Test oscillator design
- Thread starter mcs
- Start date
Help Support GroupDIY Audio Forum:
bcarso
Well-known member
That Electronics World one is kind of clever. It is using a variety of biquad with all-pass sections instead of integrators, a circuit that a few have independently discovered (Steve Dove for one---see the Consoles article in Ballou 1988) and is attributed to Moschytz in the giant Circuits and Filters Handbook of Chen's. It's easier on the opamps from a stability factor standpoint, although there is a disadvantage from the standpoint of common-mode distortion.
It might have been helpful if the author had specified an optocoupler rather than just stating that the photoresistor is "common."
Another departure for the EW circuit is making a compensated signal to drive the particular variety of full-wave rectifier.
Still no square wave output though. I guess I would see if I could get along, for the square wave, with 18V peak-to-peak and use CMOS 4000-series gates in parallel, fed from a reduced supply of +/- 9V. A little short of your +26dBu but not too bad and certainly simple. Drive the inputs from a comparator like an LM393 with a stiffish pullup R to +9V and a tap so that when the comparator output goes to the negative rail it goes to -9V, so you don't overload the inputs of the gates.
Put a teensy bit of positive feedback around the comparator so it doesn't chatter at the threshold.
If you really need a full swing square wave then somehing needs to be contrived with a few more parts. I made a generator for testing fast amps like that, that did a +/- 20V output swing with about 20nS rise/fall times. It was a fixed-frequency closely-symmetrical square wave though, and not part of a sine wave generator.
Further note: you might want to provide for ways of turning the comparator etc. off when using the sine wave---the fast swings will tend to contaminate the sine waveform.
It might have been helpful if the author had specified an optocoupler rather than just stating that the photoresistor is "common."
Another departure for the EW circuit is making a compensated signal to drive the particular variety of full-wave rectifier.
Still no square wave output though. I guess I would see if I could get along, for the square wave, with 18V peak-to-peak and use CMOS 4000-series gates in parallel, fed from a reduced supply of +/- 9V. A little short of your +26dBu but not too bad and certainly simple. Drive the inputs from a comparator like an LM393 with a stiffish pullup R to +9V and a tap so that when the comparator output goes to the negative rail it goes to -9V, so you don't overload the inputs of the gates.
Put a teensy bit of positive feedback around the comparator so it doesn't chatter at the threshold.
If you really need a full swing square wave then somehing needs to be contrived with a few more parts. I made a generator for testing fast amps like that, that did a +/- 20V output swing with about 20nS rise/fall times. It was a fixed-frequency closely-symmetrical square wave though, and not part of a sine wave generator.
Further note: you might want to provide for ways of turning the comparator etc. off when using the sine wave---the fast swings will tend to contaminate the sine waveform.
Why do you even want a sine-wave?
Most scoping works better with a triangle. You can't see if a sine is bent nearly as easily as you can see a bent triangle. Clipping is instantly obvious.
And a tri-wave is dead-simple to generate. (OK, not as simple as it looks, but simpler than a really-clean sine.)
For frequency response, yes you do need a sine, but it needn't be hyper-low THD. IMHO, those lamp-stabilized things don't work. Yes, I know thousands of audio oscillators with lamps have been sold. The ones I have gotten intimate with would not stabilize unless there were some other nonlinearity beside the lamp. And in that case, why not just hang a Zener on it?
+29dBm???? (+26dBu into 300Ω) That's a power-amp, not a precision signal source. You don't want to generate a Watt and then pad it down. Heath did that; but H-P generates 10V at 10K and then runs a couple 6F6 push-pull to generate +30dBm in 600Ω (and 2 or 3 watts in lower loads). Find an idle but clean 40-Watt speaker amp and put a 30Ω 10W resistor in series with the output: you will get your +26dBu without strain into any line input, and dead-shorts won't hurt it. I actually added a 5-Watt booster to my Heath so I could run speaker tests without dragging out a power amp, but that was in days when there were not decent power amps in every garage and attic.
Is the old 8083 function generator chip still around? The sine was glitched but usable, and the triangle and square waves were there and clean. One-resistor sweep, too.
Most scoping works better with a triangle. You can't see if a sine is bent nearly as easily as you can see a bent triangle. Clipping is instantly obvious.
And a tri-wave is dead-simple to generate. (OK, not as simple as it looks, but simpler than a really-clean sine.)
For frequency response, yes you do need a sine, but it needn't be hyper-low THD. IMHO, those lamp-stabilized things don't work. Yes, I know thousands of audio oscillators with lamps have been sold. The ones I have gotten intimate with would not stabilize unless there were some other nonlinearity beside the lamp. And in that case, why not just hang a Zener on it?
+29dBm???? (+26dBu into 300Ω) That's a power-amp, not a precision signal source. You don't want to generate a Watt and then pad it down. Heath did that; but H-P generates 10V at 10K and then runs a couple 6F6 push-pull to generate +30dBm in 600Ω (and 2 or 3 watts in lower loads). Find an idle but clean 40-Watt speaker amp and put a 30Ω 10W resistor in series with the output: you will get your +26dBu without strain into any line input, and dead-shorts won't hurt it. I actually added a 5-Watt booster to my Heath so I could run speaker tests without dragging out a power amp, but that was in days when there were not decent power amps in every garage and attic.
Is the old 8083 function generator chip still around? The sine was glitched but usable, and the triangle and square waves were there and clean. One-resistor sweep, too.
This guy has been running on my bench for about 7 years:
http://www.ele.uva.es/~jesus/datasheets/exar/xr-8038a.pdf#search='XR8038'
If you want to go the "Science Fair" route....
There's a calculus lesson in there somewhere.
http://www.ele.uva.es/~jesus/datasheets/exar/xr-8038a.pdf#search='XR8038'
If you want to go the "Science Fair" route....
There's a calculus lesson in there somewhere.
It's a good chip. The only problem I had was frequency drift due to tempeture and the cap. But I needed 60.00 cps.
As soon as I changed the cap to polycarbonate, it was steady as a rock.
EXAR is right down the street. I might be able to get a sample if they rip you too bad.
cj
As soon as I changed the cap to polycarbonate, it was steady as a rock.
EXAR is right down the street. I might be able to get a sample if they rip you too bad.
cj
The 8038 is a good waveform generator, but the sine wave is far from ideal. It is derived from soft-clipping the trianlgle wave - and distortion dosen't come under some 0.5%
So it's no good for distortion analysis, but otherwise a great little chip..
Jakob E.
So it's no good for distortion analysis, but otherwise a great little chip..
Jakob E.
BYacey
Well-known member
I have been using an 8038 based circuit for years - for general sweeping and testing it's great. For instances where I need a cleaner sine I fall back on an old Hewlett Packard generator. These can be had for little money on ebay.
NewYorkDave
Well-known member
The Wien bridge with lamp stabilization is a classic. I think that scheme can be credited to William Hewlett of H-P. Here's a super-simple, fixed-frequency version I built to replace a dead master tone generator in our TV studio:
schematic
The circuit you posted is a good implementation; I would build it myself if I didn't already have some good H-P oscillators.
No lab should be without a low-distortion sine oscillator, and "function generators" just won't do that job. You'll need to add a power output stage if you really need to go up to +26dBM and beyond.
schematic
The circuit you posted is a good implementation; I would build it myself if I didn't already have some good H-P oscillators.
No lab should be without a low-distortion sine oscillator, and "function generators" just won't do that job. You'll need to add a power output stage if you really need to go up to +26dBM and beyond.
I used the triangle wave last night to tweak the LA2 for less distortion. Can't believe I have been missing out on this. Where the heck have I been?
Works soo much better than a sine wave for clipping points.
:thumb:
Works soo much better than a sine wave for clipping points.
:thumb:
bcarso
Well-known member
Good for clipping, but if you have any kind of freq dependent phase shift it will screw up the waveform.
mhelin
Well-known member
You can build a triangle/square VCO using a single LM13600/13700.
See page 16, figure 15:
http://www.national.com/ds.cgi/LM/LM13700.pdf
See page 16, figure 15:
http://www.national.com/ds.cgi/LM/LM13700.pdf
Then there is that xr2206 chip for quick and dirty job, it is still available I believe. But don´t know if it is good enough, I only have done a drum synth with it years ago, I have the chip somewhere in the attic but where... There was long and interesting article on Elektor mag years ago, how to squeeze the best signals from it. All I can remember sine/triangle switch was replaced with mosfet transistor as switch
Don't forget to check evilbay for some generators already in da box.
AMZ-FX
Well-known member
Yes, the 2206 can do a better job than quick and dirty... I have a test oscillator on my bench using that chip and it is better than the 8038. I would post a schematic but I designed it myself and don't know if I still have a copy.
There is a version in one of the Encyclopedia of Electronics books. I'll dig it up and post if you are interested.
regards, Jack
bcarso
Well-known member
A still lower distortion with fast response to frequency changes can be achieved by using good squaring circuits off of two of the outputs that are in quadrature, and summing the results and filtering minimally before driving some variable gain element like the shunt FET.
clintrubber
Well-known member
FWIW: so far I've been using stuff that I had already around: sinewave generator of modest THD-performance vastly improved by a parametric EQ tuned to the frequency of interest.
In general, for fixed frequencies some simple filtering obviously can improve things a lot.
A dedicated lo-THD is sure less hassle (more compact), it's on my to-solder list.
In general, for fixed frequencies some simple filtering obviously can improve things a lot.
A dedicated lo-THD is sure less hassle (more compact), it's on my to-solder list.
chrissugar
Well-known member
If you need only low dist sine wave and you have a quality soundcard there is always the software alternative.
There are some very good low dist generators, one of them is the SG2102 from Dazylabs.
Unfortunately the Dazylab page do not exist anymore but I found a page where the oscillator and the spectrum analyser is uploaded, so you should download it before it dissapears:
http://home.hetnet.nl/~elektronicaenzo/
Also lots of other audio measurement applications.
chrissugar
There are some very good low dist generators, one of them is the SG2102 from Dazylabs.
Unfortunately the Dazylab page do not exist anymore but I found a page where the oscillator and the spectrum analyser is uploaded, so you should download it before it dissapears:
http://home.hetnet.nl/~elektronicaenzo/
Also lots of other audio measurement applications.
chrissugar
chrissugar
Well-known member
I discovered the Dazylab page but there are only four application on the page, there were 21 measurement and design programs.
http://www.dazyweblabs.com/shannonsoft/page3.html
Look at the SGOne function generator.
http://www.dazyweblabs.com/shannonsoft/SGOne.zip
chrissugar
http://www.dazyweblabs.com/shannonsoft/page3.html
Look at the SGOne function generator.
http://www.dazyweblabs.com/shannonsoft/SGOne.zip
chrissugar
Mark Burnley
Well-known member
My "basic" sig gen was an 8038-based unit for about 10 years. It's a great chip for getting the basic square and triangle waves, but as mentioned the sine isn't very pure. It's useful as well in that the frequncy can go very low- ideal for slow-clocking of digital circuits to check for glitches, and even for clocking analogue synth clock-inputs!
Datasheet
Extra Info
Mark
Datasheet
Extra Info
Mark
