Low-distortion Audio-range Oscillator

[dirkwright]

Originally Posted by LAJ

try the XR2206 waveform generator chip. They will operate off one 9V battery but they usually like 12-18V range.
Also it gives triangle and square wave outputs.
You can usually find them on EBay.

Thanks. That chip has 0.5% sine wave distortion. I think the whole point of this thread is to find an extremely low distortion sine wave oscillator.

It would be fun to build a precise square wave generator also. But, I'd rather focus on one project at a time.

[dirkwright]

Clarostat 2k precision wire wound pot, linearity is 0.25%, near the bottom of the page:
Bourns Rotary Pots
$5 each. minimum order is $20.
Data sheet on this pot:
http://www.cbcdirect.com/direct/Imag...ts/PDF/773.pdf

[mickeymoose]

I am in the process of doing the board, along my lines (digital pot and seperate frequency display board). Obviosly that digipot can be replaced with an off-board dual-gang pot or switch. My concerns are the errors in the app note, not corrected by Linear in an addendum (fig 47/48 switched, pin-out of LT1010). A test circuit needs to be built to verify the design. Volunteers? E

[dirkwright]

Originally Posted by mickeymoose

I am in the process of doing the board, along my lines (digital pot and seperate frequency display board). Obviosly that digipot can be replaced with an off-board dual-gang pot or switch. My concerns are the errors in the app note, not corrected by Linear in an addendum (fig 47/48 switched, pin-out of LT1010). A test circuit needs to be built to verify the design. Volunteers? E

You can call up LT no problem:
Linear Technology - USA and Canada Sales Offices
I don't know which Provence you live, but here's the list of phone numbers.

If they can't help you, fill out the technical services form:
http://www.linear.com/designtools/tech_support.php

I don't have a means of verifying that the design works. I don't have anything that measures distortion that low. Obviously, we need to find someone who has the gear to measure this thing once it's built.

[PChi]

I'm sorry my meaning wasn't clear. My first sentence was a preamble to cover my criticisms, not other posts, of the LT circuit.

The rest is my suggestion of how to get variable frequency and a reasonably low distortion. I am in the middle of laying out a PCB and learning KiCad at the same time that uses LME49990s in a phase shift loop with a sample hold controlling an analog multiplier gain control.
I prototyped up part of the circuit using components that I had and it was stable at 10 Hz and the measured distortion and noise was around 0.004 % at 1 kHz using LM318 operational amplifiers on a scrap piece of strip board. I'm expecting better but it's going to take me some while to finish.

[dirkwright]

Originally Posted by PChi

I'm sorry my meaning wasn't clear. My first sentence was a preamble to cover my criticisms, not other posts, of the LT circuit.

The rest is my suggestion of how to get variable frequency and a reasonably low distortion. I am in the middle of laying out a PCB and learning KiCad at the same time that uses LME49990s in a phase shift loop with a sample hold controlling an analog multiplier gain control.
I prototyped up part of the circuit using components that I had and it was stable at 10 Hz and the measured distortion and noise was around 0.004 % at 1 kHz using LM318 operational amplifiers on a scrap piece of strip board. I'm expecting better but it's going to take me some while to finish.

Ah, ok, thanks for explaining that.

This may not be technically the best circuit, but it apparently does perform well enough, and I'm already committed to building it. I'm not expert enough to determine which circuit is better than another in oscillators. I've ordered three of the rare part already....

[davada]

Originally Posted by LAJ

try the XR2206 waveform generator chip. They will operate off one 9V battery but they usually like 12-18V range.
Also it gives triangle and square wave outputs.
You can usually find them on EBay.

I think I might have one of those XR2206 kicking around in the basement.
If anyone is interested I can have a look.

David.

[dirkwright]

I'll build anything mickeymoose comes up with, including the digital potentiometer. I have not heard back from potentiometers.com so who knows about them. Once all the boards and power supply are figured out I can start planning the box. I assume that this thing needs a really low noise power supply. I have the Borbley FET shunt regulator from Jim's Audio:
eBay - New & used electronics, cars, apparel, collectibles, sporting goods & more at low prices
I don't recommend it because many of the transistors on it are obsolete and therefore very hard to find. Four are impossible and have to be substituted.

[dennismiller55]

Well I guess my opening post here requesting information on a very low distortion signal generator has sparked an interest in many. I have read every post numerous times. Most of what has been discussed is over my head. However, if consensus is reached and a PCB and parts list become reality, I would like in please. I like diy projects in the form of kits. I just don't have circuit design knowledge.

Thanks everyone for the great discussion.

[dirkwright]

Considering that as far as I know you'd have to spend a huge amount of money to get a regular commercial sine wave generator with 0.0003% distortion, this is going to be a really good deal. Even the much loved HP209A has only 0.1% distortion, but considering how old it is, that was really good back then. I wonder if the amps in my scope are going to be this clean.... hmmm, maybe I should upgrade those amps also!??!