low audio on K2 s/n 3641 (was RE: PLL Reference Oscillator Range Test)

[Chris Kantarjiev K6DBG]

Last night I sent out a message about my PLL reference oscillator range
test showing the range to be a bit too small; I was starting to track
down the low-audio on new-to-me s/n 3641.

Don W3FPR made several helpful suggestions off-list, none of which
panned out (including checking T7). I played around with the AGC
settings and discovered that there is reasonable (but not room-filling)
audio when the AGC is off, but turning on the AGC quiets it way
down. I also found that the S-meter response is not what I expect:

I injected a -50dBm signal at 7100KHz. I zero-beat it with SPOT on. RF
gain is fully clockwise.

- I really don't understand what the S-meter is meant to do when switching
AGC off and on: it seems that when the AGC is on, the S-meter reading goes
up even if the audio goes down; when I switch the AGC off, the S-meter
reading droops off the left edge. What's going on here?

- I expect this to be S9 or so, and it's much less. I have done CAL S LO
and CAL S HI several times, just to be sure. When I drop the signal to
-100dBm, the S-meter is zero; it takes about -70dBm to get the meter
to register.

I found a note from Gary Surrency of Elecraft in the archives, "K2: 25
reasons for low RX sensitivity or low audio. (long)". The first
things he mentions is checking CAL FIL. So I borrowed Leigh WA5ZNU's
ngen and used Spectrogram to set the filters per Tom N0SS's
excellent instructions. (Some day I'll understand why everyone
chooses different filter widths :-)

So I started on the Receiver Signal Tracing in my manual. Everything
was more or less correct until I got to the crystal filter:

4. Post-Mixer Amp Output. Expected 2.2Vrms. Got 3.02Vrms

5. -5dB Pad Output. Expected 1.40Vrms. Got 3.04Vrms

6. Crystal Filter Output. Expected 0.35Vrms. Actual 0.094Vrms.

I stopped there, because this seems very wrong.

In Gary's note, he mentions that the crystals here should be marked
4915-20, not 4915-S. Mine are marked 4.9136-S, which is what the
manual's inventory lists for the crystal filter set...

So. I'm open to suggestions. Thanks!

73 de chris K6DBG

P.S. Should these readings really be listed Vrms? It's a DC voltage
after it leaves the RF probe, isn't it? I'm certainly reading it
on my DC scale.

_______________________________________________
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 Brown-4]

Hi

I don't know if I can help but here are a few thoughts. First the S-meter is
just an indication of the AGC voltage so it is normal for it to not work
when the AGC is turned off because there is no AGC voltage when turned off.

If you are losing the signal out of the crystal filter check each crystal
pin to ground with an ohm meter. You should get a high reading not a short.
Sometimes if too much solder is used the solder will flow up under the
crystal shorting it to the ground plain on the top of the board under the
crystal. If one of them is shorted you will need to remove that crystal and
remove the excess solder.

The RF probe outputs a DC voltage to your meter but the component values on
the probe and the 10 meg ohm input resistance of the meter form a divider
that makes the DC voltage read on the meter equivalent to the RF RMS voltage
at the probe tip.

Don Brown

KD5NDB




----- Original Message -----
From: "Chris Kantarjiev" <[hidden email]>
To: <[hidden email]>
Sent: Tuesday, February 28, 2006 11:43 PM
Subject: [Elecraft] low audio on K2 s/n 3641 (was RE: PLL
ReferenceOscillator Range Test)

> Last night I sent out a message about my PLL reference oscillator range
> test showing the range to be a bit too small; I was starting to track
> down the low-audio on new-to-me s/n 3641.
>
> Don W3FPR made several helpful suggestions off-list, none of which
> panned out (including checking T7). I played around with the AGC
> settings and discovered that there is reasonable (but not room-filling)
> audio when the AGC is off, but turning on the AGC quiets it way
> down. I also found that the S-meter response is not what I expect:
>
> I injected a -50dBm signal at 7100KHz. I zero-beat it with SPOT on. RF
> gain is fully clockwise.
>
> - I really don't understand what the S-meter is meant to do when switching
> AGC off and on: it seems that when the AGC is on, the S-meter reading goes
> up even if the audio goes down; when I switch the AGC off, the S-meter
> reading droops off the left edge. What's going on here?
>
> - I expect this to be S9 or so, and it's much less. I have done CAL S LO
> and CAL S HI several times, just to be sure. When I drop the signal to
> -100dBm, the S-meter is zero; it takes about -70dBm to get the meter
> to register.
>
> I found a note from Gary Surrency of Elecraft in the archives, "K2: 25
> reasons for low RX sensitivity or low audio. (long)". The first
> things he mentions is checking CAL FIL. So I borrowed Leigh WA5ZNU's
> ngen and used Spectrogram to set the filters per Tom N0SS's
> excellent instructions. (Some day I'll understand why everyone
> chooses different filter widths :-)
>
> So I started on the Receiver Signal Tracing in my manual. Everything
> was more or less correct until I got to the crystal filter:
>
> 4. Post-Mixer Amp Output. Expected 2.2Vrms. Got 3.02Vrms
>
> 5. -5dB Pad Output. Expected 1.40Vrms. Got 3.04Vrms
>
> 6. Crystal Filter Output. Expected 0.35Vrms. Actual 0.094Vrms.
>
> I stopped there, because this seems very wrong.
>
> In Gary's note, he mentions that the crystals here should be marked
> 4915-20, not 4915-S. Mine are marked 4.9136-S, which is what the
> manual's inventory lists for the crystal filter set...
>
> So. I'm open to suggestions. Thanks!
>
> 73 de chris K6DBG
>
> P.S. Should these readings really be listed Vrms? It's a DC voltage
> after it leaves the RF probe, isn't it? I'm certainly reading it
> on my DC scale.
>
> _______________________________________________
> 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

[Chris Kantarjiev K6DBG]

>> So I started on the Receiver Signal Tracing in my manual. Everything
>> was more or less correct until I got to the crystal filter:
>>
>> 4. Post-Mixer Amp Output. Expected 2.2Vrms. Got 3.02Vrms
>>
>> 5. -5dB Pad Output. Expected 1.40Vrms. Got 3.04Vrms
>>
>> 6. Crystal Filter Output. Expected 0.35Vrms. Actual 0.094Vrms.
>>
>> I stopped there, because this seems very wrong.

DolfinDon suggested:

> If you are losing the signal out of the crystal filter check each crystal
> pin to ground with an ohm meter. You should get a high reading not a short.
> Sometimes if too much solder is used the solder will flow up under the
> crystal shorting it to the ground plain on the top of the board under the
> crystal. If one of them is shorted you will need to remove that crystal and
> remove the excess solder.

Nope, it's not a shorted crystal; all pins measure 'infinite'
resistance to ground.

I measured each pin with the RF probe. Here's what I get, which
doesn't make a lot of sense to me (this is viewed from the bottom
of the board):

        3.16v (in)

X7      1.1     3.16
X8      1.0     1.1
X9      0.58    1.0
X10     0.78    0.58
X11     0.48    0.78

These are "rough" measurements - neither the radio nor the signal
generator had been warmed up when I took them just before going to
work, but the relative values are about right.

There's no drop at the -5dB pad because I have removed a KNB2 and
jumpered 1-6 on the connector; the pad itself is 'gone'.

Should there really be such a big drop across any of the crystals? And
why would the voltage go back up?

(And should I be concerned that the crystals are marked 4.9136-S instead
of 4915-20?)

I notice, now that I look at them again, that the output measurement
is not very far from the 0.35V that's expected. There's a diode (D34)
between the last crystal pin and the jumper W3, perhaps that's the problem?

73,
chris
_______________________________________________
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

[Chris Kantarjiev K6DBG]

> (And should I be concerned that the crystals are marked 4.9136-S instead
> of 4915-20?)

Gary from Elecraft wrote to say that 4.9136-S is the current IF crystal part.

> I notice, now that I look at them again, that the output measurement
> is not very far from the 0.35V that's expected. There's a diode (D34)
> between the last crystal pin and the jumper W3, perhaps that's the problem?

Belay that - I wasn't quite away. The drop there strikes me as normal
forward drop through a diode. The varactor diodes all seem to be
installed in the right orientation.

I'm now most concerted about the 2V drop across the first crystal. If
the drop across that was similar to the drop across the others, the
voltage at W3 would be a very happy number. I don't measure any shorts.
I've aligned filters with Spectrogram. I peaked L34 with Spectrogram
(that actually made negative progress - now the output of the bandpass
at D6 is only about 0.070V, though the voltage on the can of Q22
went up to almost 5V). I've adjusted the VCO tuning for a peak.
I've retouched the solder joints on the crystal.

Other ideas?

73.
chris
_______________________________________________
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