[Elecraft] K3 TX delay

Fri Feb 17 10:43:34 EST 2012

Don,
What you are saying translates into " Do not talk into the mike or touch 
keyer for 30 ms or more after you press PTT footswitch". I know that. But 
this method can hardly be called fool proof for contesting. In fact, even 8 
ms delay woul not be needed.

73, Igor UA9CDC

> Igor,
>
> I understand your situation, but the application of PTT does not produce 
> any RF - nothing will hot switch.
> When you "hit the key" (or for SSB begin talking), then RF is produced.
> If you arrange your switching to be accomplished by either the PTT or the 
> K3 KEYOUT, and leave the K3 in non-VOX state, there is no need to increase 
> TX DLY to any other than the 8 ms default value and therefore no 
> shortening of CW elements.
>
> 73,
> Don W3FPR
>
> On 2/17/2012 6:52 AM, Igor Sokolov wrote:
>> Don,
>> I will try to explain.
>> Forget about QSK all together.
>> Apart from linear amp, there are other external relays like stack control
>> relays, T/R relays when you want to use different antennas for TX/RX etc.
>> PTT signal can either be controlled by the logger (CW, VOICE Keyer, DIGI
>> modes) or by PTT footswitch. In both cases PTT signal goes into PTT IN of
>> K3.
>> About 6ms later you get sort of PTT OUT on the KEY OUT connector. This 
>> KEY
>> OUT signal controls linear, T/R antenna relays, stack relays etc. Then 
>> 8ms
>> later you get RF out of the K3. These 8ms is not enough for all the 
>> external
>> relays to settle.
>> PTT delay of 20 ms should be more or less OK (although I would not mind 
>> it
>> to be extended to 30ms) but then CW generated by  a logger gets choppy.
>> I can overcome the problem of non adjustable TX delay of K3 by 
>> configuring
>> my logger to start sending say 30 ms after PTT was asserted. I can even
>> connect PTT footswitch to the computer, but that will only work for CW
>> sending from the keyboard, computer voice keyer and digi signals 
>> generated
>> by the computer. If I send from the K3 internal keyer or use k3 mike I am 
>> in
>> danger to hot switch some of the external relays.
>> Therefore instead of being able to extend TX delay inside K3 I will 
>> probably
>> have to build some sort of circit which will take low level from KEY OUT 
>> and
>> form some sort of positive or negative impulse of a required length (say 
>> 30
>> ms) to apply to INH input of K3. I was hoping to build it inside the 
>> DB-15
>> cabinet since KEY OUT and INH pins are available on the ACC connector. 
>> But
>> there is no any 12 v or 5v pins available on that connector. So it makes
>> life more complicated.
>> If you can suggest different approach. which does not involve building
>> external sequencer, I would really appreciate it.
>>
>> 73, Igor UA9CDC
>>
>>
>>> I fail to understand why anyone who is using PTT switching would want to
>>> set the TX DLY parameter to  greater than the default of 8 ms.  Use PTT 
>>> and
>>> a footswitch if you are using a slow amp relay.
>>>
>>> If your amplifier uses a slow relay, then you really cannot use QSK, so
>>> just use PTT and a footpedal instead - things will work out much better.
>>> There are QSK mods available for amplifiers that use slow relays.  Some
>>> mods are relatively inexpensive (the owner does the work), while other 
>>> can
>>> cost upwards of $400 (purchase a commercial version).
>>>
>>> 73,
>>> Don W3FPR
>>>
>>> On 2/16/2012 11:05 PM, Igor Sokolov wrote:
>>>>> But I am finishing my 300w HF amplifier kit and will be using a dc
>>>>> open-frame relay for TR switching of RF.  I guess I can time it using
>>>>> my dual trace scope and set K3 TX DLY accordingly (amp will be keyed
>>>>> with KEYOUT).  I do not use CW very much on HF so QSK with 300w is
>>>>> not necessary (or probably possible).
>>>> Even if you use PTT CW and set TX delay for 20 ms you will get choppy 
>>>> CW
>>>>
>>>>> There is one thing I do not see being addressed in this
>>>>> discussion.  Maybe more important than clipping of CW characters is
>>>>> avoiding QRO "Hot Switching" of RF.  Calculate the RF voltage for
>>>>> several thousand ohms impedance (open circuit) that will result if
>>>>> the amp is pumping out "globs" of RF before being connected to a good
>>>>> load.  The relay contacts will definitely get burned big time.  And
>>>>> maybe your output devices will have a short live-span.
>>>>>
>>>>>
>>>>> 73, Ed - KL7UW, WD2XSH/45
>>>> The entire topic stems from the fact that I wanted to avoid hot 
>>>> switching
>>>> of
>>>> the external relays (linear amp, stack switching etc). To acheave that 
>>>> I
>>>> have tried to increase TX delay and got choppy CW. So without external
>>>> sequencer there is a choise between the two - 'hot switching' or 
>>>> 'choppy
>>>> cw'.
>>>>
>>>> 73. Igor UA9CDC
>>>>
>>>>
>>>>> ======================================
>>>>> BP40IQ   500 KHz - 10-GHz   www.kl7uw.com
>>>>> EME: 50-1.1kw?, 144-1.4kw, 432-QRT, 1296-?, 3400-?
>>>>> DUBUS Magazine USA Rep dubususa at gmail.com
>>>>> "Kits made by KL7UW" http://www.kl7uw.com/kits.htm
>>>>> ======================================
>>>>> ______________________________________________________________
>>>>> Elecraft mailing list
>>>>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>>>>> Help: http://mailman.qth.net/mmfaq.htm
>>>>> Post: mailto:Elecraft at mailman.qth.net
>>>>>
>>>>> This list hosted by: http://www.qsl.net
>>>>> Please help support this email list: http://www.qsl.net/donate.html
>>>> ______________________________________________________________
>>>> Elecraft mailing list
>>>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>>>> Help: http://mailman.qth.net/mmfaq.htm
>>>> Post: mailto:Elecraft at mailman.qth.net
>>>>
>>>> This list hosted by: http://www.qsl.net
>>>> Please help support this email list: http://www.qsl.net/donate.html
>>>>
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:Elecraft at mailman.qth.net
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> 