LM317-based regulated PSU: how does this thing work? - Page 17 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 25th February 2004, 07:35 PM   #161
diyAudio Retiree
 
Join Date: Oct 2002
Location: Spain or the pueblo of Los Angeles
Unhappy Finally?

.... finally, the results are getting reasonable...

It is not the results that were unreasonable as much as some of the posters...... a basic understanding how an emitter follower works and how to bias it, close reading of the LM317 data sheet (and preferably ap notes), and comparing Spice results to the measured data for the circuits shown in the data sheet.

Adjustable three terminals have been around for quite awhile and are excellently documented as to performance. Data sheets and Spice programs do not a designer make without a good grasp of fundamentals. Comparison of data sheet curves and Spice model results are necessary to make sure you of pretty close results from all three.

There are many things to still be covered to examine ways to improve the performance of three terminal regulators which could be explored, but I hesitate to go in to them after the three ring circus this thread became. A Spice demo with a model of the LM317 exist at http://www.micro-cap.co.uk/focmc7.asp for those that would like to pursue Spice modeling of the regulator.

It is very simple to build a test fixture for output impedance and PSRR with a couple of IRF 510s and a fairly low noise op amp configured for an AC gain of 60 dB (1000). A signal source and scope are the only test equipment required. A PC based FFT program is also useful if one knows the circuit is stable. Maybe a new thread on design, simulation, and test of high performance three terminal regulators would be worth starting, free and clear of the morass this thread became.
  Reply With Quote
Old 25th February 2004, 08:51 PM   #162
diyAudio Member
 
Join Date: Dec 2003
Location: Munich
Default less is more

Hi Fred!
Well, I love simple solutions. Only if the results
are not satisfying I am chosing more complex
solutions. So I would use the LM317 in some standard
configuration with cap between Adj and ground....

...spending some more time on proper ground structure...

... less is more ....
It took me ten years to learn this. During the last
7 years this became more and more some sort
of life philosophy for me.

In fact I am also a little bit amazed about the ongoing
personal attacks here.

Cheers
Markus
  Reply With Quote
Old 25th February 2004, 09:31 PM   #163
Banned
 
Join Date: Feb 2002
Location: As far from the NOSsers as possible
Default Does anyone have any doubt about if the circuit is incorrectly implemented?

Maybe no one cares.

Soft start.......slow turn-on......floor wax.......dessert topping.........who really cares what it is called?

Only one person who copied it and did not understand it tries to tell us that it improves the regulator.

Too bad he is not here so that we can make fun of him.

Jocko
  Reply With Quote
Old 26th February 2004, 09:37 AM   #164
tcpip is offline tcpip  India
diyAudio Member
 
Join Date: Jan 2003
Location: Mumbai, India
Quote:
Originally posted by runebivrin
I think we all agree now that you should keep Cadj...
Thanks. Will do, in fact, have already added. I presume 10uF is a good starting point?

Quote:
To keep or not to keep the soft start circuit is a different issue. It does nothing (OK, very little) for regulation and adds to complexity. It is probably not a good idea for an adjustable supply, as it will mean raising the voltage will "soft started", which might not be what you want.
Can you please explain this a bit? I didn't get why it may cause more problems in an adjustable circuit.

Tarun
  Reply With Quote
Old 26th February 2004, 10:15 AM   #165
diyAudio Member
 
runebivrin's Avatar
 
Join Date: Jul 2003
Location: Huddinge, Sweden
Send a message via MSN to runebivrin
When you increase Radj to increase the output voltage, the voltage across the BE junction may - depending on the magnitude and speed of change - start conducting, and the transistor will clamp the voltage on ADJ, until the slow start capacitor is charged.

How much of a problem this is in reality I can't say, but it's something to consider. If you don't need a slow start, why include the circuit?

Another thing to consider with the slow start circuit in a positive/negative combination. The polarities might have substantially different startup times, which may or may not be a problem...

Rune
__________________
Do wizards use spell checkers?
  Reply With Quote
Old 26th February 2004, 01:59 PM   #166
tcpip is offline tcpip  India
diyAudio Member
 
Join Date: Jan 2003
Location: Mumbai, India
Quote:
Originally posted by runebivrin
When you increase Radj to increase the output voltage, the voltage across the BE junction may - depending on the magnitude and speed of change - start conducting, and the transistor will clamp the voltage on ADJ, until the slow start capacitor is charged.
If I've understood you right, this means that if I fix my Radj once and for all, and then switch off and switch on the circuit, there will be no such problem? In that case, I'm not worried, because this circuit is meant to sit inside a closed box, sealed from human contact, once it's been set up. A bit like the bias pot of a Class B amp.

Quote:
If you don't need a slow start, why include the circuit?
I was told that it would be nice to have one if I don't lose anything else in the process. Apparently, it can reduce audible glitches which come from the preamp output at switch-on. Is this true?

Quote:
Another thing to consider with the slow start circuit in a positive/negative combination. The polarities might have substantially different startup times, which may or may not be a problem...
Good point. In fact, I guess that in the extreme case, if one rail comes up and the other doesn't at all, then we risk damaging some downstream circuitry, right? But then I guess in typical setups, since the rail voltages will be equal on both rails, and the component values will be the same, the RC time constants will be equal to +/-25% or so, I should think? That should be safe enough, right?

Thanks for the inputs, though. When I finally get to building the amp, I may choose to not populate the soft-start part for one or two prototypes. Let's see...

Tarun
  Reply With Quote
Old 26th February 2004, 08:08 PM   #167
diyAudio Member
 
runebivrin's Avatar
 
Join Date: Jul 2003
Location: Huddinge, Sweden
Send a message via MSN to runebivrin
Slow start may reduce thumps at switch on. On the other hand, it may take the amp through some unstable voltage region for a longer time than without it, so there's probably no single correct answer. Just try it, I'd say...

Rune
__________________
Do wizards use spell checkers?
  Reply With Quote
Old 27th February 2004, 12:09 PM   #168
tcpip is offline tcpip  India
diyAudio Member
 
Join Date: Jan 2003
Location: Mumbai, India
Quote:
Originally posted by runebivrin
[B]Slow start may reduce thumps at switch on. On the other hand, it may take the amp through some unstable voltage region for a longer time than without it, so there's probably no single correct answer. Just try it, I'd say...
Yes, I will, with inexpensive NE5532 chips. I guess the dangers may be different for different circuits, right? Anyway, let's see what the proof of the pudding looks like...
  Reply With Quote
Old 27th February 2004, 12:20 PM   #169
ALW is offline ALW  United Kingdom
diyAudio Member
 
Join Date: Jun 2001
Location: UK
Default re: jcarr's link

here's the second part: -

http://www.aeng.com/pdf/regulator2.pdf

Andy.
  Reply With Quote
Old 27th February 2004, 03:46 PM   #170
Werner is offline Werner  Europe
diyAudio Member
 
Join Date: Jan 2002
Location: Belgium
I'm out of the country for a couple of days and so I missed most of the fun here.

Seeing this develop was better than Twin Peaks, although from the moment I saw Millwood's original simulations I knew who killed Laura Palmer.

Quote:
Originally posted by jcarr

I located an interesting paper on the effects of the CAdj capacitance and LM317 regulator stability <http://www.aeng.com/pdf/regulator.pdf>.

Nonetheless I would like to know if anyone here has experimented with very much smaller-than-normal capacitances for CAdj,
You may find this link slightly interesting:




TNT measurements of LM317 noise versus Cadjand Cout
  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



New To Site? Need Help?

All times are GMT. The time now is 01:55 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