sent: Thursday, June 11, 2020 2:05 PM
To: Elecraft Reflector <[hidden email]>
Cc: E.H. Russell <[hidden email]>
Subject: Re: [Elecraft] CW rise time setting

On Mon, Jun 8, 2020 at 12:31 PM E.H. Russell <[hidden email] <mailto:[hidden email]> > wrote:

I suppose shaping of the curve corners removes harmonics introduced by the abrupt transitions, allowing an accelerated ramp inbetween. But does this really reduce the total rise time to 2.5ms? It seems that the softening process must take some time. Wish I had a K3 here to scope against other radios. Anything published out there?

Ed,

Per your request, I am publishing this scope screen capture which plots the CW rise time in my K3 with the KSYN3A synthesizer upgrade. It's about 4 ms from 0 RF to full RF (2 ms per horizontal division):

https://www.kkn.net/~n6tv/N6TV_K3_Ser_1494_FW_05.64_TX_DLY_8_CW_QRQ_OFF.png

The vertical markers are there to illustrate that CONFIG:TX DLY nor 008 provides only about 6 ms of RF delay after "KEY OUT" goes to ground, not 8 ms, and there is jitter in that delay as well (not shown). If CW QRQ mode is enabled, the delay drops to about 4.6 ms and the TX DLY setting is completely ignored.

This was discussed here two years ago. See this post for suggestions on how to avoid hot-switching a non-Elecraft amplifier driven by a K3 or K3S:

http://elecraft.365791.n2.nabble.com/Default-K3-transmit-delay-may-be-too-short-for-slow-QRO-amplifiers-td7641779.html

73,

Bob, N6TV

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Bob Wilson, N6TV

Re: CW rise time setting

The plot in the *QST* review of the K3S shows the delay between "key
closure" (the KEY jack of K3S) and RF out. My plot shows the delay between
"amp relay closure" (the KEY OUT jack of K3S) and RF, which is more
critical.

Most folks assume there will be no delay between KEY closure and KEY OUT
closure, but there is an extra delay (of about 5 ms, minimum) in the K3S.
The same applies to PTT IN closure and KEY OUT closure; that is, there is
an unexplained fixed 5 ms delay in the K3S, probably due to slow firmware
logic testing for TX Inhibit or an intentional enforcement of some minimum
delay in RF output.

Most radios close KEY OUT immediately upon key closure of either the KEY
jack or the PTT IN jack. The K3 does not; it "hesitates" before closing
KEY OUT.

I'm also hoping the K4 will eliminate this unusual behavior.

73,
Bob, N6TV

On Fri, Jun 12, 2020 at 3:23 AM E.H. Russell <[hidden email]> wrote:

> Bob,
>
>
>
> Thanks for the scope shot and info. I compared the waveform to the QST K3S
> review, which seems to show a little over 10ms before RF appears. Is this
> because they used different settings?
>
>
>
> Will be interesting to see how the new radio CW looks in time and
> frequency domains. Also how the turnaround latency is managed.
>
>
>
> Tks,
>
> 73 Ed w2rf
>
>
>
>
>
>
>
> *From:* Bob Wilson, N6TV <[hidden email]>
> *Sent:* Thursday, June 11, 2020 2:05 PM
> *To:* Elecraft Reflector <[hidden email]>
> *Cc:* E.H. Russell <[hidden email]>
> *Subject:* Re: [Elecraft] CW rise time setting
>
>
>
> On Mon, Jun 8, 2020 at 12:31 PM E.H. Russell <[hidden email]> wrote:
>
> I suppose shaping of the curve corners removes harmonics introduced by the
> abrupt transitions, allowing an accelerated ramp inbetween. But does this
> really reduce the total rise time to 2.5ms? It seems that the softening
> process must take some time. Wish I had a K3 here to scope against other
> radios. Anything published out there?
>
>
>
> Ed,
>
>
>
> Per your request, I am publishing this scope screen capture which plots
> the CW rise time in my K3 with the KSYN3A synthesizer upgrade. It's about
> 4 ms from 0 RF to full RF (2 ms per horizontal division):
>
>
>
> https://www.kkn.net/~n6tv/N6TV_K3_Ser_1494_FW_05.64_TX_DLY_8_CW_QRQ_OFF.png
>
>
>
> The vertical markers are there to illustrate that *CONFIG:TX DLY nor 008*
> provides only about *6 ms* of RF delay after "KEY OUT" goes to ground,
> not 8 ms, and there is jitter in that delay as well (not shown). If CW QRQ
> mode is enabled, the delay drops to about 4.6 ms and the TX DLY setting is
> completely ignored.
>
>
>
> This was discussed here two years ago. See this post for suggestions on
> how to avoid hot-switching a non-Elecraft amplifier driven by a K3 or K3S:
>
>
>
>
> http://elecraft.365791.n2.nabble.com/Default-K3-transmit-delay-may-be-too-short-for-slow-QRO-amplifiers-td7641779.html
>
>
>
>
> 73,
>
> Bob, N6TV
>

ANDY DURBIN

CW rise time setting

In reply to this post by Richard Stutsman

"Most folks assume there will be no delay between KEY closure and KEY OUT closure, but there is an extra delay (of about 5 ms, minimum) in the K3S."

I suppose I'm not in the "most folks" group. I measured this delay when investigating the keying characterisitcs of my Kenwood TS-590S. Key down to amplifier relay closure was 2 ms for the configuration under test. RF started rising about 14 ms after key closure.

73,
Andy k3wyc
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Jim Brown-10

Evaluating Keying Characteristics

In reply to this post by Edward H Russell

On 6/12/2020 3:23 AM, E.H. Russell wrote:
> Thanks for the scope shot and info. I compared the waveform to the QST K3S review, which seems to show a little over 10ms before RF appears. Is this because they used different settings?

I repeat my advice that keying characteristics must be evaluated in the
FREQUENCY domain (as occupied bandwidth), NOT in the TIME domain
(oscilloscope). The reason has been stated here many times -- it's not
only the TIME of the transition that determines bandwidth, it is the
SHAPE of the transition from off to on and on to off that matters, NOT
the time it takes, and this cannot be seen on a scope.

73, Jim K9YC
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]