Default K3 transmit delay may be too short for slow QRO amplifiers

classic Classic list List threaded Threaded
17 messages Options
Reply | Threaded
Open this post in threaded view
|

Default K3 transmit delay may be too short for slow QRO amplifiers

wayne burdick
Administrator
It has come to our attention (thanks to N7YK and N6TV) that the default value of CONFIG:TX DLY, 8 ms, may actually be closer to 5 ms.

The associated T/R control firmware hasn’t changed in a long time, so hopefully there aren’t many users affected (or we would have heard about it). The shorter delay may be related to the improved KSYN3A synth, which switches faster than the original. We’re looking into this, and will update the displayed value to reflect the actual value.

Details:

When using a QRO amp with the K3, the KEY OUT line from the radio is used to key the amplifier, setting it up to receive RF. Since the K3 has very fast QSK, the amp’s T/R control circuitry has to be similarly fast.

Those of you who are using KPA500s and KPA1500s need not be concerned about the K3’s fast QSK speed, since our amps use PIN diode T/R switching. They can easily keep up with the K3 at any code speed and with the default setting of the CONFIG:TX DLY menu entry.

However, some other amps, especially older ones, use slow T/R relays. If the transceiver generates RF before the amp is ready, this might result in key clicks.

(In practice there’s a bit of a grace period because the RF keying envelope rises very slowly. It’s a "raised-cosine sigmoid," resulting in the transceiver itself having incredibly clean keying sidebands. See K9YC white paper on this topic.)

If you suspect your amp has a slow T/R relay, you should do a listening test, or better yet have someone else listen from a good distance. If you (or they) hear any evidence of clicks, increase the value of CONFIG:TX DLY.

Use the lowest setting that eliminates the clicks. CW keying at higher code speeds may be affected if the value is too high. Other modes not using fast QSK will generally be unaffected (voice and data).

An alternative to increasing TX DLY is to use a foot switch, as do many users of old amps regardless of transceiver type.

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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Dimitry Borzenko
Hello.
Why in the 20 century we need to use foot switch?
Why not to do longer TX delay?
Why I can not set 200mS TX DLY ?
For QSK operation or for use with new fast switched TX/RX relays or PIN
diodes user can to set minimum TX DLY.
Today is possible to change only firmware and do not need to do any
hardware changes.
TRX mast be more flexible.
You can say , What about CW??.
Nothing , in winkeyer possible to set long TX DLY why in K3S is
impossible?
All possible, need only wish to do.

Best Wishes.


------ Original Message ------
From: "Wayne Burdick" <[hidden email]>
To: "Elecraft Reflector" <[hidden email]>
Sent: 08.06.2018 2:43:13
Subject: [Elecraft] Default K3 transmit delay may be too short for slow
QRO amplifiers

>It has come to our attention (thanks to N7YK and N6TV) that the default
>value of CONFIG:TX DLY, 8 ms, may actually be closer to 5 ms.
>
>The associated T/R control firmware hasn’t changed in a long time, so
>hopefully there aren’t many users affected (or we would have heard
>about it). The shorter delay may be related to the improved KSYN3A
>synth, which switches faster than the original. We’re looking into
>this, and will update the displayed value to reflect the actual value.
>
>Details:
>
>When using a QRO amp with the K3, the KEY OUT line from the radio is
>used to key the amplifier, setting it up to receive RF. Since the K3
>has very fast QSK, the amp’s T/R control circuitry has to be similarly
>fast.
>
>Those of you who are using KPA500s and KPA1500s need not be concerned
>about the K3’s fast QSK speed, since our amps use PIN diode T/R
>switching. They can easily keep up with the K3 at any code speed and
>with the default setting of the CONFIG:TX DLY menu entry.
>
>However, some other amps, especially older ones, use slow T/R relays.
>If the transceiver generates RF before the amp is ready, this might
>result in key clicks.
>
>(In practice there’s a bit of a grace period because the RF keying
>envelope rises very slowly. It’s a "raised-cosine sigmoid," resulting
>in the transceiver itself having incredibly clean keying sidebands. See
>K9YC white paper on this topic.)
>
>If you suspect your amp has a slow T/R relay, you should do a listening
>test, or better yet have someone else listen from a good distance. If
>you (or they) hear any evidence of clicks, increase the value of
>CONFIG:TX DLY.
>
>Use the lowest setting that eliminates the clicks. CW keying at higher
>code speeds may be affected if the value is too high. Other modes not
>using fast QSK will generally be unaffected (voice and data).
>
>An alternative to increasing TX DLY is to use a foot switch, as do many
>users of old amps regardless of transceiver type.
>
>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]

______________________________________________________________
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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Jim Brown-10
On 6/8/2018 1:05 AM, Dimitry Borzenko wrote:
> Why in the 20 century we need to use foot switch?

Hi Dimitry,

I strongly agree. I HATE foot switches, because I'm very poorly
coordinated. :) I've never used anything but VOX in my own station for
any mode, and I do RTTY, CW, and SSB contesting, run FT8, JT65, JT9,
MSK144 for small signal work. With every rig I've owned since 2003 (when
I had time to back on the air) I've been able to set the delay between
transmitted RF and the DC signal to pull in the TR relay for a power amp
to the value that my amps have needed. The high quality, well designed
amps I've used have never needed more than 8 msec.

That's me. Everyone is different. :)

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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Igor Sokolov-2
In reply to this post by wayne burdick
Wayne it is a pity that TX delay setting affects CW timing. Changing TX
delay from default even one or two steps makes CW much worth. Can you
figure out a way to make CW timing independent from TX delay? With 5 ms
TX delay working in many contests I already had to change a pair of
relays in the amplifier. 10-12 ms would be fine for me. 20 ms will
probably suit the majority of users. But 5 ms is way too fast and
unfortunately cannot be altered without spoiling CW sending. I know that
I could build external sequencer but this does not look neat enough
especially for portable field operation. Every other radio I had allowed
TX delay adjustment to at least 20 ms.

73, Igor UA9CDC

08.06.2018 4:43, Wayne Burdick пишет:

> It has come to our attention (thanks to N7YK and N6TV) that the default value of CONFIG:TX DLY, 8 ms, may actually be closer to 5 ms.
>
> The associated T/R control firmware hasn’t changed in a long time, so hopefully there aren’t many users affected (or we would have heard about it). The shorter delay may be related to the improved KSYN3A synth, which switches faster than the original. We’re looking into this, and will update the displayed value to reflect the actual value.
>
> Details:
>
> When using a QRO amp with the K3, the KEY OUT line from the radio is used to key the amplifier, setting it up to receive RF. Since the K3 has very fast QSK, the amp’s T/R control circuitry has to be similarly fast.
>
> Those of you who are using KPA500s and KPA1500s need not be concerned about the K3’s fast QSK speed, since our amps use PIN diode T/R switching. They can easily keep up with the K3 at any code speed and with the default setting of the CONFIG:TX DLY menu entry.
>
> However, some other amps, especially older ones, use slow T/R relays. If the transceiver generates RF before the amp is ready, this might result in key clicks.
>
> (In practice there’s a bit of a grace period because the RF keying envelope rises very slowly. It’s a "raised-cosine sigmoid," resulting in the transceiver itself having incredibly clean keying sidebands. See K9YC white paper on this topic.)
>
> If you suspect your amp has a slow T/R relay, you should do a listening test, or better yet have someone else listen from a good distance. If you (or they) hear any evidence of clicks, increase the value of CONFIG:TX DLY.
>
> Use the lowest setting that eliminates the clicks. CW keying at higher code speeds may be affected if the value is too high. Other modes not using fast QSK will generally be unaffected (voice and data).
>
> An alternative to increasing TX DLY is to use a foot switch, as do many users of old amps regardless of transceiver type.
>
> 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]

______________________________________________________________
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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Clay Autery-2
Mmmmm.... even without a firmware change, my TX DLY already has the
range from "NOR 8ms" to "020" (8-20 ms).  Granted, that 20ms MAY be
somewhat shorter than 20ms, as described by Wayne regarding the 8ms
"NOR" setting.
I suspect that they will now push this up in the queue of things to mod
in the next firmware release.

Strange (to me) that so many find foot switch use in CW such a
non-starter.  Since I uses one exclusively in SSB, I guess it is only
natural for me to use one in CW...

73,

______________________
Clay Autery, KY5G
(318) 518-1389

On 08-Jun-18 05:22, Igor Sokolov wrote:

> Wayne it is a pity that TX delay setting affects CW timing. Changing
> TX delay from default even one or two steps makes CW much worth. Can
> you figure out a way to make CW timing independent from TX delay? With
> 5 ms TX delay working in many contests I already had to change a pair
> of relays in the amplifier. 10-12 ms would be fine for me. 20 ms will
> probably suit the majority of users. But 5 ms is way too fast and
> unfortunately cannot be altered without spoiling CW sending. I know
> that I could build external sequencer but this does not look neat
> enough especially for portable field operation. Every other radio I
> had allowed TX delay adjustment to at least 20 ms.
>
> 73, Igor UA9CDC
>
> 08.06.2018 4:43, Wayne Burdick пишет:
>> It has come to our attention (thanks to N7YK and N6TV) that the
>> default value of CONFIG:TX DLY, 8 ms, may actually be closer to 5 ms.
>>
>> The associated T/R control firmware hasn’t changed in a long time, so
>> hopefully there aren’t many users affected (or we would have heard
>> about it). The shorter delay may be related to the improved KSYN3A
>> synth, which switches faster than the original. We’re looking into
>> this, and will update the displayed value to reflect the actual value.
>>
>> Details:
>>
>> When using a QRO amp with the K3, the KEY OUT line from the radio is
>> used to key the amplifier, setting it up to receive RF. Since the K3
>> has very fast QSK, the amp’s T/R control circuitry has to be
>> similarly fast.
>>
>> Those of you who are using KPA500s and KPA1500s need not be concerned
>> about the K3’s fast QSK speed, since our amps use PIN diode T/R
>> switching. They can easily keep up with the K3 at any code speed and
>> with the default setting of the CONFIG:TX DLY menu entry.
>>
>> However, some other amps, especially older ones, use slow T/R relays.
>> If the transceiver generates RF before the amp is ready, this might
>> result in key clicks.
>>
>> (In practice there’s a bit of a grace period because the RF keying
>> envelope rises very slowly. It’s a "raised-cosine sigmoid," resulting
>> in the transceiver itself having incredibly clean keying sidebands.
>> See K9YC white paper on this topic.)
>>
>> If you suspect your amp has a slow T/R relay, you should do a
>> listening test, or better yet have someone else listen from a good
>> distance. If you (or they) hear any evidence of clicks, increase the
>> value of CONFIG:TX DLY.
>>
>> Use the lowest setting that eliminates the clicks. CW keying at
>> higher code speeds may be affected if the value is too high. Other
>> modes not using fast QSK will generally be unaffected (voice and data).
>>
>> An alternative to increasing TX DLY is to use a foot switch, as do
>> many users of old amps regardless of transceiver type.
>>
>> 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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

ke9uw
We’re thinking that the first CW character is shortened by a longer delay (QSK), and that is the issue...not that a longer delay is prevented?
Is that what I’m reading?
QSK with a foot switch sounds interesting to watch...

Chuck
KE9UW

Sent from my iPhone, cjack

> On Jun 8, 2018, at 9:02 AM, Clay Autery <[hidden email]> wrote:
>
> Mmmmm.... even without a firmware change, my TX DLY already has the range from "NOR 8ms" to "020" (8-20 ms).  Granted, that 20ms MAY be somewhat shorter than 20ms, as described by Wayne regarding the 8ms "NOR" setting.
> I suspect that they will now push this up in the queue of things to mod in the next firmware release.
>
> Strange (to me) that so many find foot switch use in CW such a non-starter.  Since I uses one exclusively in SSB, I guess it is only natural for me to use one in CW...
>
> 73,
>
> ______________________
> Clay Autery, KY5G
> (318) 518-1389
>
>> On 08-Jun-18 05:22, Igor Sokolov wrote:
>> Wayne it is a pity that TX delay setting affects CW timing. Changing TX delay from default even one or two steps makes CW much worth. Can you figure out a way to make CW timing independent from TX delay? With 5 ms TX delay working in many contests I already had to change a pair of relays in the amplifier. 10-12 ms would be fine for me. 20 ms will probably suit the majority of users. But 5 ms is way too fast and unfortunately cannot be altered without spoiling CW sending. I know that I could build external sequencer but this does not look neat enough especially for portable field operation. Every other radio I had allowed TX delay adjustment to at least 20 ms.
>>
>> 73, Igor UA9CDC
>>
>> 08.06.2018 4:43, Wayne Burdick пишет:
>>> It has come to our attention (thanks to N7YK and N6TV) that the default value of CONFIG:TX DLY, 8 ms, may actually be closer to 5 ms.
>>>
>>> The associated T/R control firmware hasn’t changed in a long time, so hopefully there aren’t many users affected (or we would have heard about it). The shorter delay may be related to the improved KSYN3A synth, which switches faster than the original. We’re looking into this, and will update the displayed value to reflect the actual value.
>>>
>>> Details:
>>>
>>> When using a QRO amp with the K3, the KEY OUT line from the radio is used to key the amplifier, setting it up to receive RF. Since the K3 has very fast QSK, the amp’s T/R control circuitry has to be similarly fast.
>>>
>>> Those of you who are using KPA500s and KPA1500s need not be concerned about the K3’s fast QSK speed, since our amps use PIN diode T/R switching. They can easily keep up with the K3 at any code speed and with the default setting of the CONFIG:TX DLY menu entry.
>>>
>>> However, some other amps, especially older ones, use slow T/R relays. If the transceiver generates RF before the amp is ready, this might result in key clicks.
>>>
>>> (In practice there’s a bit of a grace period because the RF keying envelope rises very slowly. It’s a "raised-cosine sigmoid," resulting in the transceiver itself having incredibly clean keying sidebands. See K9YC white paper on this topic.)
>>>
>>> If you suspect your amp has a slow T/R relay, you should do a listening test, or better yet have someone else listen from a good distance. If you (or they) hear any evidence of clicks, increase the value of CONFIG:TX DLY.
>>>
>>> Use the lowest setting that eliminates the clicks. CW keying at higher code speeds may be affected if the value is too high. Other modes not using fast QSK will generally be unaffected (voice and data).
>>>
>>> An alternative to increasing TX DLY is to use a foot switch, as do many users of old amps regardless of transceiver type.
>>>
>>> 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]
______________________________________________________________
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]
Chuck, KE9UW
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Elecraft mailing list
In reply to this post by Jim Brown-10
 I have a foot switch but only use it generally on Sunday of a weekend contest.  That's when my throat needs unfleming or whatever that iswhen your voice is overworked but I always just use vox if possible.  In the older days it was PTT but with a desk mic and not a foot switch.But my foot switch is always there if I need or want it, just have to drag it forward a little and place it in the "spot" where my foot can get to it.
I'm not a phone guy but when I bought my K3 in 2009 I opted for the dvk and that is more valuable than a footswitch will ever be.  Too badI only get to use it twice a year in ARRL & CQ dx phone tests, hi.
BillK3WJV

    On Friday, June 8, 2018, 4:26:22 AM EDT, Jim Brown <[hidden email]> wrote:  
 
 On 6/8/2018 1:05 AM, Dimitry Borzenko wrote:
> Why in the 20 century we need to use foot switch?

Hi Dimitry,

I strongly agree. I HATE foot switches, because I'm very poorly
coordinated. :) I've never used anything but VOX in my own station for
any mode, and I do RTTY, CW, and SSB contesting, run FT8, JT65, JT9,
MSK144 for small signal work. With every rig I've owned since 2003 (when
I had time to back on the air) I've been able to set the delay between
transmitted RF and the DC signal to pull in the TR relay for a power amp
to the value that my amps have needed. The high quality, well designed
amps I've used have never needed more than 8 msec.

That's me. Everyone is different. :)

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]
 
______________________________________________________________
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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Bob Wilson, N6TV
In reply to this post by Igor Sokolov-2
On Fri, Jun 8, 2018 at 3:22 AM, Igor Sokolov <[hidden email]> wrote:

> Wayne it is a pity that TX delay setting affects CW timing. Changing TX
> delay from default even one or two steps makes CW much worth. Can you
> figure out a way to make CW timing independent from TX delay? With 5 ms TX
> delay working in many contests I already had to change a pair of relays in
> the amplifier. 10-12 ms would be fine for me. 20 ms will probably suit the
> majority of users. But 5 ms is way too fast and unfortunately cannot be
> altered without spoiling CW sending. I know that I could build external
> sequencer but this does not look neat enough especially for portable field
> operation. Every other radio I had allowed TX delay adjustment to at least
> 20 ms.


Instead of weighting, as Wayne suggested, set "Keying Compensation" or "1st
dit extension" to match the actual TX DLY.  This will restore proper CW
timing ratios in the transmitted RF, as you increase TX DLY, since every
dot or dash (or the first one in a message) will be lengthened by a fixed
number of ms.

No footswitch is required, since external keying methods can generate a
leading PTT delay.

If you are using the internal keyer of the K3 for hand sending,
compensation is automatic.  No need for any adjustment; use any TX DLY
value you need, and the transmitted dot/dash ratios will still be perfect
(I've measured it on a scope).

If using external keying with a WinKey, use the menu setting for "Keying
Compensation" (QSK), or "1st dit extension" (Semi-Breakin).  to apply
matching keying compensation.  Example:

CONFIG: TX DLY 010
Actual measured TX Delay  7-8 ms (with jitter)
Set External Keying Compensation (for QSK) to:  7 ms
Set 1st Dit Extension (semi-breakin) to:  7 ms


If using a microHAM Device with a built-in WinKey, the menu settings for
these values were (unfortunately) moved from the WinKey tab of the Router
software.  Instead, they are now a single setting under the PTT tab, called
"R/T Delay" (improperly documented).  So, in the microHAM Router, PTT tab:

Set R/T Delay to 7 ms


If using a LogiKey K1 or similar:

Set the "K Factor" (keying compensatin) to 7


If using Win-Test with CONFIG:PTT-KEY OFF-DTR for QSK keying over the
serial port DTR pin, or an external keying circuit, set the Keying
Compensation in the Interface Configuration menu to 7 ms (this a new
feature in Win-Test 4.28.0-dev).  If using semi-breakin, set CONFIG:PTT-KEY
to RTS-DTR and tell Win-Test to use a PTT Lead time of 10-20 ms, 0 ms
keying compensation.

Most WinKey/N1MM users can simply rely on setting the WinKey PTT Lead time
to a sufficient value.  Unfortunately, the WinKey PTT lead times are set in
increments of 10 ms (0, 10, 20, 30 ms), despite the menu settings.  If you
ask for 11 ms of PTT delay in a WinKey, you actually get 20 ms of PTT delay.

Note that the K3 will safely ignore any PTT lead time less than TX DLY, so
if you set PTT LEAD to 3 ms you will actually get 7 ms and all elements
(QSK) or the first element (semi break-in) will be shortened by 7 ms.

FYI, the OM Power OM2000A+ non-QSK amplifier manual recommends a TX DLY of
20 ms (!).

Final note:  if using CW QRQ mode, the K3 ignores TX DLY completely, and
you get 4-5 ms of TX DELAY (fixed, with jitter).  Set keying compensation
to 4 ms.

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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Josh Fiden
In reply to this post by Elecraft mailing list
My poor impulse control makes it impossible to operate within FCC regs using vox during the inevitable lid encounters.

73
Josh W6XU

Sent from my mobile device

> On Jun 8, 2018, at 10:07 AM, Bill Stravinsky via Elecraft <[hidden email]> wrote:
>
> Why in the 20 century we need to use foot switch?

______________________________________________________________
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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Igor Sokolov-2
In reply to this post by Bob Wilson, N6TV
Dear Bob,

My setup is optimized for traveling/field operation. I do not need 
additional boxes in the form of Winkey or external QSK. I use external
keying from computer through USB-COM adapter. The logger that does CW
keying is TR4W. No keying compensation helps to achieve correct CW
weight if TX Delay in K3 is set to say 20ms and CW speed used is above
30WPM. Dependency of CW weight from TX Delay is only inherent in K3. All
other rigs do not have this dependency.


73, Igor UA9CDC


08.06.2018 23:12, Bob Wilson, N6TV пишет:

> On Fri, Jun 8, 2018 at 3:22 AM, Igor Sokolov <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     Wayne it is a pity that TX delay setting affects CW timing.
>     Changing TX delay from default even one or two steps makes CW much
>     worth. Can you figure out a way to make CW timing independent from
>     TX delay? With 5 ms TX delay working in many contests I already
>     had to change a pair of relays in the amplifier. 10-12 ms would be
>     fine for me. 20 ms will probably suit the majority of users. But 5
>     ms is way too fast and unfortunately cannot be altered without
>     spoiling CW sending. I know that I could build external sequencer
>     but this does not look neat enough especially for portable field
>     operation. Every other radio I had allowed TX delay adjustment to
>     at least 20 ms.
>
>
> Instead of weighting, as Wayne suggested, set "Keying Compensation" or
> "1st dit extension" to match the actual TX DLY.  This will restore
> proper CW timing ratios in the transmitted RF, as you increase TX DLY,
> since every dot or dash (or the first one in a message) will be
> lengthened by a fixed number of ms.
>
> No footswitch is required, since external keying methods can generate
> a leading PTT delay.
>
> If you are using the internal keyer of the K3 for hand sending,
> compensation is automatic.  No need for any adjustment; use any TX DLY
> value you need, and the transmitted dot/dash ratios will still be
> perfect (I've measured it on a scope).
>
> If using external keying with a WinKey, use the menu setting for
> "Keying Compensation" (QSK), or "1st dit extension" (Semi-Breakin). 
> to apply matching keying compensation.  Example:
>
>     CONFIG: TX DLY 010
>     Actual measured TX Delay  7-8 ms (with jitter)
>     Set External Keying Compensation (for QSK) to:  7 ms
>     Set 1st Dit Extension (semi-breakin) to:  7 ms
>
>
> If using a microHAM Device with a built-in WinKey, the menu settings
> for these values were (unfortunately) moved from the WinKey tab of the
> Router software.  Instead, they are now a single setting under the PTT
> tab, called "R/T Delay" (improperly documented).  So, in the microHAM
> Router, PTT tab:
>
>     Set R/T Delay to 7 ms
>
>
> If using a LogiKey K1 or similar:
>
>     Set the "K Factor" (keying compensatin) to 7
>
>
> If using Win-Test with CONFIG:PTT-KEY OFF-DTR for QSK keying over the
> serial port DTR pin, or an external keying circuit, set the Keying
> Compensation in the Interface Configuration menu to 7 ms (this a new
> feature in Win-Test 4.28.0-dev).  If using semi-breakin, set
> CONFIG:PTT-KEY to RTS-DTR and tell Win-Test to use a PTT Lead time of
> 10-20 ms, 0 ms keying compensation.
>
> Most WinKey/N1MM users can simply rely on setting the WinKey PTT Lead
> time to a sufficient value. Unfortunately, the WinKey PTT lead times
> are set in increments of 10 ms (0, 10, 20, 30 ms), despite the menu
> settings.  If you ask for 11 ms of PTT delay in a WinKey, you actually
> get 20 ms of PTT delay.
>
> Note that the K3 will safely ignore any PTT lead time less than TX
> DLY, so if you set PTT LEAD to 3 ms you will actually get 7 ms and all
> elements (QSK) or the first element (semi break-in) will be shortened
> by 7 ms.
>
> FYI, the OM Power OM2000A+ non-QSK amplifier manual recommends a TX
> DLY of 20 ms (!).
>
> Final note:  if using CW QRQ mode, the K3 ignores TX DLY completely,
> and you get 4-5 ms of TX DELAY (fixed, with jitter).  Set keying
> compensation to 4 ms.
>
> 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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Grant Youngman-2
I’m confused.  If you’re in the “field”, are you keying a high power amp with relay T/R switching??   Is it possible the problem is associated with the way your computer software keys the K3?  Have you tested “all other rigs”?  Perhaps it’s a “system” issue? How does your system behave when keying the K3 directly— e.g., no computer, just a key — manual or internal keyer.

Grant NQ5T
K3 #2091, KX3 #8342

> On Jun 8, 2018, at 3:37 PM, Igor Sokolov <[hidden email]> wrote:
>
> Dear Bob,
>
> My setup is optimized for traveling/field operation. I do not need  additional boxes in the form of Winkey or external QSK. I use external keying from computer through USB-COM adapter. The logger that does CW keying is TR4W. No keying compensation helps to achieve correct CW weight if TX Delay in K3 is set to say 20ms and CW speed used is above 30WPM. Dependency of CW weight from TX Delay is only inherent in K3. All other rigs do not have this dependency.
>
>

______________________________________________________________
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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Clay Autery-2
In reply to this post by Josh Fiden
ONE of MANY of my personal reasons NOT to use VOX.  <big smile>

73,

______________________
Clay Autery, KY5G
(318) 518-1389

On 08-Jun-18 14:05, Josh Fiden wrote:

> My poor impulse control makes it impossible to operate within FCC regs using vox during the inevitable lid encounters.
>
> 73
> Josh W6XU
>
> Sent from my mobile device
>
>> On Jun 8, 2018, at 10:07 AM, Bill Stravinsky via Elecraft <[hidden email]> wrote:
>>
>> Why in the 20 century we need to use foot switch?
> ______________________________________________________________
> 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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

k6dgw
In a simiilar vein, I try and let some newly composed emails ferment
overnight before pressing SEND. [:-)

73,

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

On 6/8/2018 1:19 PM, Clay Autery wrote:

> ONE of MANY of my personal reasons NOT to use VOX.  <big smile>
>
> 73,
>
> ______________________
> Clay Autery, KY5G
> (318) 518-1389
>
> On 08-Jun-18 14:05, Josh Fiden wrote:
>> My poor impulse control makes it impossible to operate within FCC
>> regs using vox during the inevitable lid encounters.
>>
>> 73
>> Josh W6XU
>>

______________________________________________________________
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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Bob Wilson, N6TV
In reply to this post by Igor Sokolov-2
On Fri, Jun 8, 2018 at 12:37 PM, Igor Sokolov <[hidden email]> wrote:

> My setup is optimized for traveling/field operation. I do not need
> additional boxes in the form of Winkey or external QSK.


Very good.  I assume you are using CONFIG:PTT-KEY set to RTS-DTR to have
T4RW provide CW and PTT over the K3 USB cable (or serial cable).


> I use external keying from computer through USB-COM adapter.


OK.  For the K3S, or K3 with KIO3B upgrade, no adapter is required (it is
built in to the K3).


> The logger that does CW keying is TR4W. No keying compensation helps to
> achieve correct CW weight if TX Delay in K3 is set to say 20ms and CW speed
> used is above 30WPM.


The TR4W manual <http://tr4w.net/TR4W_Reference_Manual_4.01.pdf> says:

"7.Support for PTT with programmable delay to ensure that antenna relays
are cold-switched."

But I could not find any further documentation showing how to set the CW
PTT delay to 20 ms in TR4W.  If the PTT line is held closed by the
software, no keying compensation is required, and CW output will be fine.
But, if you're trying to do QSK (full break-in) with VOX and external
keying from TR4W, with no PTT output, you will either have to use a
different radio, or use different software like Win-Test that (in
v4.28.0-dev) supports keying compensation, or an external keyer with PTT or
keying compensation, like a WinKey.

Perhaps you can convince the author of TR4W to add keying compensation (
*not* weight), a setting that lengthens every dot or dash by a fixed amount.


> Dependency of CW weight from TX Delay is only inherent in K3. All other
> rigs do not have this dependency.
>

Most rigs will shorten the first dot or dash (semi break-in) or every dot
or dash (QSK full break-in) by a few ms to compensate for the TX delay
setting, the delay between the amp RELAY closure, and the RF output.  Some
rigs may use a hardware "buffer" for external keying, to provide perfect
timing even with a  PTT delay, but I don't know which ones do this (Icom
perhaps).

Again, for hand sending with the internal keyer of the K3, all TX DLY
values work fine, no shortening of RF output.

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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

rich hurd WC3T
In reply to this post by Josh Fiden
"More seriously, I could see a foot switch as a VOX inhibit. Broadcasters
have a “cough button” to mute an otherwise live mic."

Can't you do that with the KX3 in VOX mode but with ACC2 IO set to
"LO=Inh?"  That's what I thought.  I'm far away from my KX3 so I can't test
it.

On Fri, Jun 8, 2018 at 18:24 Walter Underwood <[hidden email]> wrote:

> Maybe you could put a tone generator on the foot switch to bleep yourself.
> Leave VOX on, but stomp on the pedal when you encounter a lid. :-)
>
> More seriously, I could see a foot switch as a VOX inhibit. Broadcasters
> have a “cough button” to mute an otherwise live mic.
>
> wunder
> K6WRU
> Walter Underwood
> CM87wj
> http://observer.wunderwood.org/ (my blog)
>
> > On Jun 8, 2018, at 12:05 PM, Josh Fiden <[hidden email]> wrote:
> >
> > My poor impulse control makes it impossible to operate within FCC regs
> using vox during the inevitable lid encounters.
> >
> > 73
> > Josh W6XU
> >
> > Sent from my mobile device
>
> ______________________________________________________________
> 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]

--
72,
Rich Hurd / WC3T / DMR: 3142737
PA Army MARS, Northampton County RACES, EPA-ARRL Public Information Officer
for Scouting
Latitude: 40.761621 Longitude: -75.288988  (40°45.68' N 75°17.33' W) Grid:
*FN20is*
______________________________________________________________
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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Victor Rosenthal 4X6GP
In reply to this post by Bob Wilson, N6TV
My amplifier uses a reed relay and a vacuum relay for QSK. It uses the
so-called "AG6K circuit" <http://www.somis.org/QSK922.html> which
supposedly makes the usual Jennings RJ1A relay operate in 2-3 ms. I
never had any problems with the default delay of the K3, both before and
after installing the new synthesizers.

I do recall that with the old synthesizers and an external keyer,
increasing the delay above the default even a little seriously messed up
the CW timing. However now I have the new synths. So on reading Wayne's
announcement I decided to so some testing.

I don't notice ANY difference in CW timing, either with the internal
keyer or with an N1MM-operated Winkey, with QRQ mode on or off, with TX
Delay set from 8 to 20 ms., in semi or full QSK mode, up to 40 wpm.

I listened to both the sidetone and the transmitted signal. I didn't
measure anything with the scope, but I have always been sensitive to CW
timing, and it sounded the same to me.

So the upshot is that if you have a slow amplifier, you should be able
to just increase the TX DELAY as needed. And if Bob is right, turn off
QRQ mode too.

But it would be prudent to check the timing with your particular
amplifier with a scope before taking my advice!


73,
Victor, 4X6GP
Rehovot, Israel
Formerly K2VCO
CWops no. 5
http://www.qsl.net/k2vco/
On 8 Jun 2018 21:12, Bob Wilson, N6TV wrote:
<snip>

> If you are using the internal keyer of the K3 for hand sending,
> compensation is automatic.  No need for any adjustment; use any TX DLY
> value you need, and the transmitted dot/dash ratios will still be perfect
> (I've measured it on a scope).
>
> If using external keying with a WinKey, use the menu setting for "Keying
> Compensation" (QSK), or "1st dit extension" (Semi-Breakin).  to apply
> matching keying compensation.  Example:
>
> CONFIG: TX DLY 010
> Actual measured TX Delay  7-8 ms (with jitter)
> Set External Keying Compensation (for QSK) to:  7 ms
> Set 1st Dit Extension (semi-breakin) to:  7 ms
>
>
> If using a microHAM Device with a built-in WinKey, the menu settings for
> these values were (unfortunately) moved from the WinKey tab of the Router
> software.  Instead, they are now a single setting under the PTT tab, called
> "R/T Delay" (improperly documented).  So, in the microHAM Router, PTT tab:
>
> Set R/T Delay to 7 ms
>
>
> If using a LogiKey K1 or similar:
>
> Set the "K Factor" (keying compensatin) to 7
>
>
> If using Win-Test with CONFIG:PTT-KEY OFF-DTR for QSK keying over the
> serial port DTR pin, or an external keying circuit, set the Keying
> Compensation in the Interface Configuration menu to 7 ms (this a new
> feature in Win-Test 4.28.0-dev).  If using semi-breakin, set CONFIG:PTT-KEY
> to RTS-DTR and tell Win-Test to use a PTT Lead time of 10-20 ms, 0 ms
> keying compensation.
>
> Most WinKey/N1MM users can simply rely on setting the WinKey PTT Lead time
> to a sufficient value.  Unfortunately, the WinKey PTT lead times are set in
> increments of 10 ms (0, 10, 20, 30 ms), despite the menu settings.  If you
> ask for 11 ms of PTT delay in a WinKey, you actually get 20 ms of PTT delay.
>
> Note that the K3 will safely ignore any PTT lead time less than TX DLY, so
> if you set PTT LEAD to 3 ms you will actually get 7 ms and all elements
> (QSK) or the first element (semi break-in) will be shortened by 7 ms.
>
> FYI, the OM Power OM2000A+ non-QSK amplifier manual recommends a TX DLY of
> 20 ms (!).
>
> Final note:  if using CW QRQ mode, the K3 ignores TX DLY completely, and
> you get 4-5 ms of TX DELAY (fixed, with jitter).  Set keying compensation
> to 4 ms.
>
> 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]
>
______________________________________________________________
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]
Reply | Threaded
Open this post in threaded view
|

Re: Default K3 transmit delay may be too short for slow QRO amplifiers

Bill W4ZV
In reply to this post by Dimitry Borzenko
Dimitry Borzenko wrote
> Why in the 20 century we need to use foot switch?

Because I switch other things besides my amp.  For example:

Beverages disconnected from RX input during TX.
Parasitic TX array elements (active during RX, to ground during RX).
Detuning stub on TX antenna (during RX).

Most of these relays are slow (and cheap) open frame relays.

With a little practice, I can listen between words (not characters) using a
footswitch.

And I have NEVER burned up any TX relays in my amps.

73,  Bill  W4ZV




--
Sent from: http://elecraft.365791.n2.nabble.com/
______________________________________________________________
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]