? How to stabilize a cheap colpitts oscillator

[Dave]

I replaced the crystal in my basic Colpitts oscillator with a couple
hand-wound toroids and a variable capacitor, and the output now varies
wildly. Any ideas on how to stabilize it a little? Or how to tell if the
problem is my freq counter?

Thanks,

Dave

 

[Bruce Varley]

I replaced the crystal in my basic Colpitts oscillator with a couple
hand-wound toroids and a variable capacitor, and the output now varies
wildly. Any ideas on how to stabilize it a little? Or how to tell if the
problem is my freq counter?

Thanks,

Dave

First, what's 'wildly', and what are you expecting? You'll obviously lose
some frequency stability in changing from a xtal to LC. And is the amplitude
unstable as well, that's a key indicator of what the problem might be.

Also, IIRC a colpitts osc normally has one inductor. If you're using 2
toroids, is this 2 separate inductors? If so, should it be one inductor with
a tap, providing the mutual coupling that's required?

 

[Dave]

First, what's 'wildly', and what are you expecting? You'll obviously lose
some frequency stability in changing from a xtal to LC. And is the
amplitude unstable as well, that's a key indicator of what the problem
might be.

Also, IIRC a colpitts osc normally has one inductor. If you're using 2
toroids, is this 2 separate inductors? If so, should it be one inductor
with a tap, providing the mutual coupling that's required?

Wildly: fluctuating from 1.0xxxxx MHz to 6.7xxxx Mhz by degrees of 500 KHz
at a time or more. I was expecting the most significant digit to remain
stable, and the fluctuations to involve the two or three least significant
digits, instead of all digits changing over the span of a couple seconds.
And, no, the amplitude and frequency on the scope are stable, they don't
fluctuate at all. Every now and then it will settle down to roughly what I
expected, but that doesn't last long. No obvious clues as to why.

The two inductors are switched into the circuit one at a time, to allow me
to raise or lower the general frequency range to match that of the circuit I
am trying to test.

Thanks for the reply. Please don't hesitate to offer anything you think
of...

Dave

 

[John Larkin]

Wildly: fluctuating from 1.0xxxxx MHz to 6.7xxxx Mhz by degrees of 500 KHz
at a time or more. I was expecting the most significant digit to remain
stable, and the fluctuations to involve the two or three least significant
digits, instead of all digits changing over the span of a couple seconds.
And, no, the amplitude and frequency on the scope are stable, they don't
fluctuate at all. Every now and then it will settle down to roughly what I
expected, but that doesn't last long. No obvious clues as to why.

The two inductors are switched into the circuit one at a time, to allow me
to raise or lower the general frequency range to match that of the circuit I
am trying to test.

Thanks for the reply. Please don't hesitate to offer anything you think
of...

Dave

If it looks stable on a scope but it's jumping that much on a counter,
it's the counter's fault, likely a trigger threshold issue. You could
expect a decent LC colpitts to be short-term stable to 100 ppm sort of
turf, wiggling in maybe the 4th or even 5th digit.

What frequency does the scope suggest? Does the scope waveform look
stable while the counter is connected and jumping around?

John

 

[Dave]

If it looks stable on a scope but it's jumping that much on a counter,
it's the counter's fault, likely a trigger threshold issue. You could
expect a decent LC colpitts to be short-term stable to 100 ppm sort of
turf, wiggling in maybe the 4th or even 5th digit.

What frequency does the scope suggest? Does the scope waveform look
stable while the counter is connected and jumping around?

John

I honestly wondered about the counter as a possible source of the problem.
A trigger threshold problem would explain this perfectly in my mind. The
scope shows a stable 5 to 10 MHz, (depending on what I dial up with the
tuning cap in the Colpitts) while the counter says anything between 1 and
6.7 MHz no matter what the scope shows. Any ideas on how I could
troubleshoot the counter? I've got a lot more time than money...

Thanks,

Dave

 

[Phil Allison]

"Dave"

I honestly wondered about the counter as a possible source of the problem.

** Course it is the problem.

A trigger threshold problem would explain this perfectly in my mind. The
scope shows a stable 5 to 10 MHz, (depending on what I dial up with the
tuning cap in the Colpitts) while the counter says anything between 1 and
6.7 MHz no matter what the scope shows. Any ideas on how I could
troubleshoot the counter? I've got a lot more time than money...

** The counter needs to see a clean input signal ( not one full of
harmonics) with sufficient amplitude - likely at least 100mV.

Have you got a spec for the thing ?




........ Phil

 

[Dave]

"Dave"


** Course it is the problem.





** The counter needs to see a clean input signal ( not one full of
harmonics) with sufficient amplitude - likely at least 100mV.

Have you got a spec for the thing ?




....... Phil

A spec for the counter? No, it's old and I got it used. Would a tank
circuit produce a signal full of harmonics? I somehow thought it would be
fairly clean... And 100 mV shouldn't be a problem, I don't think. The
scope shows a 2v signal peak to peak.

Thanks,

Dave

 

[Phil Allison]

"Dave"


** The counter needs to see a clean input signal ( not one full of
harmonics) with sufficient amplitude - likely at least 100mV.

Have you got a spec for the thing ?

A spec for the counter? No, it's old and I got it used. Would a tank
circuit produce a signal full of harmonics? I somehow thought it would be
fairly clean... And 100 mV shouldn't be a problem, I don't think. The
scope shows a 2v signal peak to peak.

** Ever tried connecting the scope and counter at the same time ??

Might be instructive.




........ Phil

 

[Dave]

"Dave"


** The counter needs to see a clean input signal ( not one full of
harmonics) with sufficient amplitude - likely at least 100mV.

Have you got a spec for the thing ?




** Ever tried connecting the scope and counter at the same time ??

Might be instructive.




....... Phil

Yes, the scope and counter are connected simultaneously. That's why I
wondered about the counter...

Dave

 

[Phil Allison]

"Dave" <

Yes, the scope and counter are connected simultaneously. That's why I
wondered about the counter...

** This is getting silly and tedious.

DESCRIBE YOUR TEST GEAR FULLY !!






........ Phil

 

[Dave]

"Dave" <




** This is getting silly and tedious.

DESCRIBE YOUR TEST GEAR FULLY !!






....... Phil

I apologize, the freq counter is a Schlumberger 1252, and the scope is a Tek
2335 100MHz dual-trace, but I am only using one channel. I honestly don't
kow what else to tell you about the test equipment. Do not mean to be
tedious. I am just a hobbyist. I used to repair computers, but that was 15
years ago, and I am just getting back into the groove.

Dave

 

[Phil Allison]

"Dave"

I apologize, the freq counter is a Schlumberger 1252, and the scope is a
Tek 2335 100MHz dual-trace, but I am only using one channel. I honestly
don't kow what else to tell you about the test equipment.

** I asked you to DESCRIBE the items - not merely list the model numbers.

What is the frequency range of the counter (Goolge supplied no hits on the
model number) .

Die it have a trigger control or bandwidth filters ?

Why cannot YOU tell ME if the scope trace shows harmonics in the wave
form ?




........ Phil

 

[Dave]

"Dave"




** I asked you to DESCRIBE the items - not merely list the model
numbers.

What is the frequency range of the counter (Goolge supplied no hits on the
model number) .

Die it have a trigger control or bandwidth filters ?

Why cannot YOU tell ME if the scope trace shows harmonics in the wave
form ?




....... Phil

The counter goes up to at least 100 MHz, but that's all I know. The trace
on the scope is clean. Again, I apologize, but I don't know what to tell
you. I am still fairly new to RF, and am learning as I go along... If you
have a question, please ask. This is why I am asking for help...

Dave

 

[Frank Raffaeli]

I replaced the crystal in my basic Colpitts oscillator with a couple
hand-wound toroids and a variable capacitor, and the output now varies
wildly. Any ideas on how to stabilize it a little? Or how to tell if the
problem is my freq counter?

Thanks,

Dave

What's the purpose of the oscillaor? Perhaps a different circuit is
called for.
Was your frequency counter working with the old (xtal) version? Try:
1) attenuating the signal at the oscillator
2) decoupling the counter from the oscillator (e.g. 1 pF cap in series
between osc and counter)

Frank

 

[Dave]

What's the purpose of the oscillaor? Perhaps a different circuit is
called for.
Was your frequency counter working with the old (xtal) version? Try:
1) attenuating the signal at the oscillator
2) decoupling the counter from the oscillator (e.g. 1 pF cap in series
between osc and counter)

Frank

Hello Frank,

The oscillator is an attempt at a very basic signal generator, for the
purpose of determining the Q of the inductors in the tuning circuit of an
active shortwave antenna. Now that you mention it, the counter did function
properly when I had a xtal in the oscillator... Back to square one- how to
stabilize the oscillator. I will try your suggestions of attenuating the
signal, and decoupling the counter. Those are two things I would have never
thought of...

Many thanks,

Dave

 

[Dave]

Hello Frank,

The oscillator is an attempt at a very basic signal generator, for the
purpose of determining the Q of the inductors in the tuning circuit of an
active shortwave antenna. Now that you mention it, the counter did
function properly when I had a xtal in the oscillator... Back to square
one- how to stabilize the oscillator. I will try your suggestions of
attenuating the signal, and decoupling the counter. Those are two things
I would have never thought of...

Many thanks,

Dave

Well, a squirt of contact cleaner on the sensitivity adjustment knob (from
the outside, I'll take it apart tomorrow and do it right) helped a little,
but the decoupling capacitor had no effect, and I already had a 100K
resistor between the stator of the active antenna's tuning capacitor and the
frequency counter's input and as I recall that did help a little also. Now
fluctuates less wildly, but still too much to tell what's actually going on.
About to put it to bed for the night, and pick it up again in the morning.
Thanks for the ideas though. If you think of anything else, please don't
hesitate to suggest it. I apologize for being so ignorant about RF.

BTW, the freq counter is an autoranging model, with a 1 second, 10
millisecond and auto positions as well as a sensitivity adjustment. As I
mentioned, it counts at least to 100 MHz, but more than that I don't know.
It was a gift from my brother in law about 25 years ago, and it wasn't new
then. I suspect it was his first frequency counter.

Thanks to all. Sorry for the frustration...

Dave

 

[John]

The counter goes up to at least 100 MHz, but that's all I know. The trace
on the scope is clean. Again, I apologize, but I don't know what to tell
you. I am still fairly new to RF, and am learning as I go along... If you
have a question, please ask. This is why I am asking for help...

Dave

Does your frequency counter have a sensitivity control? If so, you may have
it set too high or too low. Try various settings to see if you can get the
reading to stabilize.

Good luck.

John

 

[Phil Allison]

"Dave"

but the decoupling capacitor had no effect, and I already had a 100K
resistor between the stator of the active antenna's tuning capacitor and
the frequency counter's input and as I recall that did help a little also.

** Now I begin to think you are a troll.

100 kohms in series with a counter's input at 6.5 MHz ???

Was that resistor in line before or after the input cable ?????

How about a JPEG or two of your set up plus test gear on ABSE ?



........... Phil

 

[D from BC]

I replaced the crystal in my basic Colpitts oscillator with a
couple
hand-wound toroids and a variable capacitor, and the output now
varies
wildly. Any ideas on how to stabilize it a little? Or how to
tell if the
problem is my freq counter?

Thanks,

Dave


..
I replaced the crystal in my basic Colpitts oscillator with a
couple
hand-wound toroids and a variable capacitor, and the output now
varies
wildly. Any ideas on how to stabilize it a little? Or how to
tell if the
problem is my freq counter?

Thanks,

Dave


..
-----------------
Stabilize an oscillator??...This is messing with my feedback
theorey. If it's too stable..wouldn't that be DC?
Or how about...
Making an oscillator by unstabilizing a DC source.

(I skipped reading the whole thread)

Practical
I recall making unstable oscillators with the following methods:
1) Ripple on power supply
2) thermal drift of feedback components
3) Unstable DC bias
4) crosstalk from other signals
5) wrong phase therefore wrong frequency (usually higher f and
fills up scope )
6) high impedance (finger approaching changes frequency)
7) 60Hz pickup from household wiring (It was very sensitive)

A bad designer knows more about what doesn't work
D
-------------

 

[Dave]

"Dave"





** Now I begin to think you are a troll.

100 kohms in series with a counter's input at 6.5 MHz ???

Was that resistor in line before or after the input cable ?????

How about a JPEG or two of your set up plus test gear on ABSE ?



.......... Phil

I may be ignorant, but I am sincere. The resistor feeds the input cable to
the counter. IIRC it was necessary to prevent overloading the counter's
input. Just thought of something: problem might be where I am taking the
input from- the tank circuit of the active antenna. Damn. Should have
thought of that sooner. I never claimed to be a genius, just persistant.

I'll see what I can do about a JPEG of the setup and test equipment. That
will have to wait until tomorrow however. It's tired and I'm late.

Thanks all, for your patience.

Dave