[Elecraft] K2 - VCO-Test (Alignment an d Test, Part II)

[Masleid, Michael A.]

Hello David,

>When I "tune up" the frequency by putting a finger on t5, followings
>happend:
> - as long the frequency is rising the voltage at R30 is 8.1 volts
> - when frequency arrives at 9.1MHz the voltage at r30 goes to 4volts. =
The
>frequency goes down to 8.9MHz

Good.  This is the PLL control circuit adjusting the varacter diodes to =
get
the frequency where it needs to be.

>when I make the finger wet, the frequency rises much faster and the =
voltage
>at r30 changes faster too.

Making the finger wet increases capacitance a bit, and decreases =
resistance
more.  That's in series with your body, a rather large =
capacitor/antenna.
But I don't know why that makes a difference.

>I tried to adjust l30. It was possible to adjust it to 6 volts (with =
the
>finger at t5). When I removed the finger, the voltage goes up to 8.1 =
volts.

Good.  This means that whatever is wrong has nothing to do with L30 or =
the
PLL control circuits.  They do as they should once T5 is oscillating =
correctly.

>Is it possible that I increase the inductivity with the finger???

No.  No one has that much iron in their blood.  Unless something is =
shorting out
or opening up, but your measurements don't indicate that.

>Can it be that the core has a invisible crack (falling down on the =
floor)?

Yes, but it doesn't usually matter.  The core is an insulator.  The =
magnetic
particles are not really touching each other, so a crack doesn't really =
hurt
anything.

>C96 is "105" -> 1000'000pF -> 1uF -> ok

OK.

>The voltage at r9 is at the front-connector 8.1 volts and at the
>backside-connector 1.6 volts. By putting a finger to t5 the voltage at =
r9
>(backside-connector) rises linear withe the frequency. At 8.9MHz the =
voltage
>is 2 volts. When die voltage is rising there ist a quiet sound in the =
DMM.
>It starts at fow frequency and stops at high frequency. At 1.6 volts =
and 2
>volts is no sound.

The 8.1 volts is a supply voltage.  It is normal.
The 1.6 volts is level control.  It should be 2.2 volts.  It is wrong, =
but
it may be a result, not a cause.
Rising linearly with frequency.  I would expect it to jump to 2 volts.  =
But
sometimes I am wrong.  It is possible that the 100 uF capacitor is =
slowing
down the circuit.
Quiet sound on the DMM?  Most don't make noise.  Mine does, because of =
the
auto range relays.  Is that what you are hearing?

>I put the finger between the connectors 2 and 4 at t5.

Can't do that.  K15 is in the way.  Are you on the bottom side of the =
board?

>The serial-number is 3613.
>I have only a DMM.

>Next intresting point: I measured the voltage between ground and =
connector 4
>of t5 - its 0.07 volts (touching t5 has no effect) but the frequency =
goes up
>to 8.8 MHz!! When I touched the connector 4 with a insulator or =
metal-piece
>the frequency goes not to 8.8 MHz.

Nice.  The DVM is a high impedance capacitance to an antenna - same as =
your finger
really.  But terminal 4?  This is nasty.  This should be at 2 to 3 =
volts.  Does this
voltage stay at 0.07 volts even when the frequency is up to 8.8 MHz?  =
The voltage is
supposed to be 2 or 3 volts to hold Q18 near cut off so that the VCO =
doesn't oscillate
at too much amplitude.  Make sure that C60, 100 uF is not backwards =
(shorted).
(I might try to grounding Q17 Gate through a 1K resistor in order to cut =
off Q17,
then voltage on C60 should rise and shut down Q18, but you might not =
want to do that.)=20

Now, a little capacitance on terminal 2 might add a little more feedback =
into Q18 and
wake it up.  A little more capacitance on terminal 4 should not do much.

What is the voltage on Q18's Drain?

What we know so for, I think:  T5 is good.  K13-K15 and capacitors are =
good.  Feedback
path and ALC path around Q18 is bad.  Q18 is being pushed too hard.  =
Might be parts
around Q18, might be Q18 is bad?  If the voltage at Q18 Drain is low, =
then Q18 is
probably OK, but something else in the Q18, Q17, Q16, LT1252 loop is =
bad.

Just had an odd thought:  It could be that your finger is loading down =
the oscillator
enough to stabalize it.

Do check C60!  If ALC doesn't control Q18, then D13 could conduct and =
quench oscillation
on Q18.  Rather like a super-regenerative receiver.  Very noisy RF.  The =
bursts of 8 MHz
RF would fool the counter into thinking it saw some fractional part of =
that.  It would
help to check this with a 'scope but any receiver antenna near T5 should =
pick it up.
Look for a very ragged signal around 8-10 MHz.

73 de Michael AB9GV