LM317T Power supply issues.

Marcus2012

Joined Feb 22, 2015
425
It seems as if there is a short across the regulator from input to output. That would put 38v across the program resistors (and pot).
I don't think that's the case. Blue is the only wire going IN and the Yellow wire and Adjust resistor are the only connections on the OUTPUT
 

Marcus2012

Joined Feb 22, 2015
425
The only thing connected to Vo is the capacitor (which is too large, it should be 47uF). When this switches on-off the only path for the discharge of the capacitor is either through the adjust resistors or the regulator output. It needs diodes and an LED dummy load.
 

eetech00

Joined Jun 8, 2013
3,858
The only thing connected to Vo is the capacitor (which is too large, it should be 47uF). When this switches on-off the only path for the discharge of the capacitor is either through the adjust resistors or the regulator output. It needs diodes and an LED dummy load.
Simple test...temporarily replace the output cap with a load resistor...
 

MrAl

Joined Jun 17, 2014
11,389
No over heating, caps are all ok and rectifiers are wired correctly.
Hi again,

First thing you have to learn is to read ALL the reply posts. You are not reading them all so you are missing some that will help you.
Second, ditch the solderless breadboard. Power circuits have no business on a solderless breadboard. The connections are too unreliable. One connections becomes undone for 10 milliseconds and something blows out.

Good luck with it.
 

Thread Starter

Anubhav Sharma

Joined Mar 21, 2016
68
Hi again,

First thing you have to learn is to read ALL the reply posts. You are not reading them all so you are missing some that will help you.
Second, ditch the solderless breadboard. Power circuits have no business on a solderless breadboard. The connections are too unreliable. One connections becomes undone for 10 milliseconds and something blows out.

Good luck with it.
Thanks sir,
I think i should go for prototype. One thing i remember, earlier i made a power supply ( very small one, just for trial ) using half wave rectifier. The input volt was 18 max and that worked well with trim pot even. Is it possible that the problem is with 37v input / 2 Amp transformer ? LM317 blows/ shorts when i reduce volts using pot.( Thats another thing that volt meter not showing dropping voltage, only pot burns and immediately after that i check regulator and its already damaged. )

If 37v is a problem then please tell me what should i do to reduce input voltage?

Thanks for guidance.
 

Marcus2012

Joined Feb 22, 2015
425
Please read my previous posts and respond, at the moment you have no load in those photographs, the output capacitor is too large and may be drawing too much current and in addition to this the charge from the capacitor has no discharge path other than the output of the regulator or the adjust resistors.
 

eetech00

Joined Jun 8, 2013
3,858
.( Thats another thing that volt meter not showing dropping voltage, only pot burns and immediately after that i check regulator and its already damaged. )
If 37v is a problem then please tell me what should i do to reduce input voltage?
It should work on the breadboard.
Take a new LM317 and check the resistance between the input and output pins before using it. Let us know the reading. Then, replace the output cap with 1K resistor, and try the LM317 without using the heat sink. There should only be a small amount of current flowing in the 1k load resistor (about 40 mA max) so it shouldn't damage the LM317.
 

MrAl

Joined Jun 17, 2014
11,389
Thanks sir,
I think i should go for prototype. One thing i remember, earlier i made a power supply ( very small one, just for trial ) using half wave rectifier. The input volt was 18 max and that worked well with trim pot even. Is it possible that the problem is with 37v input / 2 Amp transformer ? LM317 blows/ shorts when i reduce volts using pot.( Thats another thing that volt meter not showing dropping voltage, only pot burns and immediately after that i check regulator and its already damaged. )

If 37v is a problem then please tell me what should i do to reduce input voltage?

Thanks for guidance.
Hi,

Just for testing, connect the diodes to only 1/2 of the transformer secondary. That should reduce the DC voltage to around 16v or so. This is just for testing, getting it up and running, then later once it is working you can increase the voltage and see if it still works.
You should also measure the AC voltage out of the transformer to make sure it is what you think it is.
 

Thread Starter

Anubhav Sharma

Joined Mar 21, 2016
68
It should work on the breadboard.
Take a new LM317 and check the resistance between the input and output pins before using it. Let us know the reading. Then, replace the output cap with 1K resistor, and try the LM317 without using the heat sink. There should only be a small amount of current flowing in the 1k load resistor (about 40 mA max) so it shouldn't damage the LM317.
Readings between input and output pins 522 ohm
 
Last edited:

Thread Starter

Anubhav Sharma

Joined Mar 21, 2016
68
Ahh ok my bad, I just though you meant the connections because you know.....he'd already mentioned the intrinsic short.........
Actually i have changed the LED location from input side, now its connected to the positive side of output cap with 1K 2watt resistor whih goes to gnd. So i think i have a discharge path for cap, also replaced output cap with 1uf 50v and also connected 22uf 50v cap to the positive side of POT which goes to ground in series with IN4007 to output. Connected IN4007 between Input and output ( - side towards input). Still not working and shorts LM317. I think i should go with prototype now.
 

Thread Starter

Anubhav Sharma

Joined Mar 21, 2016
68
Hi,

Just for testing, connect the diodes to only 1/2 of the transformer secondary. That should reduce the DC voltage to around 16v or so. This is just for testing, getting it up and running, then later once it is working you can increase the voltage and see if it still works.
You should also measure the AC voltage out of the transformer to make sure it is what you think it is.
Already tried last night with 18-0 transformer. Output was 22 v and again shorts and pot burns.
 

MrAl

Joined Jun 17, 2014
11,389
Hello again,

Sorry to hear that. Are you sure you did not get a bad batch of LM317's or something?
This is not something i typically see. I've connected LM317's, LM217's, LM338's, and LT1083's and probably a host of other parts like that and never ONCE had one burn out. That's mostly because the circuit is so simple it's just hard to do it wrong.

This leads me to believe either you are doing something very unconventional, or all the LM317's are bad to begin with. You NEED to test at a VERY low voltage like 10v first. You dont seem to see the importance of this. Keeping the input low keeps the pot from burning out even if something else is not right.

You can test with 2.5v output and only 6v input even. This test VERIFIES the LM317 first. If you dont do that you cant know if the LM317 is good before you try to go full voltage. You dont need a pot, just another 220 ohm resistor. This way you dont burn out the pot.

Try that and see what happens. Make sure the input is very low though, like 6 to 10v. Not 12v, not 13v, not 15v, not 16v, not 18v, not 20v, not 22v, etc. You can even use batteries for this initial test.
 

dannyf

Joined Sep 13, 2015
2,197
You seem to be going out of your way to prevent others from helping you, by ignoring questions, providing incomplete or fuzzle answers, and by not following up.

Your issues are fairly easy to diagnose, as long as you are willing to.
 

Thread Starter

Anubhav Sharma

Joined Mar 21, 2016
68
Hello again,

Sorry to hear that. Are you sure you did not get a bad batch of LM317's or something?
This is not something i typically see. I've connected LM317's, LM217's, LM338's, and LT1083's and probably a host of other parts like that and never ONCE had one burn out. That's mostly because the circuit is so simple it's just hard to do it wrong.

This leads me to believe either you are doing something very unconventional, or all the LM317's are bad to begin with. You NEED to test at a VERY low voltage like 10v first. You dont seem to see the importance of this. Keeping the input low keeps the pot from burning out even if something else is not right.

You can test with 2.5v output and only 6v input even. This test VERIFIES the LM317 first. If you dont do that you cant know if the LM317 is good before you try to go full voltage. You dont need a pot, just another 220 ohm resistor. This way you dont burn out the pot.

Try that and see what happens. Make sure the input is very low though, like 6 to 10v. Not 12v, not 13v, not 15v, not 16v, not 18v, not 20v, not 22v, etc. You can even use batteries for this initial test.
Hi Again,
Tested circuit with 7.5 volts input and am getting 6.74 volts max and 1.26 volts to the minimum side. And nothing burns.
 
Last edited:

vol_

Joined Dec 2, 2015
93
I think that first you got to try this with low voltage input, as others said.
As i saw in the photos, you have connected all the three pins of the potentiometer. Meaning that you connected one pin to ground and you are using it as a potentiometer.
But as I can understand from the schematic, the potentiometer has connected on the circuit one of its side pins and the middle pin. This means that is used as an rheostat.
If am wrong I would love someone to correct me. I will solder tonight an LM338 circuit (which is in the same family with LM317), so this is interests me a lot. It almost excites me.

Also have a look here:
http://electronics.stackexchange.co...erence-between-a-potentiometer-and-a-rheostat

my thread about the LM338 is here:
http://forum.allaboutcircuits.com/threads/lm338-regulator-voltage-and-current-adjustment.122699/
 

Thread Starter

Anubhav Sharma

Joined Mar 21, 2016
68
I have tried using low volts which works prefectly. Also measures out AC volts from transformer 12-0-12 ( connecting 12-12) gives me 27 volts AC, which means there should be a drop after connecting bridge (IN4007), Its producing 22.8 volts after connecting the diodes bridge rect. which is raw DC i guess. Why final output shows 37v and after load its 35.2v ? Can i use 47K pot for that ?. Reducing voltage damages pot and regulator when working on 22 volts. Low volts makes it run perfectly 7.5 volts input and am getting 6.74 volts max and 1.26 volts to the minimum side. And nothing burns.

Also tried parallel cap and resistor, also tried diode for LM317 safety. Nothing works.
I think there is something wrong with component values. i replaced 470uf output cap with 1 uf and also placed 22uf cap in series with another IN4007 ( cap end goes to positive or wiper side then to diode and diode goes to output pin of LM317.

Currently making prototype, as am new to this so it may take little time say by tonight.

Am sorry if i did not reply to certain post but i did everything i asked to do.

Please guide
Thanks.
 
Top