Build -- Active Twin-T notch filter for distortion analysis - Page 8 - 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 17th January 2013, 06:04 PM   #71
richiem is offline richiem  United States
diyAudio Member
 
richiem's Avatar
 
Join Date: Sep 2010
Location: Grapeview, WA
@JPV -- I have done exactly that and found that using the Active Twin-T made a significant difference. While the 132k buffer definitely lowers the noise, it does not significantly diminish the distortion in the EMU caused by the high level fundamental -- in my case, 1VRMS. I suggest you build a simple passive Twin-T filter for 1kHz that attenuates the fundamental at least 30dB, then check your distortion products again -- remember to add 9.5dB to the 2nd H level and 5dB to the third.
__________________
...................
Dick Moore
  Reply With Quote
Old 18th January 2013, 09:14 AM   #72
JPV is offline JPV  Belgium
diyAudio Member
 
Join Date: Dec 2006
Quote:
Originally Posted by richiem View Post
@JPV -- I have done exactly that and found that using the Active Twin-T made a significant difference. While the 132k buffer definitely lowers the noise, it does not significantly diminish the distortion in the EMU caused by the high level fundamental -- in my case, 1VRMS. I suggest you build a simple passive Twin-T filter for 1kHz that attenuates the fundamental at least 30dB, then check your distortion products again -- remember to add 9.5dB to the 2nd H level and 5dB to the third.
I fully agree with you that decreasing the level of the fundamental will decrease the distortion in the EMU. This is the basic definition of non linear distortion: level dependent. My remark is that for normal measurements, the level of distortion is way above the combined distortion of the emu card + very low distortion oscillator at 24 bits but can be buried in noise. Using ARTA with a low distortion oscillator and a deep FFT allows you to measure amazingly low harmonics without having to tune an add on device for each measurement.


There is an interesting test to do IMHO.
At low levels of distortion, it can be shown that :

Let H2 be the ratio of the second harmonic to the fundamental ( fundamental as seen in the spectrum of the output of the DUT and not in the spectrum of the input to the DUT) , H3 the same for the third harmonic. It is well known that H2 is proportional to the square of the test sinusoid and H3 is proportional to the cube.

Let IM2 be the ratio of the second intermodulation product of two sinusoids of same amplitude at frequencies f1 and f2, f2>f1 to the fundamental f1. The second harmonic due to intermodulation will be at f2-f1 and f2+f1, the third IM3 will be at 2f1-f2 and 2f1+f2.

It can be shown that at low frequencies and low distortion levels, the following holds: IM2 = 2 times H2 and IM3=3 H3.

So, by injecting two sinusoids at same level and close frequencies we can look at the intermodulation harmonics that are generated by the EMU. these harmonics are only related to the distortion mechanism in the emu.
But also, the second IM2 will be 6db higher than the second harmonics H2 generated only by the EMU The IM3 will be 9.54 db higher than H3.

From the reading of H2 in the spectrum ( wich contains the harmonic of the oscillator and the harmonic of the EMU) we substract (IM2- 6db) and we have the pristine second harmonic of the oscillator !!!
The same for third harmonic but with 9.54db.
Of course this implies that the harmonics are amplitude adding and there is a phase problem but I think the have the same phase
I would like to do it but I have only one oscillator.

JPV
  Reply With Quote
Old 18th January 2013, 10:08 AM   #73
JPV is offline JPV  Belgium
diyAudio Member
 
Join Date: Dec 2006
Of course you will need to design an opamp based adder with virtually no distortion. I beleive that the National opamps can come close to it

JPV
  Reply With Quote
Old 18th January 2013, 12:14 PM   #74
diyAudio Member
 
Join Date: Jun 2005
Location: Zürich
Quote:
It is well known that H2 is proportional to the square of the test sinusoid and H3 is proportional to the cube.
This applies only if anything in the signal chain is "weakly nonlinear only", i.e. we're dealing with smooth nonlinearities. This does not hold at all for an ADC or DAC--there are switching devices and other nonlinearities involved which are very difficult to model (in fact, nearly impossible at the current state of the art) and which don't follow such simple rules.

In fact this simple model doesn't even hold for a standard opamp--the class B output stage is not weakly nonlinear anymore, as the transistors switch fully off at some point.

Samuel
  Reply With Quote
Old 26th January 2013, 09:36 PM   #75
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 richiem View Post
Update on the Active Twin-T and input levels -- four 9V batteries for power keep things clean with inputs up to 8VRMS or so, and with the nominal 1VRMS I usually use, the Active Twin-T appears not to add any measurable distortion. With my build of Bob Cordell's state-variable oscillator IG-18 #2, the BIG-18, configured for moderate output buffer gain, I'm resolving THDs of under 0.00004% at 1kHz, and can't prove that any of that comes from the Twin-T -- but can't disprove it either....
Hi, Dick M.,

Q: is there a pcb coming for the active portion of your twin-T? By/from you or other?

Thx - Dick M.
  Reply With Quote
Old 27th January 2013, 02:53 AM   #76
richiem is offline richiem  United States
diyAudio Member
 
richiem's Avatar
 
Join Date: Sep 2010
Location: Grapeview, WA
No board -- it's a simple circuit with three opamps, one of which could be a dual; or four opamps if you want to use an input buffer -- I don't. Easy to build on perfboard (yes, a nuisance) or small enough to get ExpressPCB's three-board $51 deal.
__________________
...................
Dick Moore
  Reply With Quote
Old 8th February 2013, 12:51 AM   #77
RNMarsh is offline RNMarsh  United States
diyAudio Member
 
RNMarsh's Avatar
 
Join Date: May 2012
Location: 2457 Cascade Trail; Cool, CA. 95614
Thanks to a contributor here, Larry Burk, for being so helpful in making and supplying a pcb to speed things up for me. I am waiting now for some 10-turn ww pots and some specific R values. Thx-RNMarsh
  Reply With Quote
Old 8th February 2013, 02:00 AM   #78
davada is offline davada  Canada
diyAudio Member
 
Join Date: Jul 2004
Location: Fort St John, BC Canada
I was going to finish up my TT this weekend but the color printer my bother in law gave me
is completely void of toner. I need this to print the water slide decals for the front panel.
$550.00 comes to mind. Sigh.
__________________
David.
  Reply With Quote
Old 9th February 2013, 05:09 PM   #79
RNMarsh is offline RNMarsh  United States
diyAudio Member
 
RNMarsh's Avatar
 
Join Date: May 2012
Location: 2457 Cascade Trail; Cool, CA. 95614
News Flash -- How about using the notch filter in an oscillator with its regenerative feedback disconnected? Built-in multi-freq notch. just add an input port and disconnect switch for the regen circuit.... any oscillator lying around could be a candidate. Wha chu tink?

Thx-RNMarsh

Last edited by RNMarsh; 9th February 2013 at 05:10 PM. Reason: Built-in notch filter
  Reply With Quote
Old 9th February 2013, 09:16 PM   #80
richiem is offline richiem  United States
diyAudio Member
 
richiem's Avatar
 
Join Date: Sep 2010
Location: Grapeview, WA
To what end, Richard? Notching out an individual harmonic? Or?
__________________
...................
Dick Moore
  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
need help wana change 60-Hz twin-T notch filter to 279 Hz filter prorms Solid State 0 9th February 2008 11:38 PM
Active Notch Filter design needed Vikash Multi-Way 63 21st May 2007 05:14 PM
Twin-T Lowpass Filter Analysis gmikol Parts 4 17th May 2007 08:16 PM
Noise out of active notch filter reins Multi-Way 0 6th July 2006 07:34 PM
Active notch filter 5th element Solid State 7 6th February 2004 10:06 PM


New To Site? Need Help?

All times are GMT. The time now is 10:05 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