Low-distortion Audio-range Oscillator - Page 132 - 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 25th December 2012, 11:27 PM   #1311
davada is online now davada  Canada
diyAudio Member
 
Join Date: Jul 2004
Location: Fort St John, BC Canada
Quote:
Originally Posted by jackinnj View Post
Some folks who built Millet's RMS converter experienced noise issues with USB.

If I had to design a piece of audio instrumentation communicating via USB I'd isolate it (Analog Devices has a development kit for around $50, which you can do for a fraction if you SMT, same with other folks), and use a linear supply on the business end.
Hi jakinnj,

Dick Moore bought one of those AD demo boards but couldn't use it because it only does full speed. The EMU0204 and QA400 use high speed.

If we could interface to ethernet and back we have the galvanic isolation or something else
balanced.


Cheers,
__________________
David.
  Reply With Quote
Old 26th December 2012, 12:40 AM   #1312
1audio is offline 1audio  United States
diyAudio Member
 
Join Date: Mar 2004
Location: SF Bay Area
Blog Entries: 3
Default QA400 interface

Here is a sketch of what I'm working out. Its missing gain calibrations, common mode trims, output protection etc. It will give the general idea however. I'm using the THAT parts because they have a pretty high level of integration and take care of a bunch of precision resistors. I don't know yet if they will limit either the noise or distortion. Alternatives would be very interesting to look at. These are also available in DIP packages making them easier to deal with and, if socketed, easier to recover after mistakes.

Input protection uses small fuses (.062A fast blow). I may be able to simplify the output amp if the QA400 can drive a lower Z load without degradation. The opamp is a placeholder. Not figured out yet is power supply and possibly a local reference oscillator (1 KHz, light bulb stabilized, lockable from the QA400).
Attached Images
File Type: jpg QA400 interface sketch.JPG (72.8 KB, 214 views)
__________________
Demian Martin
Product Design Services
  Reply With Quote
Old 26th December 2012, 02:16 AM   #1313
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 1audio View Post
Here is a sketch of what I'm working out. I don't know yet if they will limit either the noise or distortion. Alternatives would be very interesting to look at.
That is a pretty important issue -- the front end interface adding to the measured DUT distortion. -RNM
  Reply With Quote
Old 26th December 2012, 02:20 AM   #1314
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 mkc View Post
Hi,


I'm sure you know Glen's project: An Audio T.H.D. Analyser


Best regards,
Mogens
His whole analyzer looks well thought out. I'm thinking of getting just the oscillator part to check out. -RNMarsh
  Reply With Quote
Old 26th December 2012, 03:24 AM   #1315
1audio is offline 1audio  United States
diyAudio Member
 
Join Date: Mar 2004
Location: SF Bay Area
Blog Entries: 3
The THAT parts were suggested to me by Bill Whitlock of Jensen transformers. He said they would be a good option for the AP input stage. Until I get my hands on something I can't tell on way or the other. I suspect that for most audio work they will be more than good enough. The cluster of parts to do it with other circuitry gets huge. The similar input board for the Boonton is 4" X 12" with not much vacant real estate. Paralleling opamps to lower noise can work. On a Yamaha tuner I have they paralleled (without balancing resistors!!) two opamps on the same die and got a real 3 dB improvement. I don't know if that would work here. The biggest constraint is the input attenuator impedance. 10K for 100V RMS is 1W, which has impact both in the source loading and in the size of the components. Going smaller would not be an option I think. The industry standard is actually 100K.
__________________
Demian Martin
Product Design Services
  Reply With Quote
Old 26th December 2012, 03:45 AM   #1316
davada is online now davada  Canada
diyAudio Member
 
Join Date: Jul 2004
Location: Fort St John, BC Canada
Quote:
Originally Posted by 1audio View Post
The THAT parts were suggested to me by Bill Whitlock of Jensen transformers. He said they would be a good option for the AP input stage. Until I get my hands on something I can't tell on way or the other. I suspect that for most audio work they will be more than good enough. The cluster of parts to do it with other circuitry gets huge. The similar input board for the Boonton is 4" X 12" with not much vacant real estate. Paralleling opamps to lower noise can work. On a Yamaha tuner I have they paralleled (without balancing resistors!!) two opamps on the same die and got a real 3 dB improvement. I don't know if that would work here. The biggest constraint is the input attenuator impedance. 10K for 100V RMS is 1W, which has impact both in the source loading and in the size of the components. Going smaller would not be an option I think. The industry standard is actually 100K.

Hi Demian,

All the resistor in the 339a input attenuator are 2 - 3 watts. The 339a input has to be able to blow the fuse with 300V without the resistor catching fire. I would think as an added bonus the resistors in that size would be more thermally stable because of the extra mass.
__________________
David.
  Reply With Quote
Old 26th December 2012, 09:19 AM   #1317
diyAudio Member
 
Join Date: Jun 2005
Location: Zürich
Quote:
The LT supper oscillator is a tuned composite amplifier to make it work at 10 kHz. Generally the compensation of two amplifiers in series raps the phase around through 180 deg and beyond before reaching unity. So they are unstable.
To satisfy stability requirements, we need >0° phase margin at the unity loop gain frequency only. At frequencies where loop gain is large, phase margin can be what it wants, even negative.

The LT "super oscillator" doesn't oscillate because the opamps are unstable. It oscillates for the same reason as every RC oscillator oscillates--for the tuned feedback network.

Note that for low distortion the open-loop gain of the opamp needs to be high at the frequencies of the harmonics, not at that of the fundamental.

Samuel
  Reply With Quote
Old 26th December 2012, 01:41 PM   #1318
KSTR is offline KSTR  Germany
diyAudio Member
 
KSTR's Avatar
 
Join Date: Jul 2007
Location: Central Berlin, Germany
Adding to Samuels post, have a look here : www.ti.com/lit/an/sboa015/sboa015.pdf , a very good app note by Burr Brown "mastermind" Jerald G. Graeme.

There comes a heading somewhere "Phase only matters at the intercept" which took me quite some time to understand and accept. Like many, I've always thought not dipping below 45deg phase margin at all frequences of the OLG is a criterion for stability but it's not, more like only when you need the opamp/composite be stable at any arbitrary gain setup which is flat at the intercept OLG point. For a fixed gain design, though, the phase can dip down to any value as long as it comes back and hits the OLG with its phase no more than ~120° behind the OLG phase (for a ~60deg PM)

I then actually built -- deadbug style -- a variant of his circuit of fig.9, low gain inverting with noise-gain comp (50dB), using two 5534's with 33pF comp each. OLG (EDIT: No, Loop Gain) Phase now certainly must have crossed though 180deg at two points somewhere below intercept, given the additional pole from the noise-gain comp. No sign of instability was seen and the distortion performance was really good. Overload recovery was nasty, though...

Last edited by KSTR; 26th December 2012 at 01:50 PM.
  Reply With Quote
Old 26th December 2012, 02:21 PM   #1319
diyAudio Member
 
jackinnj's Avatar
 
Join Date: Apr 2002
Location: Llanddewi Brefi, NJ
Quote:
Originally Posted by KSTR View Post
There comes a heading somewhere "Phase only matters at the intercept" which took me quite some time to understand and accept. Like many, I've always thought not dipping below 45deg phase margin at all frequences of the OLG is a criterion for stability
Christophe Basso explains how to think about loop stability: http://cbasso.pagesperso-orange.fr/D...y%20factor.pdf
  Reply With Quote
Old 26th December 2012, 05:36 PM   #1320
davada is online now davada  Canada
diyAudio Member
 
Join Date: Jul 2004
Location: Fort St John, BC Canada
Quote:
Originally Posted by Samuel Groner View Post
To satisfy stability requirements, we need >0° phase margin at the unity loop gain frequency only. At frequencies where loop gain is large, phase margin can be what it wants, even negative.

The LT "super oscillator" doesn't oscillate because the opamps are unstable. It oscillates for the same reason as every RC oscillator oscillates--for the tuned feedback network.

Note that for low distortion the open-loop gain of the opamp needs to be high at the frequencies of the harmonics, not at that of the fundamental.

Samuel
I wasn't suggesting that the op amp must be unstable to oscillate. I meant D.E. tuned the the sections to manipulate the GM and PM. He did illustrate this in his write up on the oscillator.

Well yes it makes sense that we only need high feedback gain at harmonics. If we consider a network like a bridged T, a passive notch network placed inside a negative feedback loop creates a band pass filter at a very high Q. There is little to no negative feedback at resonance since this is a high impedance state therefore we have open loop gain at Fr. What ever that is at a particular Fr.

Then it stands to reason if we use a second amplifier to provide some positive feedback to the tuned network to sharpen up the shoulders, we should see better 2nd and perhaps 3rd H results. We can do this with a Twin T notch network or a bridged T. This maybe an improvement to the single amplifier oscillator. If we pull up some 9dB at 2nd and 4db at 3rd or what ever it may be. This might be easier to do than tackling a composite arrangement. Is it more gain that we need for better distortion performance or just better use of it?

It also stands to reason then that 'high Q networks provide lower distortion' is a myth.
The lower Q networks provide more negative feedback in and around the fundamental. So there is no concern for the loss of Q by sharpening the shoulders of a network by whatever means. It just adds to the complexity.

I will read these articles mentioned. Composite amplifiers are one thing I've haven't had much success with stabilizing.

Cheers,
__________________
David.
  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 11:46 AM
ultra-low distortion audio oscillator geekysuavo Analog Line Level 16 26th March 2013 04:04 PM
Low distortion oscillator? rjm Equipment & Tools 30 4th May 2011 11:45 PM
Can we improve this low distortion sine oscillator ? gaetan8888 Solid State 22 29th March 2009 01:30 PM
Simple, low distortion 1kHz oscillator jackinnj Solid State 4 6th October 2003 04:58 AM


New To Site? Need Help?

All times are GMT. The time now is 11:29 PM.


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