simple signal generator w/ variable frequency????

[Boofhead]

A wein bridge oscillator can be built with one opamp and a small light bulb (~20mA to 40mA) (which doesn't glow - it's used as a resistor).  The output is a sine-wave, the frequency can be varied with a pot, or switched with a multipole switch - the output quality is very good.

There's another version which uses a JFET instead of the light bulb.

You will find one on the web no problem.

[gez]

There's a really nice pseudo Wien bridge circuit in the March issue of Everyday Practical Electronics.  The auto gain control is done with a single transistor and a couple of diodes - very clever! - and has a low parts count, apparently low distortion (haven't tried it yet) and low current consumption.  It's all transistor, but could easily be re-jigged using an op amp.

www.epemag.wimborne.co.uk

[gez]

I have a book called practical electronics for inventors, and it has an oscilator similar to the one your posted a couple months ago, but as far as i know you cant really get a sine wave out of it with out adding annother stage.

You CAN add another op-amp stage to integrate the triangle to get a sine, and in my opinion it's worth it as the quality of the wave is surprisingly pure.  The problem is keeping the amplitude stable over the frequency range you select, but I've had good results using a dual-gang pot with resistors wired in parallel here and there.  

It's the sort of thing you need a scope for to set it up right, but I've designed some pretty mean LFOs using this method!  :twisted:

[Boofhead]

gain control is done with a single transistor

Be interesting to see if it is really low distortion.

[Paul Perry (Frostwave)]

You CAN add another op-amp stage to integrate the triangle to get a sine,

..I don't doubt you have a nice sine, but it isn't coming from integrating a triangle, or I wasted my time in that trig class :roll:

bugger, just found this link http://www.discovercircuits.com/O/o-sine.htm
now I won't get any sleep  :shock:

[gez]

Paul

I first found out about integrating a triangle to get a sine in ‘Practical Oscillator Circuits’ by A Flind.  He mentioned that the sine is surprisingly good and he’s not bloody wrong! - I’ve got extremely pure sine waves using this approach!  

He uses a op-amp and 7556 for the triangle generator then feeds the output to another op-amp wired as an integrator.  His version is fixed frequency but uses a input resistor to control the amplitude of the sine.  I did an all op-amp version and used one half of a dual gang pot (plus a resistor from wiper to outer lug) instead of a input resistor, to keep the amplitude stable over the whole range of frequencies.  The other half of the pot is used for the frequency control, and the whole thing is wired up so that as it’s resistance increases (slower frequencies) the input resistance of the integrator increases, thereby keeping the amplitude of the sine constant.  It works a treat and is remarkably stable (used a regulator to power everything).

At the very lowest frequency (less than 1Hz) there’s some triangulation of the signal granted, but only for a miniscule portion of the pot’s travel, once you’re into useable range the sine wave is as good as the one on my function generator - I was amazed when I scoped this!

The beauty of this circuit over Wien bridge is that it allows for pulse-width modulation, plus you get a triangle and square wave output.

[gez]

Be interesting to see if it is really low distortion.

It's the diodes that do the job of keeping distortion to a minimum not the trannie.  

I hope it does what it says on the tin as it's a nice little circuit, ripe for modding! (and not a dual-gang/trim pot in sight!)

[Paul Perry (Frostwave)]

thanks for the explanation gez! I'v been making my sines from triangles by overdriving CA3080s, good for 2% distortion but that is about it (of course for a LFO, or fo normal audio testing a 2% distortion sine is no problem).