Oscillator circuits 1

[LezBurns]

Hello. Can anyone help me I have a bit of a proble. My electronics skills are quite limited but I have a grasp of the basics.

I want to create a high frequency oscillator circuit in the range of around 10MHz. But I also need to be able to affect the frequency of the oscillator using an external varaible capacitor. Can I just buy any normal crystal osc oand if so how would i go about making my addition.

Any manafacturers part numbers and details of how this can be achieved would be very useful.

Many thanx.

 

[cbarn24050]

Hi, a crystal is a fixed frequency, what you need is an LC or a RC oscillator. Search for "wayne bridge" or "collpits" oscillators to get you started, there are numerous others as well.

 

[BrianR]

You don't say how much frequency adjustment you want but could pull the frequency of a 10MHz Xtal osc by 1-2KHz at the risk of it dropping out of oscillation. This is done with a series variable inductor or a variable C.

A Hartley or Colpitts circuit is probably the best for a variable frequency oscillator but keeping the frequency accurate and controlling 'drift' is quite a challenge. I've never seen a Wein bridge used at 10MHz though.

 

[BrianG]

As stated in previous posts, you don't say by how much you want to adjust the frequency. Also, you don't say whether you want a sine wave output or a logic-level square wave.

"Normal" packaged oscillators of either sine or square output generally have a buffer stage to prevent - or at least reduce - the "pulling" effect of changes in load and power supply, so you won't get much scope for adjustment there.

If you build your own oscillator with a separate crystal it is possible to adjust its freqency by a few percent with a variable capacitor. Since the crystal is tuned by the manufacturer to oscillate with a specific load capacitance, small variations can usually be made without stopping oscillation. The variable capacitor can be an external trimmer capacitor, or electrically adjusted with a control voltage applied to a varactor diode.

 

[LezBurns]

Firstly i would like to thank you all for your replies, they are much appreciated.

Ok I need the frequency to be altered as much as possible really with only very small adjustments in capacitance.

My aim is to produce a capacitance based micro-sensor that can monitor the movements of a human finger. As the finger moves toward or away from a fixed plate capacitor it will affect the frequency of an oscillator. The frequency of the oscillator is then feed into a phase lock loop chip to get a voltage out that I can then calibrate into movements.

The main problem is that I need really small movements to be monitored. In the range of 0.1mm.

Am I best of with a LC or RC oscillator and how can I test mathmatically which will produce the best frequency change over small capacitance changes. Also do I need to add the variable capacitor in series or in parallel with the other components?

Thanks again for your help.

 

[cbarn24050]

Hi, your scheme is pretty much unworkable. A better approach would be optical detection, lots of workable methods there.

 

[jimkirk]

Ignoring the detailed circuit topology for a moment, an LC oscillator frequency will vary as one over the square root of the capacitance, while an RC oscillator varies as one over the capacitance, so mathematically, an RC oscillator should given you more d(Frequency)/d(Capacitance).

One down side is that there's generally more phase noise with an RC oscillator.

Parasitics must also be considered; with the finger removed, the capacitance will not go to zero.

Can you build a test jig so you can measure the range of capacitance you'll be dealing with?

 

[logbook]

Frequency can be measured with arbitrary accuracy for your purposes. It would be relatively simple to measure to better than 1 part in 1E6 for example. Although it is true that the RC oscillator is linear with capacitance, the capacitance will not change linearly with position. The hard part of this problem is going to be calibration of the position. The resolution will change with distance from the sensor, decreasing as you get further away.

0.1mm resolution is not a problem if the total range is say 1mm. If you are trying to get to 5mm or 20mm then the scheme will probably become unworkable quite rapidly.

 

[ScottyUK]

Hi cbarn,

Correct spellings of the oscillators you mention are:

Wien Bridge Oscillator

Colpitts Oscillator

Might help someone find 'em on Google or wherever!

 

[rodar]

HI
To me the primary factor is how much percent change in
capacitance happens for a movement of 0.1mm. If there is
say 10% capacitance change it is dooable.
By the way motorola just came out with a new chip for
doing e-field capacitive sensor type stuff.
Search online for E field , motorola
You must shield the sensor carefully to eliminate capacitance changes from outside the sensor interfere.
good luck
Rodar

 

[BrianR]

Checkout a musical instrument called a Theramin which is an audio oscillator whose frequency varies as you move one hands near 2 capacitor plates and the volume changes as you move the other hand near other plates. They have been published as kit projects in many electronics magqzines over the last 20 years.

 

[itsmeRICK]

Touch switch or finger tracking device?

We used a 74HC14 connected as a schmitt trigger, driving the counter input of a Philips flash 8051 controller and monitored the frequency shift. The oscillator ran between 100 kHz and 250 kHz. The frequency depneded heavily on circuit layour and the Schmitt trigger's hysteresis.

The smallest detectable change was a few tens of femto-farads - we could sense a hand with an approaching finger 10 to 20 cm away.

The circuit and some of the operation is explained U.S. patent 6,583,676 which can be searched for a

http://www.uspto.gov

There are lots of fine points, most of them aimed at keeping the oscillator purely dependent on the capacitance, so the pickup plate has to be shielded from noise sources (like the rest of the digital circuitry, and in one case, a flourescent lamp backlight), and the power supply has to be kept very clean.

 

[Warpspeed]

There are a few other things to think about as well........

First the finger is quite likely to induce a lot of 60Hz (50 Hz ?) mains hum into your circuit which might cause trouble. Also high voltage electrostatic discharges might possibly damage or destroy components.

Its the old zap from the doorknob after walking on the synthetic carpet syndrome. While many simple circuits can be made to work, keeping them working, or mass producing them might not be so easy.

Touch circuits are quite commonly used in lifts, and other vandal proof applications. What is it you are trying to do exactly ?

 

[atferrari]

Lez,
I recall reading something describing the pad used in laptops to mimic the mouse action.

Suggest to try that in Gooogle or similar.

My experience with one laptop in Equatorial Guinea: high ambient humidity created the hell of havoc to the pad. Just in case, keep it mind. Also the opposite: too dry enviroment (static?) had similar effect but less pronounced.

Buena suerte. Feliz Año Nuevo

Agustín Tomás

 

[LezBurns]

Can anybody help ive have done some calculations and my capacitance change is going to be in the range of 0.5pf to about 3pf where do i go about getting a wien Bridge oscillator that is affected by this amount of change quite drastically so the frequency ouput changes accoridingly.