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

LM317 load capacitance
LM317 load capacitance
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 11th November 2017, 02:43 AM   #191
jackinnj is offline jackinnj  United States
diyAudio Member
 
jackinnj's Avatar
 
Join Date: Apr 2002
Location: Llanddewi Brefi, NJ
LM317 load capacitance
Quote:
Originally Posted by dotneck335 View Post
VERY cool! Thank you! Now, how 'bout noise/ripple graphs with the same combinations?
Noise you can see on Jan's website. The LM317 isn't remotely in the same class, by 60dB, as a super-reg. Not even worth discussing.

Bypassing the ADJ pin reduces the LM317 gain to unity. Therein lie the problems with it's imposition of harmonics on the supply rail.
  Reply With Quote
Old 11th November 2017, 06:19 AM   #192
Waly is offline Waly
Banned
 
Join Date: Jan 2011
Location: moderation crapper
Quote:
Originally Posted by DF96 View Post
To a first approximation you can regard the output of a regulator (almost any series regulator) as consisting of:
a DC voltage source
a noise voltage source (mix of white noise and ripple etc.)
a series inductance (possibly with a resistor across it)
a series resistance

Add a capacitor to this and you have a low pass filter, but potentially one with a high peak near the corner frequency. The peak is a peak in both noise and output impedance. You get to choose where in the spectrum you want this peak to be.
Look at the regulator output cap as a shunt frequency compensation (which it actually really is, from a small AC signal perspective). Then, it is easy to understand why a large capacitor would not create oscillation, the unity loop gain frequency will be very low and the phase doesn’t reach 180 there, but the stability comes to the price of the transient load response. It is also easy to understand why actually a ceramic nF range cap is the worst case from a stability perspective. It’s a very common beginners error to add a 100nF low ESR in parallel with the output electrolytic with a high ESR, hoping this will improve the HF performance; this is a safe recipe to get the regulator to oscillate. Very few integrated regulators will stay stable with a ceramic 10nF...100nF at the output, most require >1uF ceramic to stay stable.

A large output capacitor is a red herring; it doesn’t do anything good, but only potentially damages the regulator frequency response (although some electrolytics have a large enough ESR to look resistive at HF around the ULGF). It’s the filter capacitor at the input that plays the “reservoir” role.

My advice is to add at the output the minimum low ESR capacitor that unconditional stability requires. If the value is not in the data sheet, then experiment with, starting from 1uF and go up.
  Reply With Quote
Old 11th November 2017, 10:33 AM   #193
nigel pearson is offline nigel pearson  United Kingdom
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
Quote:
Originally Posted by dotneck335 View Post
Very hard to read those graphs, dude. What Vin-Vout voltage are they done at? Doesn't the 33Ω resistor diminish the regulation?
The resistor is 0R33 and the input voltage is typically 18VDC. I was trying all and everything for fun. I was looking at the TL431 stability grahps and noticed very high or very low capacitance values are stable, the ones in the middle not. So I asked myself what about the LM317. I was delighted to find a small resistance made any capacitor possible. When I said zero ohms 1uF zero is subject to the lead impedance. If devices are surface mount it needs more care, assume a small additional resistor could help.

The final capacitor is not especially to help regulation, it will help. If you notice my hum graphs are not very good. 20 000uF before LM317 and 10 000uF after did nothing to change that. It was a slightly wrongly placed connection. That's a story in itself. The parts were what was at hand.

Below is what pleased me most up to this point. Notice how close the LM317 came and was if critical better. You most likely could remove a few capacitors. These are my notes to me and say" yes I did try that ". TL431 is not a device to try without a scope if wanting to try this. A CCS to the input terminal was no better. The output stability better than guessed. A single MJ3001 T03 darlington was almost as good and most likey would work without complications.

Click the image to open in full size.
  Reply With Quote
Old 11th November 2017, 10:57 AM   #194
nigel pearson is offline nigel pearson  United Kingdom
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
Click the image to open in full size.
Here is the PCB version. The hum is now excellent and is almost identical with the PSU switched off. LD1084 extended the performance down to 200/100V, safely OK at 207V and 253V. As one can subtract 4% from 207V and still meet regs 198.72 was where I wanted to go. This PSU works to 2.2amps so losses are important. LM317 seems by measurement identical. The PCB was hard work. I had to hand etch crude versions before spending money. Years old chemicals put back to work. Dead Bug one was no trouble. I even recycled a failed PCB that never got etched. The first made with PVC tape and ferric chrolide that had set solid.
  Reply With Quote
Old 12th November 2017, 12:50 AM   #195
dotneck335 is offline dotneck335  United States
diyAudio Member
 
Join Date: Dec 2012
Quote:
Originally Posted by jackinnj View Post
Noise you can see on Jan's website. The LM317 isn't remotely in the same class, by 60dB, as a super-reg. Not even worth discussing.
I don't read that in those graphs--the super regulators are definitely much better than the LM317, but by nowhere NEAR 60 db---more like about 35 db.

Last edited by dotneck335; 12th November 2017 at 01:14 AM.
  Reply With Quote
Old 12th November 2017, 10:31 AM   #196
nigel pearson is offline nigel pearson  United Kingdom
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
That PSU of mine must be giving > -145 dB noise as it slams into the analyser limit. Now do a test of a 24 bit signal. I doubt it gets close to - 66dB if using real signals. The warning with super regulators is they might be a Mount Everest if you don't have test gear. As Plain Janes go LM317 can be made to be Beautiful Jane.

The maths I use is 21.6 dB ( 12 V ) + 111 dB ( analyser limit ). 0R33+10000uF is > -20 dB @ 500 Hz. The latter is conjecture except using theory of additive noise. If the two were equal we might see circa - 109 dB. I didn't. It seems resonable to think the filter was working. I got -89.5 dB ripple rejection, One has to guess the output reservoir gave 30 dB. Any battery I tried was worse, even hum was no better!

It's a shame I didn't use 35 V output. It that might have given me 142 dB range. Some would question the absolute reliability of that. I have use of an Audio Precission Analyser ( 2 in fact ). When I have compared the results they are not vastly different and usually within the maths. As NASA told me, Newton got them anywhere in the Solar System and Einstein was not required. I can't give you any of the AP data alas for reasons you might guess. At Harwell in the UK they have an external business dept. They must to complete a project use off the shelf solutions. One project was to make an advanced lens cleaner. Alas a common house cleaning product was 1/3 the price and equal. I always do the same.

My test gear came from the man who possibly designed your phone processor ( when he hung up his boots ). He has a Morris Minor car and lives near Cambridge. That chip cost about $100 000. Not all thinghs cost a fortune that change the world.
  Reply With Quote
Old 12th November 2017, 11:09 AM   #197
nigel pearson is offline nigel pearson  United Kingdom
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
This is good reading if you like the Mount Everest route.
Op-Amp Based Linear Regulators

I have a hunch this is worth a try ( I have no values to offer ). There are various theories on zeners so two shown. This is the very idea I was told in 1973 not to bother with. Mostly that was right. The compound pair has been common since 1957. A 2N3055 in 1973.

Click the image to open in full size.
  Reply With Quote
Old 12th November 2017, 11:30 AM   #198
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: SoCal
Jim Williams on his paper measuring ultra LN regulators LinearTech AN83 (2000) " speak softly but carry a big PNP"
I think that helps low drop out topology more than noise...
Attached Images
File Type: jpg Capture.JPG (38.4 KB, 154 views)
__________________
.
.

Last edited by infinia; 12th November 2017 at 11:38 AM.
  Reply With Quote
Old 12th November 2017, 01:22 PM   #199
jackinnj is offline jackinnj  United States
diyAudio Member
 
jackinnj's Avatar
 
Join Date: Apr 2002
Location: Llanddewi Brefi, NJ
LM317 load capacitance
Nigel -- put a 10k resistor across the input of your FFT and display the results.
  Reply With Quote
Old 12th November 2017, 02:05 PM   #200
jackinnj is offline jackinnj  United States
diyAudio Member
 
jackinnj's Avatar
 
Join Date: Apr 2002
Location: Llanddewi Brefi, NJ
LM317 load capacitance
Quote:
Originally Posted by dotneck335 View Post
I don't read that in those graphs--the super regulators are definitely much better than the LM317, but by nowhere NEAR 60 db---more like about 35 db.
Without Cadj the noise of an LM317 is about 20dB higher.

Trade-off transient performance vs noise.
  Reply With Quote

Reply


LM317 load capacitanceHide 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
Using an LM317/Lm337 PCB to make a zener regulated or capacitance multiplier supply doors666 Power Supplies 4 3rd January 2016 03:09 PM
High capacitance vs load on Bridge rectifiers RobertS61 Power Supplies 26 14th September 2011 11:42 AM
Load capacitance ... how to get it halved? jiiteepee Analogue Source 1 26th July 2010 10:04 PM
Amp sim output load capacitance, how much? cbdb Solid State 9 4th June 2009 04:42 PM
Cartridge capacitance load? I have some questions! Kuja Analogue Source 8 24th July 2005 02:18 AM


New To Site? Need Help?

All times are GMT. The time now is 06:54 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
Wiki