[K2 4922] IF Amplifier Alignment

[Ron Hahn (EI2JP)]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Colleagues,

I've moved on from the BFO issue (now resolved) to an IF alignment issue.

My problem is a constant S3 reading on the S-meter.

RF gain control works and is all the way turned up.  Birdie at
6.999.17MHz is clearly readable but is only strong enough to push the
S-meter up from S3 to S5.  When I move off the birdie frequency I hear
nothing on the headphones.

Mode:    CW
Filter:  FL2
RF gain: full
AF gain: 12 o'clock
Preamp:  on

The S-meter increases as the RF gain is decreased (as expected).

As the constant reading masks the S-meter reading it is hard to peak
the IF inductor L34 for maximum signal.

Where to go from here?

Thanks,

Ron

- --
- -=-------------------------o---------------------------=-
Ron Hahn, EI2JP            |         Grid Square: IO62TG
Rose Hill                  |         Fists:       10883
Rosslare Strand            |         QRP-ARCI:    12584
County Wexford, IRELAND    |         K2:          4922
                         __+__
                       
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGgtP3yJCeYjZnfkQRArYMAJ4ybS19sZJFhHvqG80a/ZmavNRvjgCgmOIP
pGISv6HI0RaYnFMapGCQQMg=
=F8Uk
-----END PGP SIGNATURE-----


--
This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.

_______________________________________________
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

[Don Wilhelm-3]

Ron,

Peak the IF amp inductor (L34) either by ear or by connecting an audio
voltmeter (or DVM in AC mode) to the speaker connection (RF Board P5) to
detect the maximum signal.

Attempting to use the S-meter at this stage is not reliable.

In any case, the peak is very broad.  If you cannot discern a peak, set
the slug 1 turn from the top of the can.

You can (and should) re-adjust the S-meter menu settings CAL S Lo and
CAL S Hi after you complete all other assembly and test items.

73,
Don W3FPR

Ron Hahn (EI2JP) wrote:

> I've moved on from the BFO issue (now resolved) to an IF alignment issue.
>
> My problem is a constant S3 reading on the S-meter.
>
> RF gain control works and is all the way turned up.  Birdie at
> 6.999.17MHz is clearly readable but is only strong enough to push the
> S-meter up from S3 to S5.  When I move off the birdie frequency I hear
> nothing on the headphones.
>
> Mode:    CW
> Filter:  FL2
> RF gain: full
> AF gain: 12 o'clock
> Preamp:  on
>
> The S-meter increases as the RF gain is decreased (as expected).
>
> As the constant reading masks the S-meter reading it is hard to peak
> the IF inductor L34 for maximum signal.
>

_______________________________________________
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

[David Woolley (E.L)]

Don Wilhelm wrote:
>
> Ron Hahn (EI2JP) wrote:
 >> My problem is a constant S3 reading on the S-meter.
 >>

I'm glad I'm not the only one to notice this!

>
> Attempting to use the S-meter at this stage is not reliable.

The manual does, however, imply one should attempt this, by suggesting
one gets a good S meter reading before starting.
>
> You can (and should) re-adjust the S-meter menu settings CAL S Lo and
> CAL S Hi after you complete all other assembly and test items.

I don't believe the manual says this, although I've only got to the 40m
receive only stage.

I'm actually having great fun trying to work out how the AGC system
really works and really should be calibrated.  One complication is that
the MC1350 is only characterised for a 12 volt supply, but the K2 uses
8. If I understand the internals of the chip properly, the AGC range
will be constant but the threshold will reduce by one volt for every two
volts reduction in supply voltage, so the 5 to 7 volts range in the data
sheet will correspond with 3 to 5 volts in the K2 (at the V AGC point,
not at U2 pin 5).  Is that right?

Another question is how one gets 88dB+ of S meter range (6dB * (9 - 1) +
40dB) from an AGC system with a maximum range of 68dB, especially when
the initial S meter calibration uses less than 1.3 volts of the
available 2 volt control range?

Another one is that the design seems to assume a gain balance between
the AF and AGC IF chains, but the differential gain is subject to
production spread in the NE612 chips (minimum to typical gain of 3dB and
  unspecified typical to maximum gain).  It also depends, to a lesser
extent, on the output impedance of the 612 in the AGC chain.  The AF
chain will depend on the loss in the second filter; I don't know enough
about crystal filters to estimate this.   There are probably some other
factors.

(I think there is a deliberate 6dB difference in that the nominal gain
of the U2a op amp is about 22 (33k explicit over 1.5k output
resistance), and the LM 380 has a nominal gain of 50).

I wonder if the feedback resistor on U2a really could do with being a
variable.  Although you can get a similar effect with R1 (because the
main part of the AGC response is exponential), you lose range on the RF
gain control.



--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
_______________________________________________
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

[Don Wilhelm-3]

David,

Jack Smith has characterized the K2 AGC curves (along with some other
rigs) and you may find some useful information on his website:
http://www.cliftonlaboratories.com/receiver_agc_curves.htm

The IF AMP AGC input is really current driven, so trying to obtain a
direct relationship between the voltage and the AGC action is
problematic at best - yes, I know the data sheet expresses the AGC
voltage in the curves, but that is based on the operating conditions
stated in the data sheet.

When you set up the AGC Threshold voltage in the K2 to 3.8 volts, you
are not reading the voltage into the MC1350, but rather reading an
indirect control point voltage - the actual AGC drive to the MC1350 is
two active devices 'down the road' from where the AGC Threshold is set.

What I can say for certain is that I have had to re-visit the AGC
Threshold and S-meter settings after completing assembly of any K2 that
I have built or repaired (I have built many and repaired many more) -
for instance, the filter passband alignment will influence the S-meter
response, so the S-meter should best be set after the final filter
alignment.

True, the manual does not explicitly state that one should re-adjust the
S-meter settings after all the other alignment is done, but it does hint
at it.

One can 'play' with the AGC Threshold voltage setting to achieve better
receive sensitivity with no input signal, but move it very far below 3.8
volts and the S-meter settings and response gets 'squirrelly', so if you
choose to do that, plan on a lot of interactive 'playing' with the AGC
Threshold, CAL S Lo and CAL S Hi - the manual settings are a compromise,
but they do work and work quite well - vary from them as you wish, but
do so at your own peril, the 'standard' settings can be easily restored.

73,
Don W3FPR

David Woolley wrote:

>
> I don't believe the manual says this, although I've only got to the 40m
> receive only stage.
>
> I'm actually having great fun trying to work out how the AGC system
> really works and really should be calibrated.  One complication is that
> the MC1350 is only characterised for a 12 volt supply, but the K2 uses
> 8. If I understand the internals of the chip properly, the AGC range
> will be constant but the threshold will reduce by one volt for every two
> volts reduction in supply voltage, so the 5 to 7 volts range in the data
> sheet will correspond with 3 to 5 volts in the K2 (at the V AGC point,
> not at U2 pin 5).  Is that right?
>
> Another question is how one gets 88dB+ of S meter range (6dB * (9 - 1) +
> 40dB) from an AGC system with a maximum range of 68dB, especially when
> the initial S meter calibration uses less than 1.3 volts of the
> available 2 volt control range?
>
> Another one is that the design seems to assume a gain balance between
> the AF and AGC IF chains, but the differential gain is subject to
> production spread in the NE612 chips (minimum to typical gain of 3dB and
>  unspecified typical to maximum gain).  It also depends, to a lesser
> extent, on the output impedance of the 612 in the AGC chain.  The AF
> chain will depend on the loss in the second filter; I don't know enough
> about crystal filters to estimate this.   There are probably some other
> factors.
>
> (I think there is a deliberate 6dB difference in that the nominal gain
> of the U2a op amp is about 22 (33k explicit over 1.5k output
> resistance), and the LM 380 has a nominal gain of 50).
>
> I wonder if the feedback resistor on U2a really could do with being a
> variable.  Although you can get a similar effect with R1 (because the
> main part of the AGC response is exponential), you lose range on the RF
> gain control.
>

_______________________________________________
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

[Clark Macaulay KE4RQ]

Don, I found your answer on the setting of R1 in the last reply of interest to me.  I'm buiding K2 #6100 and when I got to that part of the build process was unable to get 3.80: my dmm reads 3.76 but only at the extreme end of R1's range.  
   
  At first, I didn't think much of it and have started on the Phase II of the RF board.  Your message would imply that the desired 3.80 is probably a minimum and I can't even reach that!  Since I've always had some trouble chasing agc circuits, could you suggest one place to start looking for the trouble?  As they say, "it all LOOKS fine to me" and the resistance checks were good.
   
  73 de Clark ke4rq
_______________________________________________
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

[David Woolley (E.L)]

Clark Macaulay wrote:
> build process was unable to get 3.80: my dmm reads 3.76 but only at the
> extreme end of R1's range.

I forgot to mention that I can only get 3.79, but that seems to be a bug
in the manual.  If one assumes U2b (i.e. got the a and b mixed before)
doesn't load the voltage, I think the end stop voltage with all
components nominal is 3.82; that means that 3.76 is less than 2% out, so
within the component tolerances.  It looks as though the manual should
really have said set R1 at the end stop.

>    
> At first, I didn't think much of it and have started on the Phase II
> of the RF board. Your message would imply that the desired 3.80 is
> probably a minimum and I can't even reach that! Since I've always had

It's a maximum.  The original design didn't have a variable, but it was
found that many people wanted to lower R1 slightly.  The material that
describes the mod to the old version gives a better description of how
one should vary R1; the current manual simply says you may want to
adjust it, without any clue as to why.



--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
_______________________________________________
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