LM317/337 Series Tracking Preregulator - preloading with CCS. - diyAudio
Go Back   Home > Forums > Amplifiers > Power Supplies
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

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 March 2016, 08:49 PM   #1
tybysh is offline tybysh  Poland
diyAudio Member
 
Join Date: Apr 2014
Talking LM317/337 Series Tracking Preregulator - preloading with CCS.

Hi.

First of all - I'm new here but I was reading this forum for quite long time. Sorry for my bad grammar - I'm not an native speaker.

Recently I'm designing yet another op-amp based headphone amplifier for my DIY collection and I want something simple yet nice in power supply department. I have successfully built tracking preregulators in the past - no problem with that. Nonetheless I have quite a few questions for more experience users.

This is my schematic based on article on Acustica, as You can see there is another tracking preregulator after the first one to obtain better power supply rejection ratio.

Click the image to open in full size.

C9, C12, C15 and C16 are optional, I don't know if there is a need for them.
For the sake of the argument I just show You the positive reg, the same rules applies to negative one.

So, my questions:

1. Is it worth it to load the output of this regulator with CCS for a constant current draw? In the datasheet LM317/337 is measured with 500mA current draw in mind, I suppose this will minimize noise level?
2. If loading the output is a good idea - will simple resistor do the job (high wattage), or simple arrangement with LM317/337 + resistor? I'm afraid it will inject additional noise from CCS to the output, am I mistaken?
3. Last question - what is Your opinion about putting a simple capacitance multiplier with darlington transistors arrangement to put the noise level even further down? In the noise compartment it would immensely help with lowering the noise but what about output impedance - should I be afraid of it or should good supply bypassing of the op-amp take care of high impedance on power trace?

Thanks for the help

Last edited by tybysh; 17th March 2016 at 09:14 PM.
  Reply With Quote
Old 17th March 2016, 09:17 PM   #2
tybysh is offline tybysh  Poland
diyAudio Member
 
Join Date: Apr 2014
Here is link for schematic, I can't edit the first post anymore
http://s14.postimg.org/qbft70uox/trackingprereg2.jpg
Click the image to open in full size.
  Reply With Quote
Old 18th March 2016, 06:05 PM   #3
diyAudio Member
 
Join Date: Jul 2009
I've built quite a few of these tracking pre regs and am familiar with the very informative Acoustica pages on the subject. First off there a couple of issues with your schematic.

C9 and C15 if present, should not be connected to 0v but in fact to the output ends of R2 and R12. R2 and R12 could be replaced by one or a pair of LED's to further reduce the noise gain of each pre reg by reducing the dynamic impedance further. They will also tell you if the thing Is working.

You should be careful with the ESR of the capacitance gathered at the output of each TPR. Too low and it will cause a noise peak at the frequency at which the combined capacitance oscillates with the effective output inductance of the TPRs. The 10uF will likely be OK for ESR if you use a Tantalum or a high ESR Electrolytic but C22 and C24 may need either some added ESR or a good deal of wiring resistance/inductance to stop them from interacting with the regs.

Loading these regs to 500ma reduces the effective output inductance and so makes them less likely to have a noise peak interaction with the load capacitance but 500ma is more than necessary. the output inductance drops to below 2uH above 20mA and 1.5uH at 30mA. At 500mA it is still above 1uH so there is little to be gained from more than 30mA.

If your load is constant current (class A) then a cap multiplier would be OK. You might want to try inserting one either before the composite reg or after it. Putting it before the composite reg will prevent a lot of HF mains rectification hash from getting through to upset the bandwidth limited 317 regs.

John
  Reply With Quote
Old 18th March 2016, 06:49 PM   #4
tybysh is offline tybysh  Poland
diyAudio Member
 
Join Date: Apr 2014
I thought to use C9, C12, C15 and C16 as a ADJ pin bypass to ground, like in normal LM317 application, what's wrong with that?

I have perfboarded Acoustica tracking prereg once with LED in place of 720R and it burned after couple of seconds so I'm little afraid to put it in the final design, perhaps i should test it again.

Load will not be constant - AD8397, great op-amp for driving headphones. I'm thinking about inserting this simple capacitance multiplier + loading it with resistor to ground to force class A for power supply alone (lowering cap multiplier zOut):

Click the image to open in full size.


So power supply would be class A biased with 100mA (+/-12V rails) but op-amp itself will change total current draw, should I abandon cap multiplier and stick to just 2 preregs?

Capacitors on the output of regulators will be: 10uF tantal (rather high ESR) + 100uF electro Elna Silmic II (again -not very low ESR).

Right now my amp has about 10 000 uF (xD) after LM317 prereg and it sound good but I want to build it with good transient response in mind this time.

Thank You for response

Last edited by tybysh; 18th March 2016 at 07:16 PM.
  Reply With Quote
Old 18th March 2016, 11:03 PM   #5
diyAudio Member
 
Join Date: May 2011
Location: Silicon Valley
Instead of cascading four IC voltage regulators to get good ripple rejection at low frequencies, why not delete one of them and use its voltage headroom to implement RF filtering instead? Here's a quick-and-dirty demonstration, which uses up only 1 volt of headroom (500mA x 0.68R x 3), far less than a single LM317. It attenuates mains-borne RF noise quite nicely in my opinion, starting at a frequency well below the knee in the LM317's rejection curve.

_
Attached Images
File Type: png schematic.png (14.4 KB, 283 views)
File Type: png RFI_atten.png (8.0 KB, 275 views)
File Type: png LM317_ripple_plot.png (48.9 KB, 65 views)
  Reply With Quote
Old 19th March 2016, 12:12 AM   #6
tybysh is offline tybysh  Poland
diyAudio Member
 
Join Date: Apr 2014
Mark Johnson - the thing is that I need some voltage drop - I have symmetric +/-17AC toroid that I would like to utilize with this project, so it is at least +/-10V dc to drop

I'm also using nice power socket with fuses and filter inside:
Click the image to open in full size.

I have almost finished pcb project in Eagle so 4 regs are here to stay, I'm only interested in capacitor multiplier issue, is it worth it or not, this is something that I will check tomorrow - cap multiplier on protoboard and how it will impact sound quality. What will win, low noise floor vs higher supply impedance...

I'm also very curious about adding ADJ pin bypasses (C9, C12, C15 and C16 mentioned before), article in Acoustica omits them or maybe just forget to put them in schematic?

Last edited by tybysh; 19th March 2016 at 12:25 AM.
  Reply With Quote
Old 19th March 2016, 12:23 AM   #7
tybysh is offline tybysh  Poland
diyAudio Member
 
Join Date: Apr 2014
One more thing - shouldn't capacitor multiplier act as a low pass filter with freq cutoff determined by input resistor to capacitor ratio? If so, there is no need to apply this complicated RF filtering, right?
  Reply With Quote
Old 19th March 2016, 12:49 AM   #8
diyAudio Member
 
Join Date: Jul 2009
It is only the caps of the tracking pre-regulators C9 and C15 that are incorrect IMHO. These tracking pre regs are not referenced to 0v but to the voltage setting regs output. By bootstrapping the input to the regulated output these pre regs relieve the second regs of line regulation duty. The way they are connected at the moment means that the tracking pre regs will only track the output at low frequencies. All is explained, including the connections of these caps, on the Acoustica.org site.

FYI you can also improve load regulation by separating the 0 volt feed to the load into power 0v and sense 0v. To do this the only connection from the bottom of C16 and R15 should be to the 0v and the load. The power 0v connection should return from the load straight to the power source, thereby not causing common mode interference with the sense 0v line. I've done this on the regs in many amps and the resulting improvement can be heard.

John
  Reply With Quote
Old 19th March 2016, 01:02 AM   #9
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Blog Entries: 142
Send a message via MSN to abraxalito Send a message via Yahoo to abraxalito Send a message via Skype™ to abraxalito
Quote:
Originally Posted by Mark Johnson View Post
It attenuates mains-borne RF noise quite nicely in my opinion, starting at a frequency well below the knee in the LM317's rejection curve.
Which capacitor (2.2nF) did you find with 60pH of ESL? I went over to TDK's website and checked out an X5R 2.2nF in what I figured was the smallest possible package (0201) and its SRF is about 150MHz indicating somewhat higher than 60pH ESL.
__________________
The heart ... first dictates the conclusion, then commands the head to provide the reasoning that will defend it. Anthony de Mello
  Reply With Quote
Old 19th March 2016, 01:36 AM   #10
diyAudio Member
 
Join Date: May 2011
Location: Silicon Valley
Quote:
Originally Posted by abraxalito View Post
Which capacitor (2.2nF) did you find with 60pH of ESL? I went over to TDK's website and checked out an X5R 2.2nF in what I figured was the smallest possible package (0201) and its SRF is about 150MHz indicating somewhat higher than 60pH ESL.
I calculated it from a 680 pF SMD capacitor's Z-vs-f curve, then assumed a 2200pF cap would have similar inductance. Maybe I muffed the calculation.

Last edited by Mark Johnson; 19th March 2016 at 01:40 AM.
  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
LM317 Based Tracking Preregulator Dual PSU Uncle Paul Power Supplies 2 29th May 2015 01:49 AM
LM338 tracking preregulator and current limit Rodeodave Power Supplies 9 17th July 2013 05:32 AM
Maida tracking preregulator? kacernator Tubes / Valves 16 15th January 2013 04:55 PM
Lm317/337 tracking preregs 5th element Solid State 12 3rd March 2006 11:29 PM
Isolating TO-3 LM317/337 mik Parts 2 14th March 2004 12:37 PM


New To Site? Need Help?

All times are GMT. The time now is 09:49 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Copyright 1999-2016 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki