EMRFD Message Archive 4664

Message Date From Subject
4664 2010-05-08 00:41:33 ha5rxz Vackar Crystal Oscillator
I have built a Vackar oscillator, right now it is constructed 'ugly' style (see previous posts) but once I get the component values right it will be tidied up and fitted into an existing project. As a fixed 2 MHz oscillator it works very well but problems occur when I try to turn this into a crystal oscillator, it stops.

A quick look at the schematic for a Vackar http://www.qsl.net/va3diw/vackar21.jpg shows that the coil and two capacitors form a low-pass filter in the feedback path. So far I have tried replacing each of the capacitors with my 2 MHz crystal and I have also tried wiring the crystal across the coil. In each case output ceases.

How do I make a Vackar crystal oscillator?


Note: The component values in the diagram do not oscillate at 2 MHz, please ignore them.
4665 2010-05-08 03:02:45 ad5rj_slash_7 Re: Vackar Crystal Oscillator
What happens if you replace the emitter bypass capacitor with the crystal?

4666 2010-05-08 13:05:48 ha5rxz Re: Vackar Crystal Oscillator
The oscillator still runs but with the crystal wired across the emitter bypass resistor I don't think it is contributing to frequency stability. The emitter bypass capacitor is quite large and the series capacitance of the crystal (unknown) would be measured in picofarads.


4667 2010-05-08 14:57:13 victor Re: Vackar Crystal Oscillator
HA5RXZ, you did not understand what AD5RJ asked you to do. Take OUT the emitter decoupling capacitor and connect the xtal in its place. This way you will have 3900 ohm resistor in parallel with the xtal Co capacitance of 5pf, which will lower markedly the amplifier gain at all frequencies but in the xtal series resonance the xtal will show a few ohms resistance, so the amplifier gain will be high at this frequency and so allow oscillations at the xtal series resonance.
Victor - 4Z4ME

4668 2010-05-08 23:23:33 ad5rj_slash_7 Re: Vackar Crystal Oscillator
The intended circuit behavior is as Victor has described, but the resistance of a 2 MHz crystal at series resonance will be more than a few ohms (perhaps
4672 2010-05-11 09:09:35 ha5rxz Re: Vackar Crystal Oscillator
I just tried removing the emitter bypass capacitor and replacing it with the crystal and got some very strange results. The oscillator kept running but I am not sure that the crystal is aiding stability. Over thirty minutes I have seen 237 Hz of drift and it's still moving upwards. Yes, the crystal is OK as it was tested in a Colpitts circuit before this.

Someone else off-board suggested replacing the series feedback capacitor with the crystal. This might be a better option as the crystal would act as a bandpass in the feedback path but the instructions for setting up the Vackar state that the feedback capacitor should be 'adjusted for minimum value that gives oscillation'.

Any further thoughts?

Peter HA5RXZ

4677 2010-05-12 06:20:47 ad5rj_slash_7 Re: Vackar Crystal Oscillator
It may be time for the real experts to take a turn at this as I have only a few additional thoughts:
1. Prove the crystal is controlling feedback (Remove the crystal and oscillations should stop if the bypass capacitor has not been replaced)
2. The drift may be caused by excessive power dissipati
4683 2010-05-14 15:07:29 davidpnewkirk Re: Vackar Crystal Oscillator (long)
4685 2010-05-15 03:41:15 ha5rxz Re: Vackar Crystal Oscillator (long)
Well, thank you for the detailed reply Dave. Over the last two weeks I have done hours of research into Vackar oscillators including getting a friend to dig around the IEEE library and it appears that NOBODY has made a Vackar crystal oscillator. I am therefore carrying out original research 8-)

I was first attracted to the Vackar because of the low phase noise and low drift (even as an LC circuit) but if a Vackar crystal oscillator is impossible then I may have to look elsewhere. All suggestions for a stable low phase noise oscillator working at about 2 MHz will be welcome.


4686 2010-05-15 03:55:52 Roelof Bakker Re: Vackar Crystal Oscillator (long)

Maybe a bit over the top, but this might be of interest to you:


and also:


Roelof Bakker, pa0rdt
4687 2010-05-15 04:04:39 Tim Re: Vackar Crystal Oscillator (long)
4688 2010-05-15 10:11:18 ha5rxz Re: Vackar Crystal Oscillator (long)
Q Rather than invent a new kind of oscillator which has more parts and is less stable than a simple LC VFO... why not just use an LC VFO?

A It's going to be used as the reference in a 400 MHz synthesizer. Any phase noise or drift will be multiplied by 200 so this oscillator needs to be good.


4689 2010-05-15 21:08:55 ajparent1 Re: Vackar Crystal Oscillator
That looks like the schematic from the HAM RADIO Article some 38 years ago.

First item, look for the ham radio article on crystal oscillators.
The Vacker topology can be used as a crystal oscillator but the coil goes and the crystal is fitted there. The values are adjusted accordingly to match the required crystal load capacitance for series

However at 2mhz the Pierce osc is good if the values are right
and the stability in that case will be generally based on the crystals temperature coefficient.

This is covered in the book EMRFD.

4690 2010-05-15 21:28:06 ajparent1 Re: Vackar Crystal Oscillator (long)
4691 2010-05-15 22:15:33 Tayloe Dan-P26412 Re: Vackar Crystal Oscillator (long)
If the goal here is ultra low noise phase performance, you could try
another trick. I have heard of using a crystal filter to strip out the
sideband noise of a fixed frequency oscillator to make a cleaner test
source for the purposes of blocking or IP3 tests. It seems like it
would be fairly straight forward to build a conventional crystal
oscillator (like 10 MHz) and clean it up with a tight 10 MHz crystal
filter. That seems like it could give you outstanding sideband noise
performance as you multiply it up.

- Dan, N7VE


4692 2010-05-16 06:37:22 victor Re: Vackar Crystal Oscillator (long)
I designed and used a crystal oscillator that used a 25MHz crystal oscillating at its 5th overtone. Its 125MHz output was taken from the crystal current so that it was filtered by the crystal.
It was used as a low phase noise reference clock for a DDS generator.
I posted its schematics and picture in Files directory under my name: "Victor 4Z4ME"

Victor - 4Z4ME

4693 2010-05-16 14:12:51 w1kilofoxtrot Re: Vackar Crystal Oscillator (long)
I never ever had a problem with my 753. Never a drift, never chasing someone up and down the band. Of course, I almost never turned it off! So, it had plenty of time to reach thermal equilibrium. Power was cheaper back then, tho...

4694 2010-05-17 07:25:01 Ashhar Farhan Re: Vackar Crystal Oscillator (long)
Just a reminder. The first (according to my foggy memory, and i am on
a train right now hence no access to my QSTs) transistor vxo written
about my W1CER (later W1FB) was actually a vacker.
It was also reprinted in ssd. So, i guess you might as well refer to
the article.
- farhan VU2ESE

On 5/16/10, victor <vkoren@actcom.net.il> wrote:
> I designed and used a crystal oscillator that used a 25MHz crystal
> oscillating at its 5th overtone. Its 125MHz output was taken from the
> crystal current so that it was filtered by the crystal.
> It was used as a low phase noise reference clock for a DDS generator.
> I posted its schematics and picture in Files directory under my name:
> "Victor 4Z4ME"
> Victor - 4Z4ME
4697 2010-05-17 13:21:00 john lawson Re: Vackar Crystal Oscillator (long)
Ashhar, your comment inspired me to post the following as there has been a lot of comments on the Vackar Xtal Osc, etc.
If you go into the archives of QST you should find the following article from Sept. QST, 1978, Page 15. "Meet the Remarkable but Little Known Vackar VFO!"  by Floyd E. Carter, K6BSU. It also appears in the ARRL book, QRP Classics for those how can lay their hands on a copy of the book...The VFO has an output at 7 MHz. Of course, copies of the article can also be obtained from the Technical Dept of the ARRL as a last resort........John, K5IRK