Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

D-Noizator: a magic active noise canceller to retrofit & upgrade any 317-based V.Reg.
D-Noizator: a magic active noise canceller to retrofit & upgrade any 317-based V.Reg.
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 August 2019, 09:13 PM   #151
jls001 is offline jls001  India
diyAudio Member
 
Join Date: Jan 2012
Location: Mumbai
Quote:
Originally Posted by Elvee View Post
The principle should also work for the 338; the details might be different, in particular the compensation components.
If you make a PCB, I recommend you include a provision for a resistor in series with the compensation capacitor: you might end up strapping it, but if you play with the capacitor value and the resistor value (probably between 0 and 100 ohm), you will almost certainly find a correct stability range.
Almost is no certitude of course, and only an actual test will tell the definitive truth.
Thanks for the response, Elvee.

I already have two LM338-based power supplies powering an Allo Sparky and an USBridge. Someone built it for for me. I believe both are straight ahead datasheet implementations and they already sound better than the supplied wallwart.

I'm keen to rig up the D-Noizator on a general purpose board (like you did).
  Reply With Quote
Old 18th August 2019, 10:40 AM   #152
dotneck335 is offline dotneck335  United States
diyAudio Member
 
Join Date: Dec 2012
Quote:
Originally Posted by Elvee View Post
A simple circuit that transforms a good, clumsy workhorse like the LM317 into a superreg so easily that you do not even need to cut a single track.
Let us start with the basic regulator (including the additional bypass cap, since we strive for the lowest noise).
Now, let us add the D-Noizator piggy-back:
The floor drops by more than 30dB.
The circuit works by changing the output node into an AC virtual ground.
Therefore, benefits do not stop at internal noise reduction: all the performances benefit: PSRR, output impedance. Ultimately, the performances are limited by the discrete transistor.
What about the reality? I made a test with ST and ONsemi samples. They behave as perfect clones, both with and without the D-Noizator.
With just the cap, the noise is 16μV, as measured by a Levell TM3B. With the D-Noizator, it falls to 2.5μV~3μV
I am curious as to how you got such a low noise figure to begin with. The 317 data sheet lists noise as 0.001% of the output voltage. Unless I am somehow figuring this incorrectly, for a 15-volt supply, this would be 150 µVolts----WAY higher than the 16µVolts you measured.
  Reply With Quote
Old 18th August 2019, 04:58 PM   #153
Elvee is offline Elvee  Belgium
diyAudio Member
 
Elvee's Avatar
 
Join Date: Sep 2006
An important thing to take into account (I already mentionned it somewhere in the thread) is that I compared the denoiser with the best possible standard 317 implementation, ie. with the capacitors.

The capacitor reduces the effective noise gain to unity, meaning the value measured will always be for a 1.25V regulator, independent of the actual DC output voltage. This explains the ~1:10 discrepancy.

Something else to take into account is that when I mention the total rms noise, it is always for 10Hz to 10kHz bandwidth.

And finally, the 2.5µ to 3µV figure has been updated later, when my LNA became operational.
I don't remember the updated value, but it is better, and it is in the thread

Edit:

Here it is: around 0.66µV
Quote:
Originally Posted by Elvee View Post
I have been able to refine the measurements: I remembered that one channel of one of my lab amplifiers is equipped with a OPA27.

This brought the noise floor of my test set-up to 0.75µV.
When measuring the D-noizator, I got 1.0µV.
This means that the actual noise is in the vicinity of 0.66µV.
Since the noise with the cap connected normally is still 16µV, the improvement ratio is 27.7dB, much closer to what the sim predicted.

Of course, the test is made on a breadboard, and the transistor is just an ordinary BC337-40 from Philips.

With a good PCB and a super low noise transistor, it would be possible to scrape a few tens of nV
Re-edit:

This was not the final measurement, it was made with a temporary arrangement. The final value is even better (and it is also somewhere in the thread)
__________________
. .Circlophone your life !!!! . . ♫♪ My little cheap Circlophoneİ ♫♪
Electrify your daily life!!! ....⚡⚡ ELEKTRIA ⚡⚡

Last edited by Elvee; 18th August 2019 at 05:08 PM.
  Reply With Quote
Old 18th August 2019, 06:25 PM   #154
craigtone is offline craigtone  United States
diyAudio Member
 
craigtone's Avatar
 
Join Date: Apr 2006
Location: Holland, MI
Quote:
Originally Posted by jackinnj View Post
The cost per board, the one pictured below poz and neg -- about $6.60 for 100 produced in the US including shipping and sales tax. I am awaiting a quote from China.
I think C6 needs rotated 180 in the silk on the LM337 side.
__________________
Craigtone Audio
Soundsmith Boheme, WT Classic, Pearl 2, Apollo-R, Burson Swing, Hafler 915, TSE 45, ALK Klipschorns
  Reply With Quote
Old 18th August 2019, 07:22 PM   #155
Elvee is offline Elvee  Belgium
diyAudio Member
 
Elvee's Avatar
 
Join Date: Sep 2006
Quote:
Originally Posted by craigtone View Post
I think C6 needs rotated 180 in the silk on the LM337 side.
Yes indeed
Quote:
Originally Posted by Elvee View Post
I have now remeasured the output noise in good conditions: it is <0.3µV in a 10Hz to 10kHz bandwidth;
nothing particularly exceptional, but it all depends on the auxiliary transistor, an ordinary BC337 in this case.
The figure could be improved by using a really high performance, low noise transistor
Thus, the latest and final(?) noise level in a 10Hz to 10kHz range is 0.3µV.
It could be very marginally improved using a better transistor, but even with a perfect transistor, the improvement will remain marginal: the denoiser reaches its limit, dictated by the 317's noise and the gain.

To do better, you would need to switch to the nonoiser, but implementing it on existing designs is going to be much more complicated (and probably not worth the trouble)
__________________
. .Circlophone your life !!!! . . ♫♪ My little cheap Circlophoneİ ♫♪
Electrify your daily life!!! ....⚡⚡ ELEKTRIA ⚡⚡
  Reply With Quote
Old 18th August 2019, 07:31 PM   #156
jackinnj is offline jackinnj  United States
diyAudio Member
 
jackinnj's Avatar
 
Join Date: Apr 2002
Location: Llanddewi Brefi, NJ
D-Noizator: a magic active noise canceller to retrofit &amp; upgrade any 317-based V.Reg.
Quote:
Originally Posted by craigtone View Post
I think C6 needs rotated 180 in the silk on the LM337 side.
Thanks! Fixed that.

I will email the gerbers to anyone who PM's -
  Reply With Quote
Old 20th August 2019, 02:41 AM   #157
RickTH is offline RickTH
diyAudio Member
 
Join Date: May 2019
Hey LV , another question.

Often I see a diode to protect the LM317 when a big noise reduction cap is used.
The cap is pretty big here with the DeNoiser : 220 uF. Shouldn't there be diodes to not only protect the LM but the transistor too at switch off ?
Attached Images
File Type: jpg DN.JPG (15.1 KB, 105 views)
  Reply With Quote
Old 20th August 2019, 06:02 AM   #158
dotneck335 is offline dotneck335  United States
diyAudio Member
 
Join Date: Dec 2012
And, LV, one more:
Any caveats to using the DeNoiser with a slightly higher voltage (37 volts) and much higher current (7 amps) regulator (uses external pass transistors)? Should I scale up the resistor values?
  Reply With Quote
Old 20th August 2019, 06:50 AM   #159
Khadgar2007 is offline Khadgar2007  Croatia
diyAudio Member
 
Join Date: Feb 2008
Send a message via ICQ to Khadgar2007
On our local DIY forum some guy did designed workbench power supply with lm317,external pass BJT,current limit that goes up to 50V, and added denoiser to it, simulated in TinaTI and it looks really good. Other member made it, everything works just like in simulation but he don't have equipment to measure noise.
  Reply With Quote
Old 20th August 2019, 09:01 AM   #160
Elvee is offline Elvee  Belgium
diyAudio Member
 
Elvee's Avatar
 
Join Date: Sep 2006
Quote:
Originally Posted by RickTH View Post
Hey LV , another question.

Often I see a diode to protect the LM317 when a big noise reduction cap is used.
The cap is pretty big here with the DeNoiser : 220 uF. Shouldn't there be diodes to not only protect the LM but the transistor too at switch off ?
Yes, for the protection against input and output shorts, the two regular diodes should be used.
If the supply to upgrade already has the Cadj cap, it will probably have the diodes too.
If the designer deemed them unnecessary, they are not necessary either with the denoiser.

If they are included, it is preferable to add the transistor protection diode: it will protect it against a dead short at the output.
Note that D1 and Q1 are much more robust than the internals of a 317: the Ifsm for the 1N4148 is 2A, and although the peak collector current of the BC337 is only 1A, it can probably also withstand 2A under majority carriers operation (but a short will kill both of them)

Quote:
Originally Posted by dotneck335 View Post
And, LV, one more:
Any caveats to using the DeNoiser with a slightly higher voltage (37 volts) and much higher current (7 amps) regulator (uses external pass transistors)? Should I scale up the resistor values?
You mean something like this?

D-Noizator: a magic active noise canceller to retrofit &amp; upgrade any 317-based V.Reg.-denoiserboost-png

It seems to work in simulation, but since the transistor modifies the loop characteristics of the 317, the reality might be different.
In particular, it might be necessary to modify C4, and to add a small series resistor (0 to 100 ohm) with it.

BTW, I now recommend a larger value of C4 for any version: 10 to 22nF.
The HF noise reduction is slightly affected, but the stability is improved.
I do not recommend scaling up the resistors values: it will reduce the performances.
The only adaptation necessary is to use a powerful enough resistor type for R4: for 37V, 0.5W is sufficient (you can use 3/4W or 1W to have some margin)
Attached Images
File Type: png DenoiserBoost.png (67.0 KB, 92 views)
__________________
. .Circlophone your life !!!! . . ♫♪ My little cheap Circlophoneİ ♫♪
Electrify your daily life!!! ....⚡⚡ ELEKTRIA ⚡⚡
  Reply With Quote

Reply


D-Noizator: a magic active noise canceller to retrofit &amp; upgrade any 317-based V.Reg.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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
PSU upgrade for DAC Magic Plus E30NA Digital Line Level 8 1st October 2014 12:00 PM
Higher-harmonic distortion canceller jkeny Solid State 0 29th January 2009 11:39 PM
LM3886 based Amp -Upgrade advice please. mikesnowdon Chip Amps 3 26th October 2008 12:53 PM
Active crossover - Voodoo magic? ShinOBIWAN Multi-Way 5 24th February 2005 02:34 AM


New To Site? Need Help?

All times are GMT. The time now is 10:20 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki