"JLB" ) writes:
Can't quite tell from your description what your circuit is, but if I recall
correctly back when i was doing hardware design a crystal oscillator using
logic gates required two gates. I used to use three---two for the
oscillator and the third for a buffer.

The two inverters where tied together with a capacitor between them. Then
the crystal was placed from the input of the first gate to the output of the
second. The output of the second also went to the buffer stage.


!-----------------crystal------------|
Like this: - gate one---capacitor---gate two---gate three---circuit
output

As I recall, there was also a resistor from the input of gate one to ground,
and from the output of gate two to ground.

I do not recall the capacitor or resistor values, but when properly built
this circuit never failed to oscillate with any crystal within the range of
the gates (fundamental mode only).

Jim
N8EE

That was a standard oscillator with TTL. I used a 7400, and 470ohm resistors
from input to output of each gate. No coupling capacitor was needed, though
there was a low value cap in series with the crystal for load capacitance.
(The resistors linearized the gates.)

There were obviously variations on the theme, and your suggestion was
one of them.

And as you say, these tended to oscillate with most crystals. I had one
built up as a module, so I'd have a general purpose oscillator read,
to check crystals or as a signal source.

But, for some reason, when they moved to CMOS oscillators, the norm
became one gate, with a feedback resistor and a pair of capacitors, one
from input to ground, and the other from output to ground.

Michael VE2BVW

"PaoloC" wrote in message
...
Hi.
I have spent part of the weekend trying to resonate al old CB XTAL at
its fundamental frequency.

The XTAL is labelled 27.125 MHz, with a fundamental of about 9.041 MHz,
which falls into 18m HAM band when multiplied by two. I assume 27MHz
XTALs are 3rd overtone.

Since the circuitry is/will be digital, the oscillator is one gate of a
74HC14. 470ohm resistor from gate output to the parallel of 1Mohm//XTAL.
10pF and 33pF (from the junkbox) capacitors to ground on each side of
the XTAL.

No oscillation (I have no oscilloscope, I use my HF receiver to
troubleshoot oscillators at known frequencies).

If I replace the XTAL with a 10.000 MHz rock the oscillation is loud and
clear.

I have never built something with an overtone XTAL. I know that I need
an output resonating circuit if I want to extract the 3rd harmonic. Do I
need the same if I want the fundamental?

Are overtone XTALs "harder" to resonate?

Are those old CB XTALs 3rd overtone?

I assume my 27MHz XTAL works. :-) Thanks in advance for all suggestions,
Paolo IK1ZYW

"Michael Black" wrote in message
...

But, for some reason, when they moved to CMOS oscillators, the norm
became one gate, with a feedback resistor and a pair of capacitors, one
from input to ground, and the other from output to ground.

Oh, yes. That rang a bell.

CMOS logic is easier to get to work as an analog device than TTL. At least
for the good ol' RCA 4000 series. I am not familiar with the newer 74HC00
series to judge its performance. I would suspect that the 'bias' resistors
would have to be carefully chosen and that it would be sensitive to both the
crystal frequency and the ambient temperature.

When I was doing circuit desgin everyone most were still using TTL. The
74LS00 series was the hot item at the time.

Just from a logical viewpoint (pun not intended, by the way) I still think
he should try using two gates in a ring circuit.

Jim
N8EE

"Michael Black" wrote in message
...

But, for some reason, when they moved to CMOS oscillators, the norm
became one gate, with a feedback resistor and a pair of capacitors, one
from input to ground, and the other from output to ground.

Oh, yes. That rang a bell.

CMOS logic is easier to get to work as an analog device than TTL. At least
for the good ol' RCA 4000 series. I am not familiar with the newer 74HC00
series to judge its performance. I would suspect that the 'bias' resistors
would have to be carefully chosen and that it would be sensitive to both the
crystal frequency and the ambient temperature.

When I was doing circuit desgin everyone most were still using TTL. The
74LS00 series was the hot item at the time.

Just from a logical viewpoint (pun not intended, by the way) I still think
he should try using two gates in a ring circuit.

Jim
N8EE

"Michael Black" wrote in message
...
PaoloC ) writes:
Hi.
I have spent part of the weekend trying to resonate al old CB XTAL at
its fundamental frequency.

The XTAL is labelled 27.125 MHz, with a fundamental of about 9.041 MHz,
which falls into 18m HAM band when multiplied by two. I assume 27MHz
XTALs are 3rd overtone....

Could the crystal be a receive crystal? CB crystals tended to show
the channel number or the frequency of the channel, and so if you simply
looked at the marked frequency, it would not tell you if it's for
transmit or receive. Though the ones I've seen did mark them with
"R" or "T". The point is that if it's a receive crystal, it wouldn't
be 1/3 of 27.125 but 27.125-IF and then divided by three.

IFs were frequently 455KHz

I have used fundamental crystals on their overtones and overtone
crystals on the fundamental. They are not exactly 1/3 or 1/5 ratio due to
parasitics in the crystal. All crystals have the fund and OT responses, but
the responses are optimized for the intended use - by changing some of the
construction. (I'm talking AT cuts)
Overtone crystals *tend* to be used in the SERIES resonant mode, though
not always. Fundamental crystals tend to be used mostly in the parallel
resonant mode. This will also cause additinal frequency error when using
the marked frequency. The parallel freq is higher than the series (if I
recall correctly)
If I recall, the two gate oscillator is a series resonant oscillator and
the one gate is parallel. I would use a transistor Colpits oscillator
myself. The digital gate oscillators can run the crystal at a higher drive
than it should causing more crystal heating than desired (more drift
w/time).
If you are going to use a OT xtal as a fund., I'd pick a parallel type
osc like the Colpits (I think it has another name when it has a crystal
rather than a coil). I never did like the digital gate oscillators. They
tend to be a *bruit force* oscillator. Because of the Rs of the crystal,
crystals of some frequencies are easier to over drive than others.
Unfortunately, I don't remember which is which.
The fact that your 10mMHz xtal works is a good start. The other one
probably will, but since it was made for the third OT, there may be enough
difference to keep it from oscillating without some circuit change.
If you have no scope, measure the current drain. There may be a change
when the xtal is inserted if it is oscillating. Also, you could try
measuring some voltage through a large resistor or choke to see changes when
the xtal is inserted.

--
Steve N, K,9;d, c. i My email has no u's.

"Michael Black" wrote in message
...
PaoloC ) writes:
Hi.
I have spent part of the weekend trying to resonate al old CB XTAL at
its fundamental frequency.

The XTAL is labelled 27.125 MHz, with a fundamental of about 9.041 MHz,
which falls into 18m HAM band when multiplied by two. I assume 27MHz
XTALs are 3rd overtone....

Could the crystal be a receive crystal? CB crystals tended to show
the channel number or the frequency of the channel, and so if you simply
looked at the marked frequency, it would not tell you if it's for
transmit or receive. Though the ones I've seen did mark them with
"R" or "T". The point is that if it's a receive crystal, it wouldn't
be 1/3 of 27.125 but 27.125-IF and then divided by three.

IFs were frequently 455KHz

I have used fundamental crystals on their overtones and overtone
crystals on the fundamental. They are not exactly 1/3 or 1/5 ratio due to
parasitics in the crystal. All crystals have the fund and OT responses, but
the responses are optimized for the intended use - by changing some of the
construction. (I'm talking AT cuts)
Overtone crystals *tend* to be used in the SERIES resonant mode, though
not always. Fundamental crystals tend to be used mostly in the parallel
resonant mode. This will also cause additinal frequency error when using
the marked frequency. The parallel freq is higher than the series (if I
recall correctly)
If I recall, the two gate oscillator is a series resonant oscillator and
the one gate is parallel. I would use a transistor Colpits oscillator
myself. The digital gate oscillators can run the crystal at a higher drive
than it should causing more crystal heating than desired (more drift
w/time).
If you are going to use a OT xtal as a fund., I'd pick a parallel type
osc like the Colpits (I think it has another name when it has a crystal
rather than a coil). I never did like the digital gate oscillators. They
tend to be a *bruit force* oscillator. Because of the Rs of the crystal,
crystals of some frequencies are easier to over drive than others.
Unfortunately, I don't remember which is which.
The fact that your 10mMHz xtal works is a good start. The other one
probably will, but since it was made for the third OT, there may be enough
difference to keep it from oscillating without some circuit change.
If you have no scope, measure the current drain. There may be a change
when the xtal is inserted if it is oscillating. Also, you could try
measuring some voltage through a large resistor or choke to see changes when
the xtal is inserted.

--
Steve N, K,9;d, c. i My email has no u's.

Hi,
thanks to all those who replied and discussed this topic, providing me
with a lot of good suggestions.

I will try a different ex-CB rock, just in case the one I picked is
broken. I actually tried a "Colpitts" oscillator with it and had no
success either.

73,
Paolo IK1ZYW

Hi,
thanks to all those who replied and discussed this topic, providing me
with a lot of good suggestions.

I will try a different ex-CB rock, just in case the one I picked is
broken. I actually tried a "Colpitts" oscillator with it and had no
success either.

73,
Paolo IK1ZYW

Get something working with a good crystal first.

A Google search (Colpitts Crystal oscillator circuits) turned up lots of
gate oscillators. This has the type of transistor oscillators I am used to
using:

http://hem.passagen.se/communication/txo.html

There are many other references at:
http://users.telenet.be/educypedia/e...osciltypes.htm


I don't like this circuit:
http://www.electronics-tutorials.com...scillators.htm

Get something working with a good crystal first.

A Google search (Colpitts Crystal oscillator circuits) turned up lots of
gate oscillators. This has the type of transistor oscillators I am used to
using:

http://hem.passagen.se/communication/txo.html

There are many other references at:
http://users.telenet.be/educypedia/e...osciltypes.htm


I don't like this circuit:
http://www.electronics-tutorials.com...scillators.htm

I dunn-o. I had great success with this cricuit. It's used in all the
older Motorola channel elements. Just about any rock will fly at the
fundamental unless its really bad. You can play with the ratio of the two
feedback caps (Base-to-emitter and emitter-to-ground) Base-to-emitter cap
decrease to get more feedback. I even modeled one of these on P-Spice.
Really neat to see it oscillate.

like Figure 7:

http://www.northcountryradio.com/PDFs/column007.pdf


--
Steve N, K,9;d, c. i My email has no u's.

"Steve Nosko" wrote in message
...
Get something working with a good crystal first.

A Google search (Colpitts Crystal oscillator circuits) turned up lots of
gate oscillators. This has the type of transistor oscillators I am used
to
using:

http://hem.passagen.se/communication/txo.html

There are many other references at:
http://users.telenet.be/educypedia/e...osciltypes.htm


I don't like this circuit:
http://www.electronics-tutorials.com...scillators.htm