Help designing LM338 PS - diyAudio
Go Back   Home > Forums > Amplifiers > Power Supplies

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 12th April 2015, 08:09 PM   #1
diyAudio Member
 
Join Date: Sep 2010
Default Help designing LM338 PS

I need to build a linear PS with input = 25VDC and output = 5VDC @ up to 4A. It looks like a LM338 will do the job as it has Vi-Vo <35V and Io = 5A. In looking at the datasheet I have some questions. How do I calculate the needed heat sink? I do not see data on heat dissipation. It says it has "internal regulation" for max power dissipation, but what does that mean excpet I will be unlikely to kill it? Would it be (Vi-Vo)*Io? That would put me @ 80W! It also appears that the max Io is dependant on the Vi-Vo. At Vi-Vo<10, I can get 8A, but at Vi-Vo=30, only 1. Should I cascade two LM338s dropping 10V with each stage? The heat would, of course, be the same total, but the max Io would be greater. I do not have any experience with the LM338 so any help is appreciated.
  Reply With Quote
Old 12th April 2015, 10:25 PM   #2
rayma is offline rayma  United States
diyAudio Member
 
Join Date: Apr 2011
Location: Upper midwest
Quote:
Originally Posted by dmtparker View Post
Would it be (Vi-Vo)*Io? That would put me @ 80W! Should I cascade two LM338s dropping 10V with each stage?
Yes, for a linear regulator, Pd is just current times the voltage drop across it. A pre-regulator as you mentioned can be helpful
to lower the main pass device dissipation. Sometimes a simple power resistor across the pass device can work about as well.
  Reply With Quote
Old 12th April 2015, 11:07 PM   #3
paulb is offline paulb  Canada
diyAudio Member
 
Join Date: Jun 2001
Location: Calgary
Your power supply will be grossly inefficient. Consider a switching power supply or a different source.
  Reply With Quote
Old 13th April 2015, 12:39 AM   #4
diyAudio Member
 
Join Date: Sep 2010
Quote:
Originally Posted by paulb View Post
Your power supply will be grossly inefficient. Consider a switching power supply or a different source.
I am working with totally pure DC. I don't want to introduce the noise of a switcher.
  Reply With Quote
Old 13th April 2015, 12:44 AM   #5
diyAudio Member
 
Join Date: Sep 2010
Quote:
Originally Posted by rayma View Post
Sometimes a simple power resistor across the pass device can work about as well.
Are you talking about a voltage divider network (R1 in series, R2 to ground)? or something else?
  Reply With Quote
Old 13th April 2015, 01:26 AM   #6
rayma is offline rayma  United States
diyAudio Member
 
Join Date: Apr 2011
Location: Upper midwest
Quote:
Originally Posted by dmtparker View Post
Are you talking about a voltage divider network (R1 in series, R2 to ground)? or something else?
You can connect a power resistor between input and output, and if the design is carefully done the resistor will burn much of the power
that would otherwise have to be handled by the regulator. In other words, some of the load current bypasses the regulator.
This works best if the load current varies relatively little, since the resistor current can never be greater than the load current.
What is your application, a class A amplifier - how much will the load current vary?

Last edited by rayma; 13th April 2015 at 01:30 AM.
  Reply With Quote
Old 13th April 2015, 01:45 AM   #7
diyAudio Member
 
Join Date: Sep 2010
Quote:
Originally Posted by rayma View Post
You can connect a power resistor between input and output, and if the design is carefully done the resistor will burn much of the power
that would otherwise have to be handled by the regulator. In other words, some of the load current bypasses the regulator.
This works best if the load current varies relatively little, since the resistor current can never be greater than the load current.
What is your application, a class A amplifier - how much will the load current vary?
This is to power a Raspberry Pi2 plus I2S DAC. Probably in the 1.5 - 2A range, but varying with access to USB hard drive (where the music is stored). 4A is probably VERY safe upper limit, but continuous 2A fairly likely. A quick look shows 40W R in the $10-15 range, but how do you choose R? I know how to do it with a 2 R voltage divider, but just 1 R across VR?
Thanks
  Reply With Quote
Old 13th April 2015, 01:58 AM   #8
rayma is offline rayma  United States
diyAudio Member
 
Join Date: Apr 2011
Location: Upper midwest
Quote:
Originally Posted by dmtparker View Post
4A is probably VERY safe upper limit, but continuous 2A fairly likely.
Have you checked out the app circuits in the data sheet? Look at the tracking preregulator on p14. This helps if the input voltage varies much.
How much does the input voltage vary over the full range of output current? What is the actual circuit for the input supply for the regulator?
http://www.ti.com/lit/ds/symlink/lm138.pdf

If the minimum load current really is 2A, with peaks up to 4A, you might be able to add the power resistor between input and output
and allow a little less than 2A to bypass the regulator. The resistor value for example could be ( Vin - Vout )/1.8A to allow some
margin for tolerances. Use the value of Vin at a load current of 2A. This would cut regulator dissipation about in half at full load,
and to very little at 2A load. This would give you roughly 10 Ohms at 40W for the resistor. I'd use several 20W resistors in parallel,
maybe five 56 Ohm 20W. This kind of ceramic wire wound resistor is very cheap.http://www.ebay.com/itm/like/2911411...lpid=82&chn=ps

Last edited by rayma; 13th April 2015 at 02:26 AM.
  Reply With Quote
Old 13th April 2015, 02:24 AM   #9
diyAudio Member
 
Join Date: Sep 2010
Quote:
Originally Posted by rayma View Post
Have you checked out the app circuits in the data sheet? Look at the tracking preregulator on p14. This helps if the input voltage varies much.
How much does the input voltage vary over the full range of output current? What is the actual circuit for the input supply for the regulator?
http://www.ti.com/lit/ds/symlink/lm138.pdf
The input is 12 x 2VDC battery which will vary from 24.5 to 28 VDC depending on state of charge (solar) but will not vary at all with load (800 Ah @ 24V). It is part of a totally DC system I am building to include preamp, XO and power amps all powered off a 48V solar bank (0, 24, 48 = +/- 24VDC). I was going to use the standard schematic shown in Fig. 17 but with R2 fixed @ 360 for 5V. I was thinking of preceding it with the exact same schematic but R2~1k for ~12V to split up dissipation. How does the Tracking Pregulator work? With R4=360 for Vo=5V, does VR1 see equivalent R2=720+120+360=1200 so Vo=1.25*(1+1200/240)=7.5? That would put it dissipating the lion's share. Which of the Rs, if any, dissipate significant heat?
  Reply With Quote
Old 13th April 2015, 02:38 AM   #10
rayma is offline rayma  United States
diyAudio Member
 
Join Date: Apr 2011
Location: Upper midwest
Quote:
Originally Posted by dmtparker View Post
How does the Tracking Pregulator work? Which of the Rs, if any, dissipate significant heat?
A tracking preregulator keeps the main regulator input voltage constant, so there are fewer constraints on its design.
You don't want any of the gain setting Rs to dissipate much power (so the values do not drift). However, they are usually
made to draw a certain minimum current from the regulator for proper operation, around 10mA I think.
So, (R1 + R2) = 5V/10mA, or 500 Ohms maximum.

Last edited by rayma; 13th April 2015 at 02:41 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
help designing regulated PS 13v8@50A w/ IRFZ44 kroto Power Supplies 15 18th November 2009 02:11 PM
LM338 Regulated PS Puffin Chip Amps 6 22nd November 2008 05:40 PM
Where to get the lm338 reg. in Europe (for Gainclone regulated PS)? swak Chip Amps 3 15th July 2004 08:29 AM
Designing a motor speed controller/PS - querying motors GUILHERME Analogue Source 11 3rd November 2002 10:09 PM


New To Site? Need Help?

All times are GMT. The time now is 08:29 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2015 DragonByte Technologies Ltd.
Copyright 1999-2015 diyAudio

Content Relevant URLs by vBSEO 3.3.2