KX3 Operating Tip: Determining the ATU's L/C values and L-net configuration

[wayne burdick]

Heinz (OE5EEP) wrote:
> The KXAT3 internal ATU works really great! I can match a dipole of  
> 2x 6.8m with 7m of double-wire feedline erected as inverted V on all  
> bands from 40 to 6m....  I wonder if there is a way to retrieve the  
> L/C combinations and values used to achieve a match. Is this  
> somewhere hidden in the menu?
>

You can't get L-network information via the menu. But you can send an  
"AK;" (ATU relay settings) command from the KX3 Utility Command  
Tester. This command is intended for use during automated factory test  
but may be of interest to operators as well.

The response format is “AKaabbcc;”, where:

   aa = inductance bitmap in ASCII hex format
   bb = capacitance bitmap in ASCII hex format
   cc = misc relays bitmap in ASCII hex format

At present only bit 0 of byte cc is defined:

   00 = capacitance on antenna side
   01 = capacitance on transmitter side

Example: Suppose you get back "AK031A01;". First convert the hex  
quantities to binary:

   03 => 00000011 (L)
   1A => 00011010 (C)
   01 => 00000001 (network selection)

Now compare these values to the KXAT3 schematic. 00000011 for L means  
that two smallest inductors are switched in. Add these together to  
determine the total L value. Similarly, add up the selected capacitors  
based on the binary value 00011010. Finally, you can determine the  
network configuration from the value 00000001. In this case, the  
capacitance is on the transmitter side of the series inductance.

If the ATU is not installed, or if the ATU MD menu entry parameter is  
set to one of the L/C test values, “AK000000;” is returned.

If the ATU is in “bypass” mode, quasi-bypass values are shown. (These  
were selected to approximately cancel out the ATU's own stray  
reactance on each band.)

In AUTO mode, the working auto-tuned values are shown.

Wayne



______________________________________________________________
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

[WB4SON]

Wayne,

With the magic decoder ring, that is easy to understand -- thanks!

But, I can't help but think that having the KX3 Utility Program decode that
and display the values might be pretty neat.  Maybe someday in the far
distant future when there isn't anything better to do.

73, Bob, WB4SON

"You can't get L-network information via the menu. But you can send an
"AK;" (ATU relay settings) command from the KX3 Utility Command Tester.
This command is intended for use during automated factory test but may be
of interest to operators as well."
______________________________________________________________
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

[wayne burdick]

I wrote:

> You can't get L-network information via the menu. But you can send an
> "AK;" (ATU relay settings) command from the KX3 Utility Command
> Tester....compare these values to the KXAT3 schematic.

We're still working on getting the schematics formatted for  
publication. Meanwhile, here are the approximate L and C values:

  L:   8, 4, 2, 1, 0.5, 0.25, 0.12, and 0.06 uH

  C:   1360, 680, 330, 164, 82, 39, 18, 10 pF

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

[Pierfrancesco Caci]

>>>>> "Wayne" == Wayne Burdick <[hidden email]> writes:


    Wayne> You can't get L-network information via the menu. But you can send an  
    Wayne> "AK;" (ATU relay settings) command from the KX3 Utility Command  
    Wayne> Tester. This command is intended for use during automated
    Wayne> factory test  
    Wayne> but may be of interest to operators as well.

Will you please make this available on the K3 as well, at some point?

Pf


--
Pierfrancesco Caci, ik5pvx
______________________________________________________________
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

[wayne burdick]

Noted for a future release.

Thanks,
Wayne
N6KR

On Aug 2, 2012, at 11:39 AM, Pierfrancesco Caci wrote:

>>>>>> "Wayne" == Wayne Burdick <[hidden email]> writes:
>
>
>    Wayne> You can't get L-network information via the menu. But you  
> can send an
>    Wayne> "AK;" (ATU relay settings) command from the KX3 Utility  
> Command
>    Wayne> Tester. This command is intended for use during automated
>    Wayne> factory test
>    Wayne> but may be of interest to operators as well.
>
> Will you please make this available on the K3 as well, at some point?
>
> Pf
>
>
> --
> Pierfrancesco Caci, ik5pvx

______________________________________________________________
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

[Mike Morrow-3]

Wayne wrote:

> You can't get L-network information via the menu. But you can send  
> an "AK;" (ATU relay settings) command from the KX3 Utility Command  
> Tester.
>
> The response format is "AKaabbcc;", where:
>
>    aa = inductance bitmap in ASCII hex format
>    bb = capacitance bitmap in ASCII hex format
>    cc = misc relays bitmap in ASCII hex format
>
> At present only bit 0 of byte cc is defined:
>
>    00 = capacitance on antenna side
>    01 = capacitance on transmitter side
>
> Example: Suppose you get back "AK031A01;". First convert the hex  
> quantities to binary:
>
>    03 => 00000011 (L)
>    1A => 00011010 (C)
>    01 => 00000001 (network selection)
>
> Now compare these values to the KXAT3 schematic. 00000011 for L means  
> that two smallest inductors are switched in. Add these together to  
> determine the total L value. Similarly, add up the selected capacitors  
> based on the binary value 00011010. Finally, you can determine the  
> network configuration from the value 00000001. In this case, the  
> capacitance is on the transmitter side of the series inductance.

There is a quick way to convert the AKaabbcc word to actual L and C
values and configuration that does not require looking at a binary bit
pattern, correlating each set bit to a specific relay and reactance
on a schematic, and then adding up the connected values.

The KXAT3 has a maximum ATU inductance L = 15.93 uH.  The eight ATU
inductance switching relays divide this into 256 (almost) identical
steps.  When the ATU has found its best match, a hexadecimal value is
set in "aa" that is the number of those steps used by the current ATU
inductance setting.  So:

1L.  The inductance step size is 15.93 uH / 256 = 0.0622 uH.
2L.  Convert the hexadecimal value aa to the decimal value AAA.
3L.  Total inductance connected is AAA x 0.0622 uH.

The KXAT3 has a maximum ATU capacitance C = 2683 pF.  The eight ATU
capacitance switching relays divide this into 256 (almost) identical
steps.  When the ATU has found its best match, a hexadecimal value is
set in "bb" that is the number of those steps used by the current ATU
capacitance setting.  So:

1C.  The capacitance step size is 2683 pF / 256 = 10.48 pF.
2C.  Convert the hexadecimal value bb to the decimal value BBB.
3C.  Total capacitance connected is BBB x 10.48 pF.

That's all it takes to compute the effective L and C values.  The only
iffy part for a few may be converting a hexadecimal value to its decimal
equivalent.

Once the L and C values are known, all that remains is to determine
if the capacitor is connected on the antenna side (cc = 00) or the
transmitter side (cc = 01).

For Wayne's example of AK031A01:

The ATU inductance byte is 03(hex) = 003(decimal).
The ATU inductance thus is 3 x 0.0622 uH = 0.1866 uH.

The ATU capacitance byte is 1A(hex) = 025(decimal).
The ATU capacitance thus is 25 x 10.48 pF = 262 pF.

The ATU capacitance connection byte is 01(hex).
The ATU capacitor is thus on the transmitter side.

So, the ATU reactance is a series inductance of 0.1866 pF
with a capacitance of 262 pf from the transmitter end of
the inductance to common.

NOTE:  Because the eight inductances or capacitances are not
always *exact* factors of 2 from each other due to practical
individual manufacturing or value availability issues, there
will sometimes be very slight differences between the results
from this quick approach, and that from determining which bits
are set in two bytes and applying that to relay configuration
with associated reactance values in the ATU.  Applying that
painstaking approach to Wayne's example yields an inductance
value of 0.18 uH (vs. 0.1866 uH) and a capacitance value of
264 pF (vs. 262 pF).  Pretty darn close!

Were one very often interested in this detail about KXAT3 settings,
it would be great if firmware did the required work.

> If the ATU is not installed, or if the ATU MD menu entry parameter  
> is set to one of the L/C test values, "AK000000;" is returned.
>
> If the ATU is in "bypass" mode, quasi-bypass values are shown.
> (These were selected to approximately cancel out the ATU's own  
> stray reactance on each band.)
>
> In AUTO mode, the working auto-tuned values are shown.

I admire the attention to small detail that the description of the
bypass function shows.

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

[wayne burdick]

Mike Morrow wrote:
>

> That's all it takes to compute the effective L and C values.

Good idea.

I should also have pointed out that knowing the L and C values used  
doesn't necessarily tell you what your antenna impedance is. On higher  
bands, in particular, all ATUs have strays that make them look like a  
higher-order network than just an L-net (e.g., it can be modeled as a  
pi-net in some cases). We've done some modeling of the KAT500, which  
is similar, to aid in tracking the antenna impedance. It's a very  
complex set of calculations.


> The only iffy part for a few may be converting a hexadecimal value  
> to its decimal
> equivalent.

Many scientific calculators will do this. You can also find hex-
 >decimal calculators on the web.

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

[GARY-2]

Mike,
1A hex = 26 decimal not 25.
Gary
N6LRV

-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf Of Mike
Morrow
Sent: Monday, August 06, 2012 5:24 PM
To: [hidden email]
Cc: [hidden email]
Subject: [KX3] Re: KX3 Operating Tip: Determining the ATU's L/C values and
L-net configuration

For Wayne's example of AK031A01:

The ATU inductance byte is 03(hex) = 003(decimal).
The ATU inductance thus is 3 x 0.0622 uH = 0.1866 uH.

The ATU capacitance byte is 1A(hex) = 025(decimal).
The ATU capacitance thus is 25 x 10.48 pF = 262 pF.

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