negative voltage regulator problem

[peter g]

hi, I recently built a neg volt reg using an lm337. its the standard basic circuit however, its not working. the reg overheats and the adj pot smokes. this indicates something is wrong. other than the reg itself what could be causing this? confused in florida.

 

[E-design]

You most probably got the pins wired up wrong.

 

[peter g]

You most probably got the pins wired up wrong.

the lm337 is being plugged into a pc board and I checked the tracks and its laid out correctly?

 

[E-design]

Otherwise, you may have a faulty device. It can happen.
Can you post some pictures of the board for us to look at?

 

[peter g]

unfortunately, iam fairly computer illiterate, so can't post pix. I did change the lm337 though with the same result. I set up a test circuit to check the lm337 in my stock and they seem to be good.

 

[E-design]

What input voltage and output load are you testing the regulator with?

 

[peter g]

What input voltage and output load are you testing the regulator with?

the input is 14v no load.

 

[E-design]

I guess you meant -14V? What is the desired output voltage?

 

[peter g]

I guess you meant -14V? What is the desired output voltage?

yes, minus 14v----want -9v.

 

[E-design]

Since this is a standard voltage, you could use a fixed regulator like the LM7909 for the job.
https://www.fairchildsemi.com/datasheets/LM/LM7909.pdf
You don't need any programming resistors and thus fewer chances of wiring or component value mistakes.

 

[peter g]

Since this is a standard voltage, you could use a fixed regulator like the LM7909 for the job.
https://www.fairchildsemi.com/datasheets/LM/LM7909.pdf
You don't need any programming resistors and thus fewer chances of wiring or component value mistakes.

you're right. I just was very curious as to why it wasn't working.

 

[E-design]

Your basic circuit with the 337 should look something like in my diagram.

 

[peter g]

you're right. I just was very curious as to why it wasn't working.

I guess what I was asking is what scenario would have to be present to make the 5k pot burn. other than the reg itself?

- - - Updated - - -

I guess what I was asking is what scenario would have to be present to make the 5k pot burn. other than the reg itself?

actually, there's a 220 ohm res between the adj and out terminals of the lm337 from there to a 5k pot to ground.

 

[E-design]

It must be a 120 Ohm resistor as the lower spec part needs a higher minimum load current to maintain regulation with no external load than the premium LM137.

 

[peter g]

It must be a 120 Ohm resistor as the lower spec part needs a higher minimum load current to maintain regulation with no external load than the premium LM137.

the schematic calls for a 220. i'am starting to strongly suspect that I goofed something when I made the pc bd. i'am going to breadboard the circuit and see what happens. thanks so much for your help. pg

 

[E-design]

Many schematics and on the web get the 120 Ohm vs 220 Ohm wrong. You will find that with a LM337 design using a 220 Ohm that the output voltage will rise somewhat with no external load connected because the regulator can't maintain regulation.

 

[peter g]

Many schematics and on the web get the 120 Ohm vs 220 Ohm wrong. You will find that with a LM337 design using a 220 Ohm that the output voltage will rise somewhat with no external load connected because the regulator can't maintain regulation.

I didn't know that. actually, the circuit we have been discussing is the neg half of a symmetrical p/s.

 

[Audioguru]

Since you used a 220 ohm resistor from Out to Adj then its current is 1.25V/220 ohms= 5.68mA.
Then for a 9V output the pot must be set to only (9V - 1.25V)/5.68mA= 1364 ohms. Then about 1/4 of the pot's resistance track has a power of 44mW which is almost nothing so how can it burn?

I think you should look at the photo of the pins of the LM337 which is different from an LM317.

 

[peter g]

Since you used a 220 ohm resistor from Out to Adj then its current is 1.25V/220 ohms= 5.68mA.
Then for a 9V output the pot must be set to only (9V - 1.25V)/5.68mA= 1364 ohms. Then about 1/4 of the pot's resistance track has a power of 44mW which is almost nothing so how can it burn?

I think you should look at the photo of the pins of the LM337 which is different from an LM317.

I've checked the pinout of both reg many times. the question you pose is interesting. but I swear to you that pot burnt pretty dramatically.

 

[Audioguru]

that pot burnt pretty dramatically.

Then your pcb or LM317 is shorted from the Adj pin to the input or the output.