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

Simple Low Noise Voltage Regulator?
Simple Low Noise Voltage Regulator?
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
Thread Tools Search this Thread
Old 9th November 2011, 07:19 AM   #11
qusp is offline qusp  Australia
diyAudio Member
qusp's Avatar
Join Date: Oct 2009
Location: Brisbane, Australia
Originally Posted by QSerraTico_Tico View Post
Noise or output impedance do not control sound quality...
sorry but that is just to much of a generalisation to be left alone. it depends very much on what you are driving with said circuit. for headphones of low impedance for example the output impedance has a rather audible effect on sound quality
  Reply With Quote
Old 9th November 2011, 01:23 PM   #12
jackinnj is offline jackinnj  United States
diyAudio Member
jackinnj's Avatar
Join Date: Apr 2002
Location: Llanddewi Brefi, NJ
Simple Low Noise Voltage Regulator?
Originally Posted by QSerraTico_Tico View Post
Noise or output impedance do not control sound quality...
Then you may have never heard a preamp or amp whose output is modulated by power supply problems.

You can test for this pretty easily. Build a simple JFET preamplifier with a gain of 10X. Inject A-weighted noise into Vcc or Vee and measure the point at which it becomes "hearable" on a quiet recording. For a moving coil phono preamp this is going to be a pretty low number!

Output impedance is tricky to measure below 10 milliOhms or so. Walt Jung, in a lab setup was able to measure it below 10 microOhms. If you want to demonstrate for yourself the impact of supply output impedance, take a look at the power supply Vcc or Vee when a highly percussive instrument is played. Or, one could look at the effect (IM) when a clock signal is coupled to a preamplifier or buffer supply line.

It's all in the listening. I use SY's "His Master's Noise" MC phono pre which is not in the same noise category as some of the JFET designs, but it sounds sweet nonetheless. (Should mention that the current source for his design is very low impedance depletion MOSFET.)
  Reply With Quote
Old 22nd November 2011, 12:38 AM   #13
twest820 is offline twest820
diyAudio Member
Join Date: Jun 2009
Yes, in the end it's the listening which matters. But it's not particularly difficult to design for supply impedance in conjunction with control loop PSRR. The attached figure's a fairly typical example but there's a lot of information packed into it, so let me explain what it's showing. The four sets of curves indicate the magnitude of a linear supply's ripple voltage relative to an amp's output power as a function of available bypass capacitance. For example, if you have a fairly decent 10,000uF reservoir cap arrangement with power planes and your amp's playing at a total output power of 10mW RMS then the design will land right around the bottom pink curve---the supply impedance will be roughly 25mOhm + 5nH and the ripple voltage will be 10dB larger than the signal that's going out to the driver. So if you want to minimize supply artifiacts on the output---say 80dB below the signal---then the amp's control loop needs to have 90dB PSRR.

There are many different cases here as every supply's impedance varies with layout and capacitor selection, different amplifiers have different quiescent current (Iq) draws, and different drivers present different impedances and efficiencies. The basic analysis doesn't change much from design to design, though; most drivers are about 90dB efficient and most listening happens at conversational levels around 55dB. So most home audio power amplifiers spend most of their time operating in the range of 100uW to 10mW. This is the range in which the bias current in the output devices dominates the amplifier's power consumption and hence the amount of ripple in the supply---I've chosen Iq = 200mA and 3 ohm load here as that's representative of a class AB stereo amplifier driving the lower impedance range of most speakers. 100uW to 10mW is also unfortunately the range where the supply impdance has a large impact on the size of the ripple relative to the signal going out to the drivers. Since the control loop in most amplifiers only has 70 or 80dB PSRR this is why people hear differences between supply caps and various layout techniques. It's also why most THD measurements on power amplifiers don't show supply issues; typically they're conducted at atypically high output powers of 10W or more where the quality of the supply isn't particularly important.

The next part of this is interaction between channels within an amplifier via the power supply. In a typical two channel amplifier both channels share a single set of reservoir caps. So the supply ripple is a function of the combined output power of both channels. If both channels happen to be playing at the same level this isn't so bad, but usually there's a few dB difference at any given moment between the left and right stereo channel. This is what the Pvictim bit in the figure is about---whichever channel's playing quietly ends up being "victimized" by the louder channel's power consumption creating ripple in the supply. I've chosen 10dB here as a representative number but there are cases, such as biamping with an active crossover, where one's likely to want to be significantly more conservative.

So, from a power supply design standpoint, what all this means is---as a rule of thumb---one needs a way of keeping ripple voltages on the order of 40dB larger than the music off of a power amplifier's output. What exactly off the output means is a matter of personal preference but personally I like to design for 100dB down to keep the IMD floor nice and low. That means one wants PSRR on the order of 140dB, which in turn means it's desirable to operate the control loop from a regulated supply. The LM3x7 parts are attractive here due to their relatively good ripple rejection, the ability to stabilize their reference voltages with the 10uF cap discussed in the last few posts, and their low cost.

If there's demand for it I can write up a preamp version of this analysis, but I think you get the idea. One also needs to check on the regulator's load transient behavior but, usually, the control loop operates in class A and it doesn't matter much.
Attached Images
File Type: png PsrrRequirements.png (23.2 KB, 1330 views)
  Reply With Quote
Old 27th November 2011, 02:25 AM   #14
simply14prem is offline simply14prem  Switzerland
diyAudio Member
Join Date: Jan 2011
TDA 2040 is rated max supply at 20v.If I use 7818,7918 or 7815,7915 from 30v power supply ,can it drive TDA 2040 to the optimum level ?.78xx,79xx are rated for 1.5A.Or should I parallel each regulator for higher current output.I dont know how much current TDA 2040 actually needs.Im right now running too TDA 2040 from 7815,7915 regulator .

Last edited by simply14prem; 27th November 2011 at 02:30 AM.
  Reply With Quote
Old 27th November 2011, 07:46 PM   #15
CBS240 is offline CBS240  United States
diyAudio Member
CBS240's Avatar
Join Date: Aug 2005
Location: K-town
I like to do something like this.....Simple J-fets.

I happen to have need for +/-24V to +/-15V ~100mA as well.
Attached Images
File Type: png Vreg.png (16.0 KB, 1232 views)
All the trouble I've ever been in started out as fun......
  Reply With Quote
Old 27th November 2011, 08:07 PM   #16
sofaspud is offline sofaspud  United States
diyAudio Member
sofaspud's Avatar
Join Date: Nov 2005
Location: San Antonio
Originally Posted by simply14prem View Post
TDA 2040 is rated max supply at 20v.If I use 7818,7918 or 7815,7915 from 30v power supply ,can it drive TDA 2040 to the optimum level ?.78xx,79xx are rated for 1.5A.Or should I parallel each regulator for higher current output.I dont know how much current TDA 2040 actually needs.Im right now running too TDA 2040 from 7815,7915 regulator .
The TDA2040 is internally limited to 4 amps. Rather than parallel 78xx series chips, I'd look into adding a pass transistor or some of the newer vreg offerings with higher power output.
It is error only, and not truth, that shrinks from enquiry. - Thomas Paine
  Reply With Quote
Old 27th November 2011, 08:24 PM   #17
gearheadgene is offline gearheadgene  United States
diyAudio Member
gearheadgene's Avatar
Join Date: Jul 2005
Location: CT
Originally Posted by jitter View Post
Opamps themselves have a very good power supply rejection ratio (PSRR). That PSR suppresses the very little noise that might still be there after the regs even quite a bit more.
I agree - but the story doesn't end there. So I'd like to add a little bit here -

Yes, opamp psrr is very good. You may see this spec'd at 70db or more. *but* - there are caveats:

1) PSRR is not always the same on both rails. Many opamps have better performance on the positive rail, some on the negative rail. Check the data sheets and you will see. Those that spec psrr+ and psrr- are the best, those that spec only psrr+ scare me.

2) PSRR is frequency dependent. Some, not all, manufacturers provide curves of PSRR vs frequency. It falls off very quickly. For audio, maybe not an issue - but check the specs.

3) Some opamps have a very nice (joke) unfiltered path from the -rail to the output. Something to be aware of.
  Reply With Quote


Simple Low Noise Voltage Regulator?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
Best low noise regulator? analog_sa Power Supplies 573 31st December 2016 10:00 AM
Simple Low Power Voltage Regulator rjm Power Supplies 18 28th September 2005 03:48 PM
Ultra low noise JFET voltage regulator Kochkurov Maxim Digital Source 6 25th January 2004 11:33 AM
ForSale: LT1085CT Low noise Regulator tone Swap Meet 0 10th September 2003 06:22 AM

New To Site? Need Help?

All times are GMT. The time now is 01:48 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio