KXAT2 vs. Other Elecraft Transceiver ATUs

[Mike Morrow-3]

Bill wrote:

> As to the ATU in the KX2, 7L/7C is the same range as the KX1, T1 and
> I think the K2 ATU.

That is wildly inaccurate, especially for the KXAT1 ATU!!!

The reactance configuration for all Elecraft ATUs is an L-network of series inductance, with a capacitance connected to common before (radio side) or after (antenna side) the inductance (but not both) by a relay.  It is a classic simple low pass filter.

The total number of possible configurations in such a filter of L inductors and C capacitors, with the capacitor bank capable of being connected to either side of the series inductor, is:

Different Configurations = ( 2^L x 2^C x 2 )

Four (4) "don't care" (non-unique) configurations should be eliminated from total:
1.  Capacitor bank connected to inductor input, but all capacitors open.
2.  Capacitor bank connected to inductor output, but all capacitors open.
3.  Capacitor bank connected to inductor input, but all inductors shorted.
4.  Capacitor bank connected to inductor output, but all inductors shorted.

USEFUL Different Configurations = ( 2^L x 2^C x 2 ) - 4

For Elecraft transceiver ATUs before the new KXAT2, here are summaries (N is the network variable...capacitor connected to input OR output of inductor):

KAT1 (4L/5C/N) Configs=1020, L=0 to 4.9 uH (16 steps), C=0 to 300 pF (32 steps)

KAT2 (8L/8C/N) Configs=131068, L=0 to 20.7 uH (256 steps), C=0 to 2420 pF (256 steps)

KAT3 (8L/8C/N) Configs=131068, L=0 to 17.4 uH (256 steps), C=0 to 2600 pF (256 steps)

KXAT1 (3L/3C/N) Configs=124, L=0 to 4.5 uH (8 steps), C=0 to 140 pF (8 steps)

KXAT3 (8L/8C/N) Configs=131068, L=0 to 15.9 uH (256 steps), C=0 to 2680 pF (256 steps)

T1 (7L/7C/N) Configs=32764, L=0 to 7.5 uH (128 steps), C=0 to 1300 pF (128 steps)
 
Many already have the new KXAT2, but no circuit value information has yet been provided to those of the great unwashed masses like myself.  All I've seen is its description as 7L/7C/N, just like the T1.  If that is correct:

KXAT2 (7L/7C/N) Configs=32764, L=0 to ?.?? uH (128 steps), C=0 to ???? pF (128 steps)

The KXAT1 generates 124 unique matching network configurations.
The KAT1 generates 1020 unique matching network configurations.
The KXAT2, T1 generate 32724 unique matching network configurations.
The KAT2, KAT3, KXAT3 generate 131068 unique matching network configurations.

Conclusion:  The KXAT2 is comparable ONLY to the T1.

73,
Mike / KK5F
______________________________________________________________
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]

Mike,

Your analysis of combinations is correct. But the KXAT2 should be compared to the KXAT3 (etc.), not the T1. Here's why:

The KXAT2 only has to cover 80-10 meters. The KXAT3 has one additional L and C, but it has to cover 160-6 meters. Over the range of 80-10 meters, for virtually any field antenna, the KXAT2 will perform just as well.

The T1 also has to cover 160-6 meters. To do this with 7L/7C, the L and C values in its array are spread out with a ratio of slightly greater then 2:1 from one to the next. This allows the array to cover a wider range of load impedances, but with an average final SWR more like 1.5:1 than 1:1.

The KXAT2, like the KXAT3, uses a ratio of very close to to 2:1, so average final SWR is very close to 1:1.

73,
Wayne
N6KR



On May 24, 2016, at 9:05 AM, Mike Morrow <[hidden email]> wrote:

> Bill wrote:
>
>> As to the ATU in the KX2, 7L/7C is the same range as the KX1, T1 and
>> I think the K2 ATU.
>
> That is wildly inaccurate, especially for the KXAT1 ATU!!!
>
> The reactance configuration for all Elecraft ATUs is an L-network of series inductance, with a capacitance connected to common before (radio side) or after (antenna side) the inductance (but not both) by a relay.  It is a classic simple low pass filter.
>
> The total number of possible configurations in such a filter of L inductors and C capacitors, with the capacitor bank capable of being connected to either side of the series inductor, is:
>
> Different Configurations = ( 2^L x 2^C x 2 )
>
> Four (4) "don't care" (non-unique) configurations should be eliminated from total:
> 1.  Capacitor bank connected to inductor input, but all capacitors open.
> 2.  Capacitor bank connected to inductor output, but all capacitors open.
> 3.  Capacitor bank connected to inductor input, but all inductors shorted.
> 4.  Capacitor bank connected to inductor output, but all inductors shorted.
>
> USEFUL Different Configurations = ( 2^L x 2^C x 2 ) - 4
>
> For Elecraft transceiver ATUs before the new KXAT2, here are summaries (N is the network variable...capacitor connected to input OR output of inductor):
>
> KAT1 (4L/5C/N) Configs=1020, L=0 to 4.9 uH (16 steps), C=0 to 300 pF (32 steps)
>
> KAT2 (8L/8C/N) Configs=131068, L=0 to 20.7 uH (256 steps), C=0 to 2420 pF (256 steps)
>
> KAT3 (8L/8C/N) Configs=131068, L=0 to 17.4 uH (256 steps), C=0 to 2600 pF (256 steps)
>
> KXAT1 (3L/3C/N) Configs=124, L=0 to 4.5 uH (8 steps), C=0 to 140 pF (8 steps)
>
> KXAT3 (8L/8C/N) Configs=131068, L=0 to 15.9 uH (256 steps), C=0 to 2680 pF (256 steps)
>
> T1 (7L/7C/N) Configs=32764, L=0 to 7.5 uH (128 steps), C=0 to 1300 pF (128 steps)
>
> Many already have the new KXAT2, but no circuit value information has yet been provided to those of the great unwashed masses like myself.  All I've seen is its description as 7L/7C/N, just like the T1.  If that is correct:
>
> KXAT2 (7L/7C/N) Configs=32764, L=0 to ?.?? uH (128 steps), C=0 to ???? pF (128 steps)
>
> The KXAT1 generates 124 unique matching network configurations.
> The KAT1 generates 1020 unique matching network configurations.
> The KXAT2, T1 generate 32724 unique matching network configurations.
> The KAT2, KAT3, KXAT3 generate 131068 unique matching network configurations.
>
> Conclusion:  The KXAT2 is comparable ONLY to the T1.
>
> 73,
> Mike / KK5F
> ______________________________________________________________
> 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]

[Mike Morrow-3]

Thanks for your reply, Wayne.  I should have noted at least one of the important points you make in my conclusion about KXAT2 vs. T1, since the reduced required frequency span is pointed out in other KX2 materials.  I just did not want to get much more long-winded than I do naturally.

Thanks for making the KXAT2 instructions available at:

http://www.elecraft.com/manual/E740294%20KXAT2%20Installation%20Instructions%20Rev%20A.pdf

I wrote earlier:

> T1 (7L/7C/N) Configs=32764, L=0 to 7.5 uH (128 steps), C=0 to 1300 pF (128 steps)

> KXAT2 (7L/7C/N) Configs=32764, L=0 to ?.?? uH (128 steps), C=0 to ???? pF (128 steps)

There is no KXAT2 schematic nor component values in its user instructions.  Those would show the hard evidence of KXAT2 design variance with the T1.  Obviously the ATU's total and incremental inductance or capacitance values spread through 128 steps of each in the KXAT2 will be smaller than those for the T1.

Will there be a system schematic similar to the KX3's available for download in the near future?  The beauty is not only in the unit itself, but also in what lies within that small case.

The size of the KX2 makes it just beg to be taken along on ALL wilderness day (or longer) hikes, without any exception.  All I'm waiting for is the initial mad mad sales rush to dissipate.  A fantastic achievement...thanks much!

73,
Mike / KK5F

-----Original Message-----

>From: Wayne Burdick <[hidden email]>
>Sent: May 24, 2016 11:33 AM
>To: Mike Morrow <[hidden email]>
>Cc: [hidden email]
>Subject: Re: [Elecraft] KXAT2 vs. Other Elecraft Transceiver ATUs
>
> Mike,
>
> Your analysis of combinations is correct. But the KXAT2 should be compared to the
> KXAT3 (etc.), not the T1. Here's why:
>
> The KXAT2 only has to cover 80-10 meters. The KXAT3 has one additional L and C, but
> it has to cover 160-6 meters. Over the range of 80-10 meters, for virtually any field
> antenna, the KXAT2 will perform just as well.
>
> The T1 also has to cover 160-6 meters. To do this with 7L/7C, the L and C values in
> its array are spread out with a ratio of slightly greater then 2:1 from one to the
> next. This allows the array to cover a wider range of load impedances, but with an
> average final SWR more like 1.5:1 than 1:1.
>
> The KXAT2, like the KXAT3, uses a ratio of very close to to 2:1, so average final SWR
> is very close to 1:1.
>
> 73,
> Wayne
> N6KR
>
> On May 24, 2016, at 9:05 AM, Mike Morrow <[hidden email]> wrote:
>
>> Bill wrote:
>>
>>> As to the ATU in the KX2, 7L/7C is the same range as the KX1, T1 and
>>> I think the K2 ATU.
>>
>> That is wildly inaccurate, especially for the KXAT1 ATU!!!
>>
>> The reactance configuration for all Elecraft ATUs is an L-network of series inductance,
>> with a capacitance connected to common before (radio side) or after (antenna side) the
>> inductance (but not both) by a relay.  It is a classic simple low pass filter.
>>
>> The total number of possible configurations in such a filter of L inductors and C
>> capacitors, with the capacitor bank capable of being connected to either side of
>> the series inductor, is:
>>
>> Different Configurations = ( 2^L x 2^C x 2 )
>>
>> Four (4) "don't care" (non-unique) configurations should be eliminated from total:
>> 1.  Capacitor bank connected to inductor input, but all capacitors open.
>> 2.  Capacitor bank connected to inductor output, but all capacitors open.
>> 3.  Capacitor bank connected to inductor input, but all inductors shorted.
>> 4.  Capacitor bank connected to inductor output, but all inductors shorted.
>>
>> USEFUL Different Configurations = ( 2^L x 2^C x 2 ) - 4
>>
>> For Elecraft transceiver ATUs before the new KXAT2, here are summaries (N is the
>> network variable...capacitor connected to input OR output of inductor):
>>
>> KAT1 (4L/5C/N) Configs=1020, L=0 to 4.9 uH (16 steps), C=0 to 300 pF (32 steps)
>>
>> KAT2 (8L/8C/N) Configs=131068, L=0 to 20.7 uH (256 steps), C=0 to 2420 pF (256 steps)
>>
>> KAT3 (8L/8C/N) Configs=131068, L=0 to 17.4 uH (256 steps), C=0 to 2600 pF (256 steps)
>>
>> KXAT1 (3L/3C/N) Configs=124, L=0 to 4.5 uH (8 steps), C=0 to 140 pF (8 steps)
>>
>> KXAT3 (8L/8C/N) Configs=131068, L=0 to 15.9 uH (256 steps), C=0 to 2680 pF (256 steps)
>>
>> T1 (7L/7C/N) Configs=32764, L=0 to 7.5 uH (128 steps), C=0 to 1300 pF (128 steps)
>>
>> Many already have the new KXAT2, but no circuit value information has yet been provided
>> to those of the great unwashed masses like myself.  All I've seen is its description as
>> 7L/7C/N, just like the T1.  If that is correct:
>>
>> KXAT2 (7L/7C/N) Configs=32764, L=0 to ?.?? uH (128 steps), C=0 to ???? pF (128 steps)
>>
>> The KXAT1 generates 124 unique matching network configurations.
>> The KAT1 generates 1020 unique matching network configurations.
>> The KXAT2, T1 generate 32724 unique matching network configurations.
>> The KAT2, KAT3, KXAT3 generate 131068 unique matching network configurations.
>>
>> Conclusion:  The KXAT2 is comparable ONLY to the T1.
>>
>> 73,
>> Mike / KK5F

______________________________________________________________
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]

[Phil Wheeler-2]

Interesting analysis, Mike -- though relating the
various *AT* acronyms to rigs is a bit mind
bending :-)

But another point, an important one, is that the
KX2 has fewer band (no 160 and no 6) compared to
some of the other rigs, so needs an ATU with less
L and C range. So it's capabilities for the bands
it covers is likely just as good as that of, say,
the KX3.

73, Phil W7OX

On 5/24/16 9:05 AM, Mike Morrow wrote:

> Bill wrote:
>
>> As to the ATU in the KX2, 7L/7C is the same range as the KX1, T1 and
>> I think the K2 ATU.
> That is wildly inaccurate, especially for the KXAT1 ATU!!!
>
> The reactance configuration for all Elecraft ATUs is an L-network of series inductance, with a capacitance connected to common before (radio side) or after (antenna side) the inductance (but not both) by a relay.  It is a classic simple low pass filter.
>
> The total number of possible configurations in such a filter of L inductors and C capacitors, with the capacitor bank capable of being connected to either side of the series inductor, is:
>
> Different Configurations = ( 2^L x 2^C x 2 )
>
> Four (4) "don't care" (non-unique) configurations should be eliminated from total:
> 1.  Capacitor bank connected to inductor input, but all capacitors open.
> 2.  Capacitor bank connected to inductor output, but all capacitors open.
> 3.  Capacitor bank connected to inductor input, but all inductors shorted.
> 4.  Capacitor bank connected to inductor output, but all inductors shorted.
>
> USEFUL Different Configurations = ( 2^L x 2^C x 2 ) - 4
>
> For Elecraft transceiver ATUs before the new KXAT2, here are summaries (N is the network variable...capacitor connected to input OR output of inductor):
>
> KAT1 (4L/5C/N) Configs=1020, L=0 to 4.9 uH (16 steps), C=0 to 300 pF (32 steps)
>
> KAT2 (8L/8C/N) Configs=131068, L=0 to 20.7 uH (256 steps), C=0 to 2420 pF (256 steps)
>
> KAT3 (8L/8C/N) Configs=131068, L=0 to 17.4 uH (256 steps), C=0 to 2600 pF (256 steps)
>
> KXAT1 (3L/3C/N) Configs=124, L=0 to 4.5 uH (8 steps), C=0 to 140 pF (8 steps)
>
> KXAT3 (8L/8C/N) Configs=131068, L=0 to 15.9 uH (256 steps), C=0 to 2680 pF (256 steps)
>
> T1 (7L/7C/N) Configs=32764, L=0 to 7.5 uH (128 steps), C=0 to 1300 pF (128 steps)
>  
> Many already have the new KXAT2, but no circuit value information has yet been provided to those of the great unwashed masses like myself.  All I've seen is its description as 7L/7C/N, just like the T1.  If that is correct:
>
> KXAT2 (7L/7C/N) Configs=32764, L=0 to ?.?? uH (128 steps), C=0 to ???? pF (128 steps)
>
> The KXAT1 generates 124 unique matching network configurations.
> The KAT1 generates 1020 unique matching network configurations.
> The KXAT2, T1 generate 32724 unique matching network configurations.
> The KAT2, KAT3, KXAT3 generate 131068 unique matching network configurations.
>
> Conclusion:  The KXAT2 is comparable ONLY to the T1.
>
> 73,
> Mike / KK5F

______________________________________________________________
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]