f/w 2.73/1.98 power calibration

[Steve Ward]

I just applied the latest beta firmware as I have done many times
before, and it looks like this time my power calibration went out the
window.  I'm putting out about 70w on PSK31 when I dialed in 45.

Just wanted to give folks a heads-up to watch their power output when
they start operating under new firmware, it might be time to re-calibrate!

73,
Steve AD7OG
K3 #1544
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

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

[Richard Ferch]

Steve Ward said:

> I just applied the latest beta firmware as I have done many times
> before, and it looks like this time my power calibration went out the
> window.  I'm putting out about 70w on PSK31 when I dialed in 45.
>
> Just wanted to give folks a heads-up to watch their power output when
> they start operating under new firmware, it might be time to re-calibrate!

It's more than just the power calibration. Even after calibrating power and
setting the requested power below 50W, with either 2.73/1.98 or 2.72/1.98
the power output on PSK31 (in DATA A) mode often exceeds the requested
power, which results in splatter. I have been having to constantly ride the
LINE IN and/or sound card output level controls in this weekend's PSK
contest to keep the power below 50W and PSK signal quality acceptable; I
have now switched back to 2.67/1.96 for the remainder of the contest.

73,
Rich VE3KI
K3 #1595

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

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

[Bob Cunnings NW8L]

Have you tried adjusting TXG VCE? With latest firmware, a setting of
-2dB puts my PSK31 power right where it belongs for a given requested
power out. There's then no need to ride the audio input level, which
is a bad idea given the way the K3 ALC works.

Bob NW8L

On Sun, Dec 14, 2008 at 11:44 AM, Richard Ferch <[hidden email]> wrote:

> Steve Ward said:
>
>> I just applied the latest beta firmware as I have done many times
>> before, and it looks like this time my power calibration went out the
>> window.  I'm putting out about 70w on PSK31 when I dialed in 45.
>>
>> Just wanted to give folks a heads-up to watch their power output when
>> they start operating under new firmware, it might be time to re-calibrate!
>
> It's more than just the power calibration. Even after calibrating power and
> setting the requested power below 50W, with either 2.73/1.98 or 2.72/1.98
> the power output on PSK31 (in DATA A) mode often exceeds the requested
> power, which results in splatter. I have been having to constantly ride the
> LINE IN and/or sound card output level controls in this weekend's PSK
> contest to keep the power below 50W and PSK signal quality acceptable; I
> have now switched back to 2.67/1.96 for the remainder of the contest.
>
> 73,
> Rich VE3KI
> K3 #1595
>
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
>  http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com
>

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

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

[Greg - AB7R]

With DataA you need to adjust the LineIn gain so that you have 4 bars  
of ALC action.

73
Greg

On Dec 14, 2008, at 6:06 PM, Bob Cunnings wrote:

> Have you tried adjusting TXG VCE? With latest firmware, a setting of
> -2dB puts my PSK31 power right where it belongs for a given requested
> power out. There's then no need to ride the audio input level, which
> is a bad idea given the way the K3 ALC works.
>
> Bob NW8L
>
> On Sun, Dec 14, 2008 at 11:44 AM, Richard Ferch <[hidden email]>  
> wrote:
>> Steve Ward said:
>>
>>> I just applied the latest beta firmware as I have done many times
>>> before, and it looks like this time my power calibration went out  
>>> the
>>> window.  I'm putting out about 70w on PSK31 when I dialed in 45.
>>>
>>> Just wanted to give folks a heads-up to watch their power output  
>>> when
>>> they start operating under new firmware, it might be time to re-
>>> calibrate!
>>
>> It's more than just the power calibration. Even after calibrating  
>> power and
>> setting the requested power below 50W, with either 2.73/1.98 or  
>> 2.72/1.98
>> the power output on PSK31 (in DATA A) mode often exceeds the  
>> requested
>> power, which results in splatter. I have been having to constantly  
>> ride the
>> LINE IN and/or sound card output level controls in this weekend's PSK
>> contest to keep the power below 50W and PSK signal quality  
>> acceptable; I
>> have now switched back to 2.67/1.96 for the remainder of the contest.
>>
>> 73,
>> Rich VE3KI
>> K3 #1595
>>
>> _______________________________________________
>> Elecraft mailing list
>> Post to: [hidden email]
>> You must be a subscriber to post to the list.
>> Subscriber Info (Addr. Change, sub, unsub etc.):
>> http://mailman.qth.net/mailman/listinfo/elecraft
>>
>> Help: http://mailman.qth.net/subscribers.htm
>> Elecraft web page: http://www.elecraft.com
>>
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com

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

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

[Steve Ward]

Richard Ferch wrote:

Steve Ward said:


> > I just applied the latest beta firmware as I have done many times
> > before, and it looks like this time my power calibration went out the
> > window.  I'm putting out about 70w on PSK31 when I dialed in 45.
> >
> > Just wanted to give folks a heads-up to watch their power output when
> > they start operating under new firmware, it might be time to re-calibrate!
>  

> It's more than just the power calibration. Even after calibrating power and
> setting the requested power below 50W, with either 2.73/1.98 or 2.72/1.98
> the power output on PSK31 (in DATA A) mode often exceeds the requested
> power, which results in splatter. I have been having to constantly ride the
> LINE IN and/or sound card output level controls in this weekend's PSK
> contest to keep the power below 50W and PSK signal quality acceptable; I
> have now switched back to 2.67/1.96 for the remainder of the contest.


Agreed that calibration is apparently not the issue.  I did some testing
with a dummy load.  Sending a PSK signal (just "TX" mode no data) I see
9 bars on the RF Power display on the K3.  The MFJ dummy load's meter
(FWIW) shows approx 50w average output but much higher peak reading.  I
have no idea how accurate the meter on the MFJ is, but something is
definitely changed with this revision.

73,
Steve
AD7OG
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

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

[Julian, G4ILO]

Richard Ferch wrote

It's more than just the power calibration. Even after calibrating power and
setting the requested power below 50W, with either 2.73/1.98 or 2.72/1.98
the power output on PSK31 (in DATA A) mode often exceeds the requested
power, which results in splatter. I have been having to constantly ride the
LINE IN and/or sound card output level controls in this weekend's PSK
contest to keep the power below 50W and PSK signal quality acceptable; I
have now switched back to 2.67/1.96 for the remainder of the contest.

I don't think it is anything to do with this revision. I first noticed this a few months ago, after one of the updates that made some changes to the ALC. Adjusting TXG VCE may be the answer, but the other change I made was to avoid tuning at 5.0 or 50W. The K3 recalibrates the power when you press TUNE at those exact power levels, and if it is looking into a mismatched (or varying, as you adjust the ATU) load then it mucks up the calibration.

[Richard Ferch]

Julian, G4ILO said:

> I don't think it is anything to do with this revision. I first noticed
this
> a few months ago, after one of the updates that made some changes to the
> ALC. Adjusting TXG VCE may be the answer, but the other change I made was
to
> avoid tuning at 5.0 or 50W. The K3 recalibrates the power when you press
> TUNE at those exact power levels, and if it is looking into a mismatched
(or
> varying, as you adjust the ATU) load then it mucks up the calibration.

Except when performing a power calibration, I never set my requested power
to 5.0W or 50W for the reason Julian describes.

Actually, this is not a power calibration issue at all, but it is a
difference in the radio's performance as a result of the firmware revision.
Even after performing the power calibration at 50W into a dummy load to
eliminate calibration as a concern, I find that version 2.73/1.98 delivers
significantly higher power than 2.67/1.96 at the same PWR control setting.
As AD7OG did, I observe that with a PWR control setting of 48 watts and TXG
VCE at 0.0 dB I get a reading of 90 watts (9 bars) delivered power at an
audio frequency of 1000 Hz in DATA A mode. The IMD in PSK31 is
correspondingly bad (at 90W delivered power, the measured IMD is around -18
dB instead of the desired -24 dB or better). The delivered power is not
quite this high at higher audio frequencies, but still well above the
requested 48 watts.

AFSK A mode behaves similarly to DATA A, but unlike PSK31 the consequences
in RTTY are relatively minor (no IMD issues to worry about - if I were using
low tones, I would hold the delivered power to 100 watts or less simply to
avoid the possibility of transmitting audio harmonics, but with high tones
this is not likely to be a problem).

In order to hold the delivered power down to no more than 50 watts so as to
give acceptable IMD values for PSK31, I must either adjust the PWR control
down to a setting much lower than the power I actually want, or set TXG VCE
to a value of -2.0 dB or so as suggested by NW8L. The precise power
delivered varies somewhat with audio frequency, but settings of 48W for the
PWR control and -2.0 dB for TXG VCE seem to deliver reasonable PSK31
performance here. I expect that these settings may continue to change with
future firmware revisions.

73,
Rich VE3KI
K3 #1595

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

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

[Steve Ward]

After reading about the TXG VCE parameter I reloaded MCU 2.73/DSP 1.98
and attempted to adjust the gain back to a reasonable level.  By setting
TXG VCE to -3.0 dB I came close to my requested power output on the K3's
meter, when set to 40w I get between 4-5 bars on the graph.  This is
with the 5th bar flickering slightly on the ALC meter.

At the same time I also tried the ADC REF calibration and came up with
4.967 v on my Fluke meter, so I set the K3 to 4.97.

It seems to me that having to set the TXG VCE to it's minimum setting
suggests something is still not quite right with the new firmware...

73,
Steve
K3/100 #1544
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

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