Project: Crystal Oscillator, LC Oscillator and PLL help needed

[estelle.ros06@live.com]

Hi everyone.

I am new to electronics design and I will really appreciate some guidance with a project I am busy with.

Here is an overview of what I need to achieve:
1. Build a 2Mhz crystal oscillator circuit and filter out the second harmonic (which will be 4Mhz).
2. Build a 4MHz LC VCO circuit using varactors.
3. Use the phase comparator in the PLL chip to phase and frequency lock the two oscillators.

Issues:
1. I have constructed the crystal oscillator circuit, which oscillates at 2MHZ. When connecting the 4MHz bandpass filter, no waveform is seen up on the oscilloscope. Is this due to loading and should a buffer amp be used?
2. The LC oscillator only seems to oscillate at 1.5MHz when built on a pcb, although the component values are simulated for 4MHZ? Must a buffer amp be used as well? I am also not sure if the placement of the varactors is correct to provide tuning range that will be wide enough to ensure a definite lock condition.
3. Will the phase detector section work on the PLL chip if I input the two sine waves (xtal to signal in and lc to PD1)? If not,is there a bjt circuit be used to convert the sines to squares? The opamp circuits are too slow.

I have attached a hand sketch of my circuit (not the best in the world!)

Thank you in advance

Regards

Estelle

 

[betwixt]

There are some very suspicious values in there!

C10 probably should be 15nF not 15uF
C1 has no value but it should be about 1nF
C2 has Xc of ~17K so in conjunction with R5 you reduce the signal voltage by about 65%
R6 and C3 reduce it by a further ~20%
C4 should probably be 150pF
C5 should probably be 330pF
C9 is far too small to filter the PLL output pulses, something like 3.3uF would be more suitable
The varactors have no DC return so they will have no tuning effect
The LC ratio of the oscillator is too much L and not enough C. I suggest L should be ~25uH and C should be ~50pF (combine C7 & C8 into one)
You also need decoupling capacitors across the supply and after the regulator.

I'm not sure what kind of oscillator it is supposed to be. I would put the tuned circuit between the base and ground instead of base and collector.

Brian.

 

[FvM]

I have difficulties to relate both circuits to a known working oscillator topology. The colpitts respectively pierce oscillator circuits needs phase shift capacitors for stable operation.

 

[BigBoss]

There are many mistakes in especially varicap controlled oscillator.
There are 2 resonance point, varicaps are floating,phase shift ???

 

[D.A.(Tony)Stewart]

Since you are new it would help to learn to specify specifications for the design.
purpose, inputs, process & outputs
stability, jitter, capture time, SNR, tuning range, environmental drift or capture range.

There are dozens of implementations, which are not measurable without a spec.


Keep mind VCXO"s are more stable but have less tuning range than V controlled LC OSC and the the CD4046 also has a wide tuning range VCO internally with two different mixer types.

 

[mar0791]

for such frequency , hard construct bandpass filter of component RC . I suggest used coil and capacitor and construct resonance circuit , serial or parallel . Have you got waveform on C transistor ( quartz generator) ? How is voltage (amplitude) on collector?
with reference to varicap . Anodes must connect for DC to GND , by resistor , example 10k . Search you in google "generator LC with varicap"

 

[chuckey]

The first thing that is wrong is your 4MHZ "filter", replace it all by a parallel LC tuned circuit (tuned to 4MHZ) to earth. So remove C2,R5,R6 and C3. Then you might have enough RF for the PLL to lock up on , once the oscillator is sorted out.
Frank

 

[estelle.ros06@live.com]

Hi everyone -

Thank you for the valuable feedback. I really appreciate it. I did some modifications to the circuit (see attached) on Multisim, but still struggling to get the crystal oscillator into oscillation state (should be able after the weekend). I have not constructed the physical circuit yet, first want to get some feedback from the forum.

I have a few questions:
1. Will the varactor configuration allow for a tuning range of at least 0.5MHz? I have NTE610/612/618 that can be used.
2. When simulating the LC oscillator, the Vp-p reading as about 2.5V. Will this be sufficient to ensure proper detection and ultimately a lock state on the PLL?
3. Any comments on component values.

Thank you

Estelle

 

[D.A.(Tony)Stewart]

Hi everyone -

Thank you for the valuable feedback. I really appreciate it. I did some modifications to the circuit (see attached) on Multisim, but still struggling to get the crystal oscillator into oscillation state (should be able after the weekend). I have not constructed the physical circuit yet, first want to get some feedback from the forum.

I have a few questions:
1. Will the varactor configuration allow for a tuning range of at least 0.5MHz? I have NTE610/612/618 that can be used.
2. When simulating the LC oscillator, the Vp-p reading as about 2.5V. Will this be sufficient to ensure proper detection and ultimately a lock state on the PLL?
3. Any comments on component values.

Thank you

Estelle

It's a good attempt but puzzling topology.

To get the second harmonic of an oscillator, it has to be distorted.
Square and sine waves don't have much 2nd harmonic. ( in theory , none)
To generate a 4MHz PLL output, all you need is to use the built in VCO and a /2 Flipflop to mix both 2MHz signals and use the type II phase comparator charge pump with Cap to control VCO and use the lock detector built in.

A better approach is tune the Xtal to a standard with ac coupled varicaps. The 62 pF caps are for older 35pF rated Xtals. Most now are 15~20pF load rated.

broken link removed

No need for any LC filters , LC tunable oscillators..

YOu can even run the 74HC4046 up to 12MHz at 5V and /6 to create 12MHz locked to the 2MHZ crystal and other possible dividers.
To get a sine wave out then use a BPF.