Problem with LM317 adjustable regulator

[thebadtall]

hello, i have a problem with lm317 adjustable regulator.
when i supply it with 12vdc it is fine,and adjustable down to 1,5v
but when i supply it with 35vdc and trying to adjust it drops only down to 34.5 and the 220ohm resistor (adjust - out ) burns and smokes.

the circuit is the test circuit of the datasheet
the capacitors are 50v rated
the psu are both smps

thank you

 

[enjunear]

Re: problem with regulator

We need more information about your circuit in order to best help. You should draw a circuit diagram (use Paint, Powerpoint, Visio, any schematic capture tool or simulator...) so we can see all of the circuit device values.

An easy guess is that you're getting too much current going through your 220 ohm resistor (which is why it's burning up). Recall that power dissipated in a resistor is P = V*I = V^2/R = I^2*R, where V is the voltage drop across the resistor. You may need to scale up the value of both resistors in your reference voltage divider (R1 and R2).

See sheet 7: TI LM317 datahseet

 

[thebadtall]

Re: problem with regulator

i destroyed already 2 ics to realise that something is wrong, and i dont know what! maybe i use fake ic's?
undortunatelly i cannot currently draw , i am not using a comouter at the moment. will tomorrow, however its the typical test circuit of datasheet, similar of the one page 7 abobe, without adj cap and diodes.

while this circuit is good with 12v it burns instantly the ics at 35v, and it is rated for 40v, anyway , i will post the schematic thank you

 

[Jestin_cubetech]

Re: problem with regulator

try with high current rating LM317 ic

 

[crutschow]

Re: problem with regulator

What are the values of your two voltage adjustment resistors?

 

[Jestin_cubetech]

Re: problem with regulator

R1 Should be 12E >= R1 <= 240E

 

[thebadtall]

Re: problem with regulator

r1 is 220ohm and r2 10k pot. log.

 

[Jestin_cubetech]

Re: problem with regulator

is it high current rating LM317 ?

 

[thebadtall]

Re: problem with regulator

no its the typical 1.5A but why is that a problem since there is no load at all , and the pmax owf psu is 500ma@35v

 

[godfreyl]

Re: problem with regulator

Is your circuit the same as the one Jestin posted? If so, it's impossible for R1 to burn unless you wired it up wrong or the potentiometer is short-circuit.

 

[Jestin_cubetech]

Re: problem with regulator

make sure the pin out is correct.

 

[thebadtall]

Re: problem with regulator

i am sure that is correct, i have made 20 or more circuits with lm317 this summer , all the same way, for tube filament supply.
thats the same circuit as posted above. the pinout is ok(watching the ic from the front is Adj Out In) . the potentiometer is horizontial, like those in graphic equalisers. i bought this pot 10+ years ago, it is fully dusted, but i measure it with multimeter and it has resistance range from 50 to 13500 ohm.
maybe is the ic's fake (i get them for $.23 each)?

 

[alexan_e]

Re: problem with regulator

I don't think it is is a fake IC, first of all because it works fine with low voltages and secondly there is no point in making cheap fake IC compared to more expensive opamps for example.

I'm not sure what goes wrong if your circuit, the output should always be about 1.2v higher that the ADJ pin and the OUT to ADJ voltage should be equal to the REF voltage (about 1.2v).

Have you measured the voltages in the pins?

 

[godfreyl]

Re: problem with regulator

If R1 is getting burned, there's a fairly high current flowing through it. The question is: from where to where? If there's no load, the current must be flowing through R2. It doesn't matter how badly damaged the LM317 is or what it's doing.

The pot (R2) must be set to quite a low resistance for R1 to burn. I'm surprised the pot didn't start smoking before the resistor.

 

[tpetar]

Re: problem with regulator

Maybe you are give AC to LM317 without Graetz rectification. :-x

 

[thebadtall]

Re: problem with regulator

you mean that my circuit is bad. it may be, i am human ! (but why it works with low voltage,i dont know)
anyway, to not bother you with this issue anymore, i will make another circuit, with new parts at this time, and will post results!

 

[tpetar]

Re: problem with regulator

Try this, and of course adjust voltage to your needs :

http://www.electronicecircuits.com/electronic-circuits/lm-317-power-supply

And replace that smoked LM317 with new.

Keep in mind :

:wink:

 

[thebadtall]

Re: problem with regulator

thank you , i have used this for "improved ripple rejection"
the lm317 is not smoked, the 220r resistor is.
the 317 doesnt function because in the circuit it has vin=vout

 

[alexan_e]

Re: problem with regulator

If R1 is getting burned, there's a fairly high current flowing through it. The question is: from where to where? If there's no load, the current must be flowing through R2. It doesn't matter how badly damaged the LM317 is or what it's doing.

The pot (R2) must be set to quite a low resistance for R1 to burn. I'm surprised the pot didn't start smoking before the resistor.

R2 can't create this problem, even if it is set to 0 the output voltage will be still be 1.25v , the voltage across R1 will be equal to Vref (1.25v) while the regulator is operating properly no matter what the output voltage is.
I also don't think R2 can be destroyed even if set to 1 or 2ohm, the current across is not high enough.

Alex

 

[jeffrey samuel]

Re: problem with regulator

thank you , i have used this for "improved ripple rejection"
the lm317 is not smoked, the 220r resistor is.
the 317 doesnt function because in the circuit it has vin=vout

resistor blows up only if the power dissipated across it is very high so kindly take care and also post the current flowing through the resistor for the voltage

the biasing error is also possible try going through the same ckt again as some errors are overlooked when we repeat an act over many times