Mini DIY signal generator

[merlinb]

Finally got this into a proper enclosure.
PCB, schematic and other files are here:
http://s81.photobucket.com/albums/j207/merlinblencowe/Sine%20Wave%20Generator/

[iccaros]

Good Ideal, Did I miss the schematic for this change?

[Perrow]

Mind posting the graphics and/or drill plan? Looks like something I'll have to build.

[merlinb]

Mind posting the graphics and/or drill plan? Looks like something I'll have to build.

Sure, here's the graphic. Dimensions 55 x 105mm

http://dl.dropbox.com/u/57831278/SigGen55x105mm.bmp

[PRR]

What's a Cambridge configuration? I understand the principles of a Wien Bridge oscillator, but the Cambridge thing is new.

The Wein Bridge plus one diff-amp makes an oscillator; but to change frequency you must vary two parts or face severe gain-change trouble.

IF you allow two amplifiers, specifically a paraphase (one non-invert one invert common input), there's nine ways to connect the 2Cs and 2Rs and get an oscillator. Two of these allow large frequency change with one part without gain trouble.

http://i.imgur.com/rkmIF.gif

With other multi-amplifier configurations you can get "An Abundance Of Sinusoidal Oscillators" (R.S. Sidorowicz, Proc IEE (London) Mar 1972). This Cambridge/Merlin configuration is probably touched-upon in there (it's dense).

[PRR]

>>> W. Cambridge published a modification for the Wein bridge
>> Cite? Hint?
> The source I have is the Newnes Circuit ideas Pocket Book Part Three, p258. Year unknown, probably early 90s.

I found the book. Very-very interesting. All these technical projects and then "Guard the vegetables without frying the cat".

Newnes Circuit Ideas Pocket Book
ISBN: 0750623365 / 0-7506-2336-5
Publication Date: September 1995

(FWIW, I can't find any "Part Three".)

I paid $16 on ABE for a copy shipped from the UK. Offers on this side of the pond were $20++. But on ABE today I'm seeing only $40 (to $235!) while Amazon shows five at $25 delivered in USA.

W A Cambridge's 'Circuit Idea' does not include any references.

http://i.imgur.com/XPUfBCv.gif

[Thunk]

Pardon my ignorance, but why do you use an LDR instead of just a 100k resistor?

I know this thread is old, but it looks like a fun project and I'm curious....


Thanks!

[PRR]

> why do you use an LDR instead of just a 100k resistor?

Gain around the loop must be *exactly* 1.0000......

If less, it won't sing.

If more, it tries to build-up to infinity, hitting the rails and distorting.

(You know this from waving a microphone around a PA speaker. There's really no stable point below "howl".)

Yes, there are simpler ways. Such as a 110K resistor (gain of 1.1) shunted with diodes and a 500K resistor (ultimate gain of 0.9). When the diodes barely conduct, the wave has a faint flat-top, average gain hovers at 1.0000. You can even build a useful oscillator with initial gain of 1.5 and let it bang itself against a grid and de-bias down to average gain of 1.000. While there is gross distortion inside the loop, sometimes an optimum choice of output point gives acceptably low distortion.

Merlin choose to use the better way. A level detector, a smoothing filter, and a gain control. Filament lamps and FETs are classic gain controllers. The opamp and LDR do the same thing and, in this case, quite neatly.

[R.G.]

Yes, there are simpler ways... Filament lamps and FETs are classic gain controllers. The opamp and LDR do the same thing and, in this case, quite neatly.

As an interesting (to me, anyway) is that the variable resistance of a filament lamp was the subject of Bill Hewlett's graduate work at Stanford. He and Dave Packard decided to start an electronics company, the first product of which was the HP 200 sine wave oscillator. This was a Wien bridge oscillator that used a filament lamp for keeping gain precisely at the oscillation point.

The garage they worked in is viewed as the birthplace of Silicon Valley - which itself has now passed into history, I fear.

See http://www.hpmuseum.org/garage/garage.htm

[Thunk]

Thanks for the replies!

Now, I just need to find an LDR and then to get started...

[PRR]

> Bill Hewlett's graduate work at Stanford.

Was probably Prof/Dean Fred Terman told him what to try. Terman was a giant. I have his main book in my bedroom.

> an electronics company, the first product of which was the HP 200 sine wave oscillator.

There were a couple of "200"s. The early ones were black rack-boxes, sales were slow. What made it the Must-Have! of 1950 is probably the sleek case on the 200AB. (You yourself know how packaging influences sales.)

> keeping gain precisely at the oscillation point.

What's also interesting is that the lamp does NOT work alone. It *needs* nonlinearity in the amplifier. 200AB got it with mis-bias on the 2nd stage. I foolishly "fixed" that, and it wouldn't stabilize. Similar experience with a Heath transistor brige-Tee singer. Once found a tech paper proving the same thing (can't find it again).

[merlinb]

>> Cite? Hint?
> The source I have is the Newnes Circuit ideas Pocket Book Part Three, p258. Year unknown, probably early 90s.
I found the book. Very-very interesting. All these technical projects and then "Guard the vegetables without frying the cat".

Today I stumbled on this book which has a very interesting collection of single-element-control versions of the Wein bridge oscillator:
https://archive.org/details/CircuitDesigns3CollectedCircards

[Morocotopo]

I have a DIY sine generator that uses a lamp for stabilizing the gain, it´s funny to watch in the oscilloscope how, when you change frequency, the signal "bounces" up and down till it stabilizes. Uses two batteries, though. A hint for getting exact frequency, since the pots/case will have no precise markings: generate a sine in some audio soft of the frequency you want, loop it, start the gen and tune it as you would tune a guitar string against another one! We are all musicians that know how to tune, right?



Nice project Merlin, I´ll give it a try.

[deadastronaut]

Finally got this into a proper enclosure.
PCB, schematic and other files are here:
http://s81.photobucket.com/albums/j207/merlinblencowe/Sine%20Wave%20Generator/

i love the look/style of this ...its got that new but old classic look to it, the dial graphics are excellent..very

[merlinb]

i love the look/style of this ...its got that new but old classic look to it, the dial graphics are excellent..very

Thanks! It was inspired by this:

[tca]

There is a Wienbridge oscillator (1kHz) at the LM386 datasheet (pg. 6), if someone would like to play with it

[Seljer]

I recently made a pretty clean sinewave generator by heterodyning two crystal oscillators against each other. I used two identical crystals and by adding extra capacitance in series with one the crystals I can detune it to about 1khz away from the other one (I used and on-off-on switch for three fixed settings). For a frequency mixer I used a simple 1n914 diode, followed by a low pass filter, leaving me with about 200mV peak to peak of relatively clean sinewave. One NPN transistor is used for each oscillator, and another NPN is used as a buffer after the low pass.

It's an entertaining approach to the final goal 

[PRR]

> I stumbled on this book which has a very interesting

Wow! Yes, a great collection of basic cook-book blocks.

Thanks for posting.

[newperson]

"Guard the vegetables without frying the cat".

Newnes Circuit Ideas Pocket Book
ISBN: 0750623365 / 0-7506-2336-5
Publication Date: September 1995

What is this circuit about?  Some sort of replacement electric fence?

[PRR]

It IS an electric-fence charger, home-brew.

Supposed to "deter squirrels without excessive violence". However it runs straight 230V past an unspecified gas tube, SCR, 1uFd cap, to a 6V/12V ignition coil, and you do something unspecified with the sparky end of the coil.

On further study, it scares the piss out of me.

If you need electric fencing, Tractor Supply et al will sell you a low-power ""SAFE"" fence charger. As with any electric shock, results can be unpredictable.