Low-distortion Audio-range Oscillator - Page 40 - 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 15th September 2012, 03:40 PM   #391
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
If some one is looking for a pcb for the Jim Williams 10khz oscillator, i believe that Frex is still selling them. Just search for Username: Frex and Topic: oscillator .

The following link is mid point of the thread showing the performance. A few pages prior, I believe Frex posted his results.

DIY Analog-to-Digital Converter project.Audio measurements tool
  Reply With Quote
Old 15th September 2012, 03:45 PM   #392
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Here is the spectra from Frex

http://oneaudio.net/Pict/THDwithNotch.pdf

See post number 58 in previously posted thread.

Last edited by klewis; 15th September 2012 at 03:47 PM.
  Reply With Quote
Old 15th September 2012, 04:59 PM   #393
diyAudio Member
 
Join Date: Aug 2009
Location: Tallinn
Send a message via MSN to trioflex
Default Victors 1KHz measurements, no notch method

Hi

after some fine tuning we got some pictures that look already pretty nice:

Click the image to open in full size.

Click the image to open in full size.

Both screens are from same FFT run, Blackman window, 65K points, sample rate is 625KHz. I guess those are not so bad plots for DIRECT without notch filter measurements.

Can someone comment if this is all that is possible, or would it be really possible to get rid of the last harmonics spurs?

Setup: Victors oscillator (batterie powered) > DC coupled to AD4940 +input, AD4940 -input to GND via 620 Ohm, AD4940 OCM floating (Decoupled with 0.1uF to GND). ADC is AD7760. Victors oscillator is not shielded just PCB.

To have on the desk an ADC with 160dB noise floor is already nice, but we want more to get rid of the last few ppm of THD.

As of ALL measurements done with Victors generator, the spurs of 2nd and 3rd harmonic what we see at -138dB and -143dB are not in the generated signal?

Ok, we still have some fine tuning todo, the resistors in the differential path are maybe not all 0.1% ones and the power supply for AD4940 is a bit asymmetrical (+3V and -3V).

BTW, we will maybe be doing replica of the claimed PPB range oscillator, the funky one with those 180dB gain blocks.

I wonder if anyone has any experience with that design?
Its the Figure 67 in LT appnote collection AN67F
  Reply With Quote
Old 15th September 2012, 06:31 PM   #394
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
The oscillator I was refering to in prior posts was Linear Audio AN67.
  Reply With Quote
Old 15th September 2012, 07:02 PM   #395
davada is offline davada  Canada
diyAudio Member
 
Join Date: Jul 2004
Location: Fort St John, BC Canada
Quote:
Originally Posted by trioflex View Post
Hi

after some fine tuning we got some pictures that look already pretty nice:

Click the image to open in full size.

Click the image to open in full size.

Both screens are from same FFT run, Blackman window, 65K points, sample rate is 625KHz. I guess those are not so bad plots for DIRECT without notch filter measurements.

Can someone comment if this is all that is possible, or would it be really possible to get rid of the last harmonics spurs?

Setup: Victors oscillator (batterie powered) > DC coupled to AD4940 +input, AD4940 -input to GND via 620 Ohm, AD4940 OCM floating (Decoupled with 0.1uF to GND). ADC is AD7760. Victors oscillator is not shielded just PCB.

To have on the desk an ADC with 160dB noise floor is already nice, but we want more to get rid of the last few ppm of THD.

As of ALL measurements done with Victors generator, the spurs of 2nd and 3rd harmonic what we see at -138dB and -143dB are not in the generated signal?

Ok, we still have some fine tuning todo, the resistors in the differential path are maybe not all 0.1% ones and the power supply for AD4940 is a bit asymmetrical (+3V and -3V).

BTW, we will maybe be doing replica of the claimed PPB range oscillator, the funky one with those 180dB gain blocks.

I wonder if anyone has any experience with that design?
Its the Figure 67 in LT appnote collection AN67F
I believe Jackinnj has quite a bit of experience with the one with the 180dB blocks.
I don't remember the thread but he did build this and analyze it with comparison to other oscillators. If you do a search I think you can find the thread.

"the spurs of 2nd and 3rd harmonic what we see at -138dB and -143dB are not in the generated signal?"

If these are not in the generated signal then where are they coming from?
I wonder if notching the fundamental would reduce this.

David.
__________________
David.
  Reply With Quote
Old 15th September 2012, 07:11 PM   #396
davada is offline davada  Canada
diyAudio Member
 
Join Date: Jul 2004
Location: Fort St John, BC Canada
Quote:
Originally Posted by trioflex View Post
Hi

after some fine tuning we got some pictures that look already pretty nice:

Click the image to open in full size.

Click the image to open in full size.

Both screens are from same FFT run, Blackman window, 65K points, sample rate is 625KHz. I guess those are not so bad plots for DIRECT without notch filter measurements.

Can someone comment if this is all that is possible, or would it be really possible to get rid of the last harmonics spurs?

Setup: Victors oscillator (batterie powered) > DC coupled to AD4940 +input, AD4940 -input to GND via 620 Ohm, AD4940 OCM floating (Decoupled with 0.1uF to GND). ADC is AD7760. Victors oscillator is not shielded just PCB.

To have on the desk an ADC with 160dB noise floor is already nice, but we want more to get rid of the last few ppm of THD.

As of ALL measurements done with Victors generator, the spurs of 2nd and 3rd harmonic what we see at -138dB and -143dB are not in the generated signal?

Ok, we still have some fine tuning todo, the resistors in the differential path are maybe not all 0.1% ones and the power supply for AD4940 is a bit asymmetrical (+3V and -3V).

BTW, we will maybe be doing replica of the claimed PPB range oscillator, the funky one with those 180dB gain blocks.

I wonder if anyone has any experience with that design?
Its the Figure 67 in LT appnote collection AN67F
Hi trioflex,

I can't find a part for an AD4940. Do you mean ADA4940?
http://www.analog.com/static/importe...0-1_4940-2.pdf

David.
__________________
David.
  Reply With Quote
Old 16th September 2012, 08:19 AM   #397
diyAudio Member
 
Join Date: Aug 2009
Location: Tallinn
Send a message via MSN to trioflex
Quote:
Originally Posted by davada View Post
Hi trioflex,

I can't find a part for an AD4940. Do you mean ADA4940?
http://www.analog.com/static/importe...0-1_4940-2.pdf

David.
Hi

sorry yes ADA4940 this looks like the BEST available part to perform single ended to differential conversion, this is the only reason we use it, we do not need to change the common mode level, as AD7760 already is centered around 0V using the on-chip differential amplifier.

If there exist better IC to do low noise, low distortion we would sure consider using other part, but well we need bandwidth up to 1MHz, this is a bit above audio band..

Issues problems we have, maybe: AD7760 has 680 ohms all resistors on the differential ampliefier, so its input impedance bit above 680 ohm, now we accidentially have 549 ohm series with 51pF to GND after ADA4940, this not right of course the 549 ohm is too large there, but that only explains some extra attenuation we see on fundamental signal.

What is maybe bigger problem is that ADA4940 has about 260 ohms input impedance, and this is connected directly to Victors oscillator, well the oscillator has 600 ohm series, and it should I think work pretty well with low impedance loads too. But for sure oscillator output impedance is higher then ADA4940 input. ADA4940 has 1.2nF in parallel to the feedback resistor, so basically Victors oscillator is loaded with 200 ohm in series with 1.2nF.

Where do the harmonics come? This is to be found out.. When trying out Analog devices Differential Opamp calculator tool we see that ADA4940 with with the settings we are using should have SNR of around 105dB, well we are measuring SNR 106 for complete setup that includes ADA4940.

What is a bit mystery is that we have seen 3rd harmonics COMPLETLY disappearing in output spectrum! This was when we used 0.33uF large T/H capacitor as DC block in the oscillator output. 47uF MLCC 1206 added 5 db to the SNR, and 0.33uF DC bloc did hide the second harmonic into 150dB noise floor! If Victors oscillator would have 3rd harmonic present (at measurable with our system levels) would DC block capacitor type change be able to push 3rd harmonics (but only 3rd) down by 20+ dB?

Ah stupid, it is probably related to the loading of the oscillator!

Ok, the last 2 FFT plots are done with DC coupled measurement, without block capacitor, full path from oscillator to the ADC is DC coupled.
  Reply With Quote
Old 16th September 2012, 12:57 PM   #398
1audio is offline 1audio  United States
diyAudio Member
 
Join Date: Mar 2004
Location: SF Bay Area
Blog Entries: 3
Why don't you bypass the single ended to diff converter for the test? You can take the oscillator output through a series divider with the cenert tap tied to ground. With battery power it can work pretty well. put the oscillator in a metal box and tie the box to the center tap. Tie that point to the bias for the inputs of the ADC.

I published what I got for Victors oscillator earlier in the thread. All the harmonics were less than -110 as I remember.

THAT has some very good single ended to diff converters and vice versa. Certainly worth looking at.

At these levels the phase relationships of the harmonics generated can cause odd things like cancellation to happen. Also dropping the level into the ADC by 1-2 dB can reduce the harmonics a lot.
__________________
Demian Martin
Product Design Services
  Reply With Quote
Old 16th September 2012, 01:17 PM   #399
diyAudio Member
 
Join Date: Aug 2009
Location: Tallinn
Send a message via MSN to trioflex
Hi

well, initially we had the ADA4940 bypassed. We did see some very bad signal, well the best we did see was fundamental 20dB below noise! While troubleshooting this problem we installed the ADA4940 SMD module to the test base board we use. After 3 days of troubleshooting with Logic analyzer and VHDL simulations we found that the problem we did witness (signal below noise) was because in the GUI visualization software, well not even sure what was wrong, but I can tell what we changed to make it all work: We changed the size of the read buffer where we did take the samples for oscilloscope display. We did not change any functional code. The results we started to see then looked already very promising.

Taking the ADA4940 differential conversion out is possible but due to the small thermal resistance, we would have to PREHEAT the baseboard in order to suck off some solder from the pads that connect ADC module to the ADA4940 module.

AD7760 module uses 0201 components and well cost a bit, so we do not want to solder and re-solder too many times.

We can connect victors oscillator directly to ADC + input and tie -input to ground. This is acceptable (with 6dB performance loss). Of course when powered from battery, the trick with divider center point would work as well.

Harmonics are down almost 110 dB in our measurements already, I am hunting for 20dB more... as per measurements Victors oscillator it should be possible? The best THD reading we got was 5.3 ppm, Victors oscillator has been measured to be below 0.3 ppm. So we would love to see our ADC to actually measure below 1 ppm
  Reply With Quote
Old 16th September 2012, 03:33 PM   #400
diyAudio Member
 
dirkwright's Avatar
 
Join Date: May 2011
Location: Virginia
My results with Victor's oscillator. It's running on 4x9V lithium-ion cells, which measure about 32VDC. I don't see any 60Hz noise, and no harmonics. Of course, this is just with a cheap Behringer ADC.
Attached Images
File Type: jpg Victors oscillator.JPG (155.0 KB, 114 views)
  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 02:30 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