VCXO frequency isn't high enough

[Anthony Fremont]

If you want a VXCO, you use the capacitor and inductor to achieve
maximum range of the crystal frequency.

An inductor in series will lower the frequency.

Ok, just like an antenna

A capacitor in series will raise the frequency.

Same thing again, just like an antenna. However, when I increase the series
capacitance, the frequency drops instead of going higher. This I don't
really get.

Eliminate the coil and just use the series trimmer to see how
how high the crystal frequency can be raised before it becomes
unstable or quits oscillating.

Did this, it helps, but when capacitance is below about 10pF, it seems to
get unstable and I still can't achieve the frequency I desire.

As Ian suggested, you can add an inductor ACROSS the
crystal to raise the frequency. Think of the crystal as being
a parallel LC circuit (over simplification) and you'll
be better able to visualize how this works.

Can you suggest a value to start with?

Try a few other color burst xtals in the circuit for the heck
of it. You may not be able to pull the crystal that far.

That's what I'm thinking now, it just won't pull that far high. I can sure
pull it low though. Maybe I should try a 4MHz crystal and pull it down to
3990, this is where the voice transmissions are anyway.

 

[Anthony Fremont]

I know that I'm "violating the specs", but that's the intent here. Most of
the circuits I've encountered tend to pull the xtal down in frequency (just
add capacitance), but this case is odd in that W1AW transmits just above the
colorburst frequency. I can pull it down several kcs pretty easily, I just
can't get it above the specced value by more than about 500Hz. I think I
just need to try a different crystal or maybe two in parallel. The parallel
inductance idea is interesting as well, never done anything like that
before.

As for stability, as long as it's not causing warbling audio, I'm not too
concerned.

 

[Anthony Fremont]

My W1AW receiver uses the crystal with just a 5-47 pf variable
capacitor...no inductor, and it nets right on frequency. Try shorting

Story of my life. ;-)

the inductor and see if this gets you closer to the required
frequency. Not all color burst crystals were created equal. If you
have another crystal you might want to try it.

That seems to be the common consensus. I suspect my crystal is just too
good. ;-)

Your ferrite will not be saturating at the small amount of signal you
are sending through it as part of an SA-602 oscillator.

Thanks, I know very little about these things.

 

[colin]

Its only a pull of ~100ppm, this should be easily pullable for most
fundamental xtals.
ofc if the tolerenace all add up against you you might find it hard.

I would also try reduce the 100pf caps on the sa602 too.
100pf is higher than most crystals I use would like,
50pf or 20pf or if youve got some spare trimmers ...
you can adjust the ratio too, say just reduce the one accross pin 6-7

Colin =^.^=

 

[[email protected]]

Hello all,

I was playing around and saw that my junk box had all the parts so I started
tossing this together:http://newenglandqrp.org/files/w1aw-receiver.jpg
The problem is (well I think it's a problem) is that I'm all the way down to
a 10pF cap for the crystal trimmer and the highest frequency I can get out
of it is still less than 3580kHz. Pleae correct me if I'm wrong, but I'm
thinking that the 20uH inductor is supposed to pull the colorburst crystal
high in frequency

The inductor will pull the crystal down in frequency, as others have
suggested.

In fact it is extremely difficult to pull a crystal's series resonance
up in frequency more than a few Hz. This is because the crystal's
parallel resonance is just above its series resonance. If you put a
capacitor in series with the crystal the series resonant frequency
goes up...BUT...if you approach the parallel resonant frequency you
can no longer get a low impedance resonance condition since the
crystal's parallel resonance makes the crystal look like an open
circuit, regardless of what you put in series with it.

If you want to understand this better, try to find a reference with a
good discussion of the equivalent circuit of the quartz crystal
resonator. the only one I know of at present is Kenneth K. Clarke and
Donald T. Hess, Communication Circuits: Analysis and Design, Addison-
Wesley Publishing Co., 1971. It may be a bit hard to find outside a
good university library.

As Ian Jackson suggested a parallel inductor might work....this
modifies the parallel resonance. Ian..is there a schematic available
for that ? I'd be interested in what actually worked.

Steve

 

[Anthony Fremont]

Its only a pull of ~100ppm, this should be easily pullable for most
fundamental xtals.

I'd like to get close to 500ppm if possible.

ofc if the tolerenace all add up against you you might find it hard.

I would also try reduce the 100pf caps on the sa602 too.
100pf is higher than most crystals I use would like,
50pf or 20pf or if youve got some spare trimmers ...
you can adjust the ratio too, say just reduce the one accross pin 6-7

I figured that they were voltage dividers to set the amount of feedback, but
I can certainly see how they could have an effect. Since I'm close to where
I need to be, I will try a couple of 33pF caps to see what happens.

 

[Anthony Fremont]

The inductor will pull the crystal down in frequency, as others have
suggested.

I understand now. I knew that adding capacitance tended to slow xtals down.
I was unfamiliar with how an inductor would act in series. I knew that this
circuit was trying to pull the crystal higher than spec so I, quite wrongly,
assumed that the inductor would have the opposite effect of capacitance.
And it seems that it does when used in parallel and that seems sensible to
my feeble mind. But in series, it has the opposite effect. What still
throws me is that raising series capacitance doesn't have the opposite
effect of adding parallel capacitance.

In fact it is extremely difficult to pull a crystal's series resonance
up in frequency more than a few Hz. This is because the crystal's
parallel resonance is just above its series resonance. If you put a
capacitor in series with the crystal the series resonant frequency
goes up...BUT...if you approach the parallel resonant frequency you
can no longer get a low impedance resonance condition since the
crystal's parallel resonance makes the crystal look like an open
circuit, regardless of what you put in series with it.

Hmm....food for thought.

If you want to understand this better, try to find a reference with a
good discussion of the equivalent circuit of the quartz crystal
resonator. the only one I know of at present is Kenneth K. Clarke and
Donald T. Hess, Communication Circuits: Analysis and Design, Addison-
Wesley Publishing Co., 1971. It may be a bit hard to find outside a
good university library.

As Ian Jackson suggested a parallel inductor might work....this
modifies the parallel resonance. Ian..is there a schematic available
for that ? I'd be interested in what actually worked.

Doodling with reactance formulas, it appears that 20uH (coincidence?) would
offset 100pF of capacitance fairly well by having a an opposing reactance
(well resistance at this point) of about 450 Ohms at 3581kHz, the same as
100pF. I'll try putting the coil I made in parallel and see what happens.
Hopefully it won't short the oscillator and kill my 15 year old NE602, I
only have two spares.

 

[Anthony Fremont]

Doodling with reactance formulas, it appears that 20uH (coincidence?)
would offset 100pF of capacitance fairly well by having a an opposing
reactance (well resistance at this point) of about 450 Ohms at

s/.well resistance at this point.//
It's just inductive reactance, I need more coffee. ;-)

3581kHz, the same as 100pF. I'll try putting the coil I made in
parallel and see what happens. Hopefully it won't short the
oscillator and kill my 15 year old NE602, I only have two spares.

Should I be afraid to do this? Does it need something to block DC current?

 

[colin]

I'd like to get close to 500ppm if possible.

do you need that much ?
what freq you need ?
whats the colourburst crystal freq ?

I figured that they were voltage dividers to set the amount of feedback,
but I can certainly see how they could have an effect. Since I'm close to
where I need to be, I will try a couple of 33pF caps to see what happens.

no they are involved in setting the frequency too,
in order for you circuit to work it needs to resonate,
with the 2 100pf the input where the crystal is looks like a capacitor with
some negative impedance,
the circuit with your crystal, inductor, and trimmer must be inductive,
it then forms a resonant ciruit with the capacitance of the input.
usualy the crystal would just be operated so that it looks inductive.

the negative part of the input impedance must be stronger than the loss in
the tuned circuit.
this is affected by the ratio of the 2 100pf capacitors.

a crystal can apear to be a very high inductance at resonance,
at the point where you want to operate it probably has very high inductance
indeed.
you can determnine the eqv inductance by using the equivalent internal
inductance and capacitance. you need to find the mutual capacitance of the
crystal wich is hard to find man specs wich tel you this but it is often
something like 14ff for example.
(0.014pf) you can then work out the eqv series inductance for it to resonate
with the std load wich may be 20pf.

you can then work out what inductance the crystal will apear to have at the
frequency you want. and hence the series capacitance you need.
you might find the inductance is so high that you need less than 1pf or it
has become capacitive.


Colin =^.^=

 

[colin]

s/.well resistance at this point.//
It's just inductive reactance, I need more coffee. ;-)


Should I be afraid to do this? Does it need something to block DC
current?

the trimmer should block all the dc.
im not convinced about the inductor in parallel with the crystal
it might work though.

 

[[email protected]]

I believe this is direct conversion and that the crystal should exactly
match the desired station (3581kHz) I want to recieve, that's right isn't
it? Maybe this is close enough? I have to wait til tonight to see if I can
actually hear anything.

Depends on what sort of signal you are trying to receive.

To receive a single our double sideband signal (or, in a very touchy
way, an AM one), you want your direct conversion receiver's local
oscillator exactly on the transmitter frequency.

But to receive a CW signal, you want your local oscillator 700 Hz or
so above or below the transmit frequency. That will cause you to hear
a 700 Hz audio tone for the CW.

Unless I'm mistaken, 3.581 MHz is in the CW portion of the 80 meter
band.
If you can get to 1 KHz below the transmitter, you should be able to
copy CW as 1KHz audio. If you can only get to 1.5 KHz below, you'll
get 1.5 KHz audio... not ideal listening, but probably workable.

 

[Arv]

s/.well resistance at this point.//
It's just inductive reactance, I need more coffee. ;-)


Should I be afraid to do this? Does it need something to block DC current?

Anthony

Your trimmer cap will block any DC flow to ground. If you are really
paranoid then put a 0.1 mfd in series with the crystal.

If you do accidentally ground the crystal input terminal on that
SA602, it only pulls the base of a transistor to ground and thus turns
it off. That shouldn't hurt anything.

While you may not be able to purchase new "602's" now (they are long
ago declared obsolete) the Phillips SA-612 is the same unit and is
readily available from a number of outlets.

A datasheet for this device is located at:

http://www.nxp.com/pip/SA612AD_01.html

Arv - K7HKL
_._

 

[Anthony Fremont]

do you need that much ?

I "need" an increase of about .0546%. Isn't that about 546ppm?

what freq you need ?

3581.5kHz to zero beat with the desired signal.

whats the colourburst crystal freq ?
3579.545kHz.



no they are involved in setting the frequency too,
in order for you circuit to work it needs to resonate,
with the 2 100pf the input where the crystal is looks like a
capacitor with some negative impedance,
the circuit with your crystal, inductor, and trimmer must be
inductive, it then forms a resonant ciruit with the capacitance of the
input.
usualy the crystal would just be operated so that it looks inductive.

Ok, I had to read that a few times to get it. A period between ".....100pf"
and "the input..." there would have been quite helpful. ;-)

the negative part of the input impedance must be stronger than the
loss in the tuned circuit.
this is affected by the ratio of the 2 100pf capacitors.

So they do act like a voltage divider of sorts and shunt some of the
oscillator output to ground and some back to the input.

a crystal can apear to be a very high inductance at resonance,
at the point where you want to operate it probably has very high
inductance indeed.
you can determnine the eqv inductance by using the equivalent internal
inductance and capacitance. you need to find the mutual capacitance
of the crystal wich is hard to find man specs wich tel you this but
it is often something like 14ff for example.
(0.014pf) you can then work out the eqv series inductance for it to
resonate with the std load wich may be 20pf.

you can then work out what inductance the crystal will apear to have
at the frequency you want. and hence the series capacitance you need.
you might find the inductance is so high that you need less than 1pf
or it has become capacitive.

Cool, a way to figure out just how high you can pull it and how to attain a
certain frequency. I'll probably stick to tinkering though. ;-) Thanks
allot for the detailed explanation.

 

[Anthony Fremont]

Anthony

Your trimmer cap will block any DC flow to ground. If you are really
paranoid then put a 0.1 mfd in series with the crystal.

Doh, I see that now. For some reason I was thinking that they crystal had
both pins connected to the oscillator on the 602, and not with one leg
grounded as the circuit has it. Having yet another senior moment I guess.
Only about the tenth one so far this week.

If you do accidentally ground the crystal input terminal on that
SA602, it only pulls the base of a transistor to ground and thus turns
it off. That shouldn't hurt anything.

While you may not be able to purchase new "602's" now (they are long
ago declared obsolete) the Phillips SA-612 is the same unit and is
readily available from a number of outlets.

I saw that 612 part when I was poking around on the net. A fellow ham gave
me three NE602s about 15 years ago. I had fogotten about them and recently
found them. I even have a couple of tuning cap vernier drives he gave me.

A datasheet for this device is located at:

http://www.nxp.com/pip/SA612AD_01.html
Thanks

Arv - K7HKL
_._

_.. . _. ..... __._ __ __.

 

[[email protected]]

3581.5kHz to zero beat with the desired signal.

But why do you want to zero beat?

 

[colin]

I "need" an increase of about .0546%. Isn't that about 546ppm?


3581.5kHz to zero beat with the desired signal.


3579.545kHz.

aha ok, thats a fair bit, but maybe within range, as as been said at a
certain frequency the crystal becomes a complete open circuit.

Ok, I had to read that a few times to get it. A period between
".....100pf" and "the input..." there would have been quite helpful. ;-)

Yeah I kinda got lost in my own explanation myself. didnt have much time to
explain.

So they do act like a voltage divider of sorts and shunt some of the
oscillator output to ground and some back to the input.

this circuit is an emiter folower wich has no voltage gain,
so actually they operate in the opposite way wich is kinda confusing but let
me explain ...
consider a typical tuned circuit with LC and a tap in the L,
driving the circuit at the tap acts as a step up,
but the crystal is the inductor wich make it difficult to put a tap here,
but at resonance the 2 capacitors can act in the same way and provide a
voltage step up.

but you can see there is a curent loop involving all these components in
series,
this is what sets the frequency.


If you decrease the capacitor accross the 2 pins of the ic this will
increase the voltage accros it and so give more drive, as wel as increase
the frequency.

another way to look at it is if you consider that ground is at the base then
the capacitors are in fact a voltage divider wich feed into the emiter of a
comon base amplifier.

Cool, a way to figure out just how high you can pull it and how to attain
a certain frequency. I'll probably stick to tinkering though. ;-)
Thanks allot for the detailed explanation.

The pulling range is usually equal to the motional capcitance over the
crystal self capacitance, so for 14ff and 20pf this gives 700ppm, but
actually at this extreme its unusable in this circuit as its required to be
inductive.

as said by some1 else the farther you pull it the worse the performance.

Colin =^.^=

 

[Anthony Fremont]

Depends on what sort of signal you are trying to receive.

To receive a single our double sideband signal (or, in a very touchy
way, an AM one), you want your direct conversion receiver's local
oscillator exactly on the transmitter frequency.

But to receive a CW signal, you want your local oscillator 700 Hz or
so above or below the transmit frequency. That will cause you to hear
a 700 Hz audio tone for the CW.

Unless I'm mistaken, 3.581 MHz is in the CW portion of the 80 meter
band.

It is, I mistakenly thought they did the 80 meter voice announcement down
there. The voice bulletin is actually on 3990kHz.

Since I have most of what I need (tuning caps, vernier and toroids/wire) I
started looking around at different VFO circuits. I'm trying to find
something that will span from 3500 to 4000kHz and still be fairly simple. I
could then inject this into pin 6 of the 602 and be done with it.

If you can get to 1 KHz below the transmitter, you should be able to
copy CW as 1KHz audio. If you can only get to 1.5 KHz below, you'll
get 1.5 KHz audio... not ideal listening, but probably workable.

Right. I was hoping to be able to "get on the other side" of the signal to
have a second chance at avoiding possible QRM.

 

[Anthony Fremont]

But why do you want to zero beat?

For CW I don't, but I would want to be able to get on the other side for QRM
reasons.

 

[[email protected]]

Right. I was hoping to be able to "get on the other side" of the signal to
have a second chance at avoiding possible QRM.

My guess is the project's original author hadn't planned on that...
most likely, they'd planned on rubbering the crystal only far enough
to get a reasonable audio tone from the RF-LO combination.

Obviously you can get more range out of a VFO, though building nice
VFO's isn't simple.

Even the tuning caps can be a pain. I spent a while as a teenager
knocking alternate plates out of old 365 pf AM broadcast caps to try
to make some suitable for 40 meters - or you can use a series
capacitor. Beware hand capacitance when you go to tune it. Today,
varacter tuning is another option - stable regulated supply and a
multi-turn pot.

Another thing you might do is google the "poundshop" (dollar store)
receiver projects. Those are little KHz-IF varactor tuned auto-
scanning FM radios that people have been modifying into direct
conversion ham band receivers.

 

[Eamon Skelton]

I guess I'm at the limits of that particular crystal.

I think you are getting close to the limits You might get a
few Hz more by reducing the value of the 2 100pF capacitors. Try
another crystal or a ceramic resonator.

73, Ed. EI9GQ.