Low-distortion Audio-range Oscillator - Page 199 - diyAudio
Go Back   Home > Forums > Design & Build > Equipment & Tools

Equipment & Tools From test equipment to hand tools

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 29th March 2013, 02:11 AM   #1981
RNMarsh is offline RNMarsh  United States
diyAudio Member
 
RNMarsh's Avatar
 
Join Date: May 2012
Location: 2457 Cascade Trail; Cool, CA. 95614
Hmmmm. The AD633 has a unused and grounded summing input to the buffered/multiplier output. :-) Hmmmmm.

Thx-RNMarsh
  Reply With Quote
Old 29th March 2013, 02:55 AM   #1982
diyAudio Member
 
Join Date: Nov 2008
Location: Oakmont PA
A couple of issues, if you average the FFT analysis of an oscillators output you might miss amplitude modulation of the harmonics that would indicate gain hunting. A cepstrum would be better for that.

The wider the sweep range the more the gain shifts and the more range the gain control element requires. The more range of the gain control element the more its distortion contributes.

All of the component parts have improved over the years. It was very interesting to see in Jim Williams last design the use of "tweak" parts from Oscon capacitors to really low distortion resistors in locations he thought were critical.

I currently use a transversal filter for my basic oscillator design. The longer it is the lower the distortion to the component and switch accuracy. In the basic form it is nowhere near any of the units considered here. It's strength is accuracy of frequency control and consistent output level. Issues that don't seem as big here, but are critical for some of my tests.
  Reply With Quote
Old 29th March 2013, 06:57 AM   #1983
RNMarsh is offline RNMarsh  United States
diyAudio Member
 
RNMarsh's Avatar
 
Join Date: May 2012
Location: 2457 Cascade Trail; Cool, CA. 95614
It isnt that it is of no interest. rather the amplitude regulation has been so good as not to worry. But, what do you use for a Cepstrum?

I have replaced quit a few passive parts to lower distortion. Resistors, capacitors and trim pots/controls... switch contacts, etc.

What do you test that needs a super accurate freq control and output level?

Thx-RNMarsh
  Reply With Quote
Old 29th March 2013, 07:04 AM   #1984
RNMarsh is offline RNMarsh  United States
diyAudio Member
 
RNMarsh's Avatar
 
Join Date: May 2012
Location: 2457 Cascade Trail; Cool, CA. 95614
Quote:
Originally Posted by RNMarsh View Post
Hmmmm. The AD633 has a unused and grounded summing input to the buffered/multiplier output. :-) Hmmmmm.

Thx-RNMarsh
And, another AD633 can be made as a freq doubler off the osc output...... and a sample of it put into the first 633's summing input. Got some AD633 on order. But will try to do a test of concept first.

-RNM
  Reply With Quote
Old 29th March 2013, 08:17 AM   #1985
diyAudio Member
 
Join Date: Jun 2005
Location: Zürich
Quote:
I just wonder exactly how important the rectification is given some adequate method of filtering ripple -- and when in the Cordell SVO the ripple frequency was cut in half and the ripple level increased by 6dB without any consequent penalty, well, I wonder.
Very much a function of how much the multiplier is decoupled. In the design I'm working on the multiplier is decoupled about 4x more than it is in the AP System One. That's about the limit I've found to still get a dependable 10 s settling time to 0.1% for a 100 Hz to 10 Hz range change. With such a configuration, the level detector ripple at the lower harmonics (fundamental to 3rd or 4th) needs to be in the "few dozen uVrms" range to keep its distortion contribution below the -150 dB level. This is feasible with a very good dual sample-and-hold level detector, without additional low-pass filtering. With less multiplier decoupling, the requirements for level detector ripple quickly become hopeless.

Quote:
So, where do You think the weak link is in these existing products/designs?
No idea, I haven't analyzed them in such detail. In a well-designed product such as the System One the various distortion sources are likely kept at similar level, such that there is an overall well-balanced cost-performance ratio.

Quote:
My goal has been for multiple freqs output at below -140 THD+N or better.
As all I'm writing is old news for you I'm confident you've already measured the noise floor of a typical oscillator topology--just for the others which may have not: this can be done by snipping the multiplier control voltage port off and feeding it with an external low-noise DC voltage which is just a weeny bit lower as the one which makes the oscillator oscillate. Then feed the oscillator output, which is now a funny narrow-band noise signal, to a THD+N analyzer manually set to the same frequency as the oscillator (the notch is required as near the oscillation frequency the voltage noise density becomes very large, and this frequency band is removed in a THD+N measurement too). This measurement procedure ignores amplitude modulation noise from the leveling loop, but is still a pretty good way for a first estimate.

With reasonable impedance levels and available IC opamps, it's difficult to push the noise floor much below -110 dBu (22 kHz BW) up to 20 kHz. So for your -140 dB THD+N the oscillator operating level would need to be in the order of +30 dBu!

Quote:
At -140 dB and below passives may become a roadblock.
Definitely, particularly as the operating level of the oscillator needs to be relatively high for good THD+N, as the above discussion showed. I'm currently working out the details of the output attenuator, and it looks like I'll need several dozen resistors as series/parallel configuration!

Samuel
  Reply With Quote
Old 29th March 2013, 09:28 AM   #1986
HpW is offline HpW  Switzerland
diyAudio Member
 
Join Date: Mar 2002
Location: Switzerland (Bern)
Well,

Quote:
Originally Posted by RNMarsh View Post
We are already at the -140dB/ <.00005% THD+N level using this fast-track approach. Can we go lower - into the noise?
Thx-RNMarsh
You are already there! You are already playing in a high league.

-140dB is a 100.0E-9 ratio and for a 5V RMS it's 0.50 uV = 500nV

Cheers

Hp
__________________
www.hpw-works.com
  Reply With Quote
Old 29th March 2013, 12:37 PM   #1987
diyAudio Member
 
Join Date: Nov 2008
Location: Oakmont PA
Quote:
Originally Posted by RNMarsh View Post
It isnt that it is of no interest. rather the amplitude regulation has been so good as not to worry. But, what do you use for a Cepstrum?

I have replaced quit a few passive parts to lower distortion. Resistors, capacitors and trim pots/controls... switch contacts, etc.

What do you test that needs a super accurate freq control and output level?

Thx-RNMarsh
For the cepstrum I am still using a DOS program as it seems I am one of the very few who look at it.

I am only 1/3 way through reading this thread so I apologize for repeat some of the earlier issues.

There are two reasons for looking at amplitude variations. The dynamic changes really are amplitude modulation. A .1 db variation is about 1% modulation which then raises the issues of at what frequency is that dynamic variance. Of course these appear as sidebands. Such sidebands can mask "base spreading" which is an indicator of voltage coefficient noise.

Now overall level change with frequency is a sign that the gain control element has to have greater range and thus more contribution to distortion.

Most of the tests that I am getting to require the oscillator is locked to the analyzer. Then you can use a very fine time delay to look at the phase artifacts.

Although doing fine and finer THD and noise is quite challenging, looking else where seems to be more productive for measuring artifacts that affect perceived quality.

ES
  Reply With Quote
Old 29th March 2013, 01:07 PM   #1988
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by Samuel Groner View Post
I see. Well, I'm pretty confident that it is possible to design oscillators with better distortion performance than presented in this thread, without resorting to fundamentally new approaches. The first step to improve a given design is to understand the dominant distortion mechanism. We have four known fundamental distortion sources--passives, opamps, multiplier and level detector ripple. To find out which one's the culprit, it is best to measure each one in isolation. Just looking at the oscillator output (as you've done it so far) is not very helpful. This applies particularly because various distortion sources may cancel. If you swap in another opamp type and observe that the oscillator output has lower distortion, this may have two reasons. First, it could be that indeed the old opamp was the dominant distortion mechanism, and the new one is better. Or it could be that the new one actually has higher distortion, and that happend to cancel better with another distortion source. There's no way to decide which one's the case if you just look at the oscillator output. But if you hit the second one and believe it's the first, you'll get tremendously fooled and run in circles forever. So looking at each of the four known fundamental distortion sources in isolation is absolutely inevitable.

So we need to know how this can be done. The passives may be checked on a suitable bridge setup, or you can replace them with series-parallel combination which lowers their distortion contribution. The opamps can be measured at high noise gain (luckily a Swiss chap has already done much of this hard work for you). The multiplier can be isolated in the oscillator and tested by feeding it with a suitable sine input and DC control voltage. Level detector ripple distortion contribution can be estimated by operating the oscillator and taking an FFT of the voltage at the multiplier control voltage port. The various frequencies can be translated into resulting harmonic distortion levels by the simple amplitude modulation equation.

That's a lot of work, but before thinking how to make it better we need to know what's the issue...

Samuel
Samuel, this is a great observation. I've tried to do a bit of this when I was upgrading the oscillator in my THD analyzer, and it is indeed a lot of work. I had focused on the agc detector and the JFET control element, but not in as rigorous a way as you suggest. I piddled around with changes in each of those circuits that should have made a difference, and was able to infer what was dominating. For example, I might double the filtering on the agc control signal, temporarily sacrificing settling time, to see if the oscillator distortion would go down. For the JFET, I altered the relationship between JFET operating signal level and the injection factor of its correction signal into the oscillator loop.

But I like your more thorough approach more.

Cheers,
Bob
  Reply With Quote
Old 29th March 2013, 02:22 PM   #1989
diyAudio Member
 
Join Date: Sep 2006
Quote:
Originally Posted by RNMarsh View Post
yes. I include noise and distortion reduction in the process. I believe progress is made by getting away from the jFET to reg level. It improved THD byat least 6db across the board with same oscillator IC.

Thx-RNMarsh
Hi Richard,

Sorry if the answer came up before. This thread is great, but fast-moving and I sometimes can't keep up.

Can you describe specifically how you measure the oscillator noise, and what numbers you got for one of the better oscillators you measured?

I'd like to measure my oscillator for noise in the same way to see how it compares.

Cheers,
Bob
  Reply With Quote
Old 29th March 2013, 04:41 PM   #1990
RNMarsh is offline RNMarsh  United States
diyAudio Member
 
RNMarsh's Avatar
 
Join Date: May 2012
Location: 2457 Cascade Trail; Cool, CA. 95614
Quote:
Originally Posted by Bob Cordell View Post
Hi Richard,

Can you describe specifically how you measure the oscillator noise, and what numbers you got for one of the better oscillators you measured?

I'd like to measure my oscillator for noise in the same way to see how it compares.

Cheers,
Bob
I am looking at the notched output of residual harmonics from the analyzer monitor output. I am using an analog swept wave filter type analyzer - HP-3580A.

The monitor output is scaled to the analyzer range selected with a normalized output level of 1v rms = full scale (fs) of the range selected. The range is on its lowest from a ShibaSoku AD725D analyzer which is -110dB (.0003% fs). The noise floor from this monitor port is about -160dB re fs.

The dynamic range of the HP is >85dB on its 0db/1v range.

The signal level from the generator being measured is 7v rms.


Thx-RNMarsh

Last edited by RNMarsh; 29th March 2013 at 05:07 PM. Reason: osc. noise -
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Radford Low Distortion Oscillator Series 2 audiomik Equipment & Tools 21 19th February 2014 10:46 AM
ultra-low distortion audio oscillator geekysuavo Analog Line Level 16 26th March 2013 03:04 PM
Low distortion oscillator? rjm Equipment & Tools 30 4th May 2011 10:45 PM
Can we improve this low distortion sine oscillator ? gaetan8888 Solid State 22 29th March 2009 12:30 PM
Simple, low distortion 1kHz oscillator jackinnj Solid State 4 6th October 2003 03:58 AM


New To Site? Need Help?

All times are GMT. The time now is 05:21 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2