Author Topic: LM317AHVT and Heatsink  (Read 15532 times)

0 Members and 1 Guest are viewing this topic.

Offline mike_mikeTopic starter

  • Super Contributor
  • ***
  • Posts: 1033
  • Country: ro
Re: LM317AHVT and Heatsink
« Reply #75 on: December 14, 2019, 07:51:35 pm »
Probably I will not use the power supply in constant current mode... I need only some protection for the PNP transistors which is available through the 2 1N5408 diodes.
The next question is: if I want to use a 100VA/24V transformer for this project, how can I be sure that in all situations, the output current will not be more than 3 A ?
I checked this way: the maximum current through LM317 is about 2.2A (maximum) and the current through the PNP's is about 1V/1.2R=0.833A. For 2 PNP's it is 1.66A. So the maximum output current will be 2.2A + 1.66A = 3.86A, but I am getting more than 3.86A when I have a overload at output. For example, I am getting 2.5A through LM317T ... and the total output current more than 4A...
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: LM317AHVT and Heatsink
« Reply #76 on: December 14, 2019, 07:58:25 pm »
I tested the attached schematic to see if the output voltage rise when the transformer is unplugged from the mains.
The result was that the voltage does not rise when powering a load that sink a current which is below the maximum output current.
If the load sink more current than the maximum output current then when unplugging the transformer from the mains then the output voltage rise.

What you are describing is completely normal.  The LM317 and most integrated regulators include foldback current limiting which lowers the current limit when the input to output voltage is high to protect the regulator from high power dissipation and secondary breakdown of the output transistor.  This has a side effect of preventing startup into a high current load.

Quote
I made the above mentioned tests using LM317 AHVT from On Semiconductor. Will the behavior of the circuit remain the same if I will use LM317T instead of LM317AHVT ?

All LM317s and most integrated regulators include foldback current limiting and will display the behavior that you describe.
 

Offline mike_mikeTopic starter

  • Super Contributor
  • ***
  • Posts: 1033
  • Country: ro
Re: LM317AHVT and Heatsink
« Reply #77 on: December 15, 2019, 09:53:01 am »
I made some tests to see if the output voltage rise when the power supply is disconnected from mains.
I have got the attached results:
with no load at output: 0097.jpg
with about 2A load: 0112.jpg
with overload: 0124.jpg
The output voltage was 25V in all 3 cases.

If I will make a higher current power supply, for example 10A (with more power transistors and a bigger transformer), then the power supply behavior will remain the same as described in this reply ?
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: LM317AHVT and Heatsink
« Reply #78 on: December 15, 2019, 12:49:27 pm »
The LM317 datasheet includes a graph of maximum output current versus input-to-output voltage as shown below.  Solutions to this problem include:

1. Use multiple 317s in parallel to limit current of each one to about 1/4 amp.  The lab power supply design in National Semiconductor linear brief 28 uses this method to some extent.
2. Use an external power pass transistor to multiple the current rating of the LM317.  Note that this moves the problem to the external transistor which needs to be derated accordingly but there are many single power transistors with a secondary breakdown rating which can support more than 1.5 amps at more than 30 volts.  Multiple transistors in parallel may still be desirable to handle the power dissipation.
3. Use an external power pass transistor in a cascode configuration to limit the input-to-output voltage across the LM317 to a maximum of about 12 volts.

In the case of 1 and 2 above, the built in LM317 current limit will be much higher than the power supply can support so it may be desirable to add a lower external current limit circuit.
 

Offline mike_mikeTopic starter

  • Super Contributor
  • ***
  • Posts: 1033
  • Country: ro
Re: LM317AHVT and Heatsink
« Reply #79 on: December 15, 2019, 04:50:17 pm »
Can you please also have a look at the below reply and clarify the answer ?
If I will make a higher current power supply, for example 10A (with more power transistors and a bigger transformer), then the power supply behavior will remain the same as described in this reply ?
« Last Edit: December 15, 2019, 06:51:38 pm by mike_mike »
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: LM317AHVT and Heatsink
« Reply #80 on: December 15, 2019, 07:09:53 pm »
Can you please also have a look at the below reply and clarify the answer ?
If I will make a higher current power supply, for example 10A (with more power transistors and a bigger transformer), then the power supply behavior will remain the same as described in this reply ?

It all depends on the current level and voltage across the LM317.
 

Offline mike_mikeTopic starter

  • Super Contributor
  • ***
  • Posts: 1033
  • Country: ro
Re: LM317AHVT and Heatsink
« Reply #81 on: December 15, 2019, 07:55:03 pm »
This means that the output voltage is possible to rise when disconnecting the power supply from 230V mains, if the load is 9A, although the voltage didn't rise when the laod was 3A ?
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: LM317AHVT and Heatsink
« Reply #82 on: December 15, 2019, 08:03:35 pm »
This means that the output voltage is possible to rise when disconnecting the power supply from 230V mains, if the load is 9A, although the voltage didn't rise when the laod was 3A ?

Yes, that is possible.  When the AC input is disconnected, the input DC voltage to the regulator falls gradually.  As it falls, the current limit increases so the output current increases while the input voltage is falling.
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: LM317AHVT and Heatsink
« Reply #83 on: December 15, 2019, 08:27:10 pm »
And of course the output voltage will not rise above its setting.
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17429
  • Country: us
  • DavidH
Re: LM317AHVT and Heatsink
« Reply #84 on: December 15, 2019, 10:55:18 pm »
And of course the output voltage will not rise above its setting.

It should not but some power supply designs have flaws.  A bare LM317 will not have a problem.

Sometimes failsafe circuits are added which clamp the drive to the output transistor during power up and power down.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf