Author Topic: Help with lm317 power supply smoking potentiometer  (Read 16222 times)

0 Members and 1 Guest are viewing this topic.

Offline jeffnebel

  • Contributor
  • Posts: 6
Help with lm317 power supply smoking potentiometer
« on: June 22, 2013, 08:09:33 pm »
Hi guys I am trying to build a 3.5 amp 0-18.5v power supply. I have a problem though with my potentiometer burning out and sometimes R2 turns black. I know its being connected properly. I was thinking it was because the schematic didn't include any protection diodes. Could someone help me with this? One more thing, I would like some advice on making the components run cooler. I didn't know if i should add more pass transistors or anything else. I'm open to alterations of the circuit so it would be more reliable. In the schematic R1 is a 5K pot, R2 is a 220 ohm .5 watt, Q1 is a TIP41C, C1 is a 22uF, C2 is a 3300uf. Thanks in advance
« Last Edit: June 22, 2013, 08:12:25 pm by jeffnebel »
 

Offline Alana

  • Frequent Contributor
  • **
  • Posts: 296
  • Country: pl
Re: Help with lm317 power supply smoking potentiometer
« Reply #1 on: June 22, 2013, 08:47:04 pm »
Check your circuit, proper one for bypass transistor is for example one of those http://ourlibro.com/2012/04/08/how-to-boost-current-in-lm317-regulator-circuit/
changing this should be enough for non-smoking pot
 

Online mariush

  • Super Contributor
  • ***
  • Posts: 3754
  • Country: ro
  • .
Re: Help with lm317 power supply smoking potentiometer
« Reply #2 on: June 22, 2013, 09:01:53 pm »
I don't see how you'd get 0v from that.... you'd get 1.25v - ~30v. You'd need negative voltage on the adjust pin to get less than 1.25v at the output.

And yeah, like Alana says, I don't understand what you're trying to do with that transistor there.

here's example circuits straight from the datasheet : ( http://www.ee.buffalo.edu/courses/elab/LM117.pdf , page 17-18)

 

Offline jeffnebel

  • Contributor
  • Posts: 6
Re: Help with lm317 power supply smoking potentiometer
« Reply #3 on: June 22, 2013, 09:09:01 pm »
Check your circuit, proper one for bypass transistor is for example one of those http://ourlibro.com/2012/04/08/how-to-boost-current-in-lm317-regulator-circuit/
changing this should be enough for non-smoking pot
so I would have to change the Transistor from a TIP41C?
 

Online mariush

  • Super Contributor
  • ***
  • Posts: 3754
  • Country: ro
  • .
Re: Help with lm317 power supply smoking potentiometer
« Reply #4 on: June 22, 2013, 09:32:53 pm »
There's also a couple of resistors at least in the other designs that are important. Just replacing the transistor is not enough.
 

Offline jeffnebel

  • Contributor
  • Posts: 6
Re: Help with lm317 power supply smoking potentiometer
« Reply #5 on: June 22, 2013, 09:52:41 pm »
There's also a couple of resistors at least in the other designs that are important. Just replacing the transistor is not enough.
Could you recommend a good circuit that would work correctly preferably one that uses some of the components that i have already
 

Offline Alana

  • Frequent Contributor
  • **
  • Posts: 296
  • Country: pl
Re: Help with lm317 power supply smoking potentiometer
« Reply #6 on: June 22, 2013, 10:06:54 pm »
Either what i posted or first schematic from mariush's post. Transistor - i used BD912 as power PNP transistor mounted on the same heatsink as LM317. No insulation pads needed.
Or if you do not want to mess with pass transistor check out LM350 - 3A version of LM317 but if i remember correctly its like 5x more expensive.
 

Online mariush

  • Super Contributor
  • ***
  • Posts: 3754
  • Country: ro
  • .
Re: Help with lm317 power supply smoking potentiometer
« Reply #7 on: June 22, 2013, 10:08:36 pm »
The ones in the attachments I posted will work correctly. The extra resistors are there for good reason, try to understand why those matter. There's no magic way to make something work with just what you have right now.

If you don't have LM195 like it says in the picture, search for the datasheet for those power transistors and then search local stores for something compatible.

Same with the other example circuit, that one only needs an opamp, a few resistors and capacitors and two leds - all can be easily found anywhere.
If not locally then on eBay or buy from real stores like Digikey, Mouser, Newark.com  in US,  RS-Components.com or Farnell.com in Europe etc
 

Offline jeffnebel

  • Contributor
  • Posts: 6
Re: Help with lm317 power supply smoking potentiometer
« Reply #8 on: June 22, 2013, 10:54:23 pm »
Either what i posted or first schematic from mariush's post. Transistor - i used BD912 as power PNP transistor mounted on the same heatsink as LM317. No insulation pads needed.
Or if you do not want to mess with pass transistor check out LM350 - 3A version of LM317 but if i remember correctly its like 5x more expensive.
In the schematic you posted what were the values you used for the components? Is there a regulator bigger than a lm350?
 

Offline jeffnebel

  • Contributor
  • Posts: 6
Re: Help with lm317 power supply smoking potentiometer
« Reply #9 on: June 22, 2013, 11:36:01 pm »
would this circuit work?
 

Offline Alana

  • Frequent Contributor
  • **
  • Posts: 296
  • Country: pl
Re: Help with lm317 power supply smoking potentiometer
« Reply #10 on: June 22, 2013, 11:46:33 pm »
R1 = will work anything between 1.5 and 20 ohms, i used 3.3ohm 2W power resistor, R2 = 0 [haven't used it, short], R3 = 5k pot, R4 = 220ohms something bigger than those tiny 1/8W resistors - usually 1/4W is enough here
C1 - main filter cap, 4700uF/50V C2 = 100nF ceramic, C3 = 10uF/35V and i had 10uF/35V and 100nF ceramic between output and ground that are not on this drawing.

I haven't seen a single chip regulator with more current than LM350 but even with LM350 it gets tricky because of max dissipated power  TO220 case chip can withstand.
 

Online mariush

  • Super Contributor
  • ***
  • Posts: 3754
  • Country: ro
  • .
Re: Help with lm317 power supply smoking potentiometer
« Reply #11 on: June 22, 2013, 11:49:36 pm »
Quote
Is there a regulator bigger than a lm350?

There are all kinds of linear regulators, for example you can look here:

http://uk.farnell.com/ldo-voltage-regulators
http://uk.farnell.com/linear-voltage-regulators

http://www.digikey.com/product-search/en/integrated-circuits-ics/pmic-voltage-regulators-linear-ldo/2556290

and select at "output current" 3A or more  (because LM350 is rated for 3A).

Then you can enter "ADJ" in the search box on the left side to see only the adjustable regulators capable of more than 3A.

A good linear regulator I worked with is LT1084, it can do up to 5A.  LT1083 can do 7.5A , LT1085 can do 3A...

Just be aware that just because a linear regulator says it can do 3A or more, it doesn't mean in practice it's possible. 
A linear regulator works by dissipating the voltage difference between the input voltage and output voltage as heat.
So, for example, if you have 30v at the input and you want 12v 1A at the output, the linear regulator will generate (30v-12v)x1A  = 18 watts. Most linear regulators can only do 15 watts, some can do up to 25 watts. 
Linear regulators simply aren't designed to work with such wide voltage difference.

If you really want to have 0-18v or something like that, your best bet would be to find a transformer that has two secondary windings or one that has a center tap.

For example, get a 18v AC transformer with center tap, or one that has two 9v AC secondary windings.
When you want a voltage between 0 and 8v DC, connect the center tap to the bridge rectifier, so that the bridge rectifier receives 9v AC.  The bridge rectifier converts this 9v AC to 9x1.414 = 12.5v, minus about 2v in the bridge rectifier, so you have about 10.5 volts.
The LT108x has a voltage drop of about 1.1 volts, a LT350 or similar needs about 2 volts, so you can reliably output 0-8v DC.
Now, if you want - let's say - 3.3v, then the regulator will dissipate 10.5v - 3.3v = 7.2 watts.  Since the regulator can dissipate 15 watts at best, this means the regulator can do 2 A at best, even if it's rated for 3-5 A.

The design I attached in one of my previous posts, uses those LM195 power transistors for this instead of dissipating power into the LM317 linear regulator... each one of those can dissipate more watts than a simple linear regulator could, and since they're several of them in parallel, that's how that design can do 5 A like it says in the picture.

PS.  A TIP147 transistor like the one in your schematic has a 125w power dissipation rating, much better than 15-25w of a linear regulator... but the wasted power issue still applies. You need a HUGE heatsink to keep that TP147 below 150 degrees.
« Last Edit: June 22, 2013, 11:55:56 pm by mariush »
 

Offline Paul Price

  • Super Contributor
  • ***
  • Posts: 1419
Re: Help with lm317 power supply smoking potentiometer
« Reply #12 on: June 23, 2013, 11:51:07 am »
Just go with a switcher and forget about large heatsinks and burnt resistors and pots.

There are quite a few varieties of the switching chip such as LM2575, LM2696, LM2648 etc.

In every case you just need to find the -ADJ version of the switching regulator chip.

Whatever voltage input to this switcher,  you can get a max regulated voltage output about 3V less then the raw rectified input when the output is drawing full  current output.
« Last Edit: June 23, 2013, 12:06:23 pm by Paul Price »
 

Offline jeffnebel

  • Contributor
  • Posts: 6
Re: Help with lm317 power supply smoking potentiometer
« Reply #13 on: June 23, 2013, 08:05:01 pm »
Just go with a switcher and forget about large heatsinks and burnt resistors and pots.

There are quite a few varieties of the switching chip such as LM2575, LM2696, LM2648 etc.

In every case you just need to find the -ADJ version of the switching regulator chip.

Whatever voltage input to this switcher,  you can get a max regulated voltage output about 3V less then the raw rectified input when the output is drawing full  current output.

So if I bought a lm2596 the other components would stay the same as the lm2576 circuit that you posted. I looked them up and the difference to me seems to be that the two chips oscillate at a different frequency
 

Online mariush

  • Super Contributor
  • ***
  • Posts: 3754
  • Country: ro
  • .
Re: Help with lm317 power supply smoking potentiometer
« Reply #14 on: June 23, 2013, 08:59:28 pm »
The problem with switching regulators is that they're not a magical solution and can't be used instead of linear regulators for all purposes without getting into issues.
For example, you shouldn't use a switching regulator to power some audio opamps or something that requires little ripple and noise.. linear regulators are much better for that.

By design, as they work at 70-110kHz, they make noise, emi etc You also have to be careful about the layout and how the components are connected to the chip ... If you put it on a breadboard, there's a good chance it will be unstable.

If you want good efficiency, you also have to pay attention to what inductor you're using - an inductor selected to  3.3v  3A out of a switcher will not work great to get 12v 2A or something like that. Then you have to be careful about peak currents in the shottky diodes, the inductors etc...  Linear regulators are much easier to make and use, especially for beginners.

Anyway... if you decide to go with a LM2xxx, have a look at the TI site... they have a "Switchers made simple" free software in which you can enter input voltage, output voltage, current, and they recommend a switcher and show component values and everything you need.
Actually, they may no longer have download link for software as they push for the web based stuff, but you can download the software from here: http://www.datasheetarchive.com/files/national/0,1768,6.htm  (scroll down and get sms6.24.exe )

There's also a web based solution on this page on their site: http://www.ti.com/lsds/ti/analog/webench/overview.page .. enter voltages and currents and click on "Start design" and a new window should show up with various LM chips that would do what you entered there... click on open design and you get example circuits (free registration may be needed to actually see design- the software shows it directly)




 

Offline Stonent

  • Super Contributor
  • ***
  • Posts: 3824
  • Country: us
Re: Help with lm317 power supply smoking potentiometer
« Reply #15 on: June 24, 2013, 04:28:52 am »
That's almost exactly the rating of a dell laptop power supply... you could save yourself some trouble....
The larger the government, the smaller the citizen.
 

Offline Paul Price

  • Super Contributor
  • ***
  • Posts: 1419
Re: Help with lm317 power supply smoking potentiometer
« Reply #16 on: June 24, 2013, 01:37:01 pm »
   mariush, I have successfully breadboarded many different switching power supply circuits without any problems, just using thin strips of un-etched printed circuit board for the common ground across one side of the circuit being built. They are not as touchy as you might think in operation, the circuit I show is from a spec sheet and it is a variable power supply.

While a different size inductor can optimize the efficiency and assure continuous mode operation, the use of a somewhat larger inductor than shown will only result in a reduced maximum output current at some voltage output settings and a slower transient response. 

Choosing at least a 5-amp schottky diode, mounting the switcher IC on a small heatsink,  and selecting  an inductor that has a 150% current rating (or better) above the max output current ensures the power supply will protect itself against short-cicruited condItions.

The switcher output is so low in noise and ripple so can to be used to power sensitive low-level amplifiers or any other type of bench power supply work.
 

Offline xelion

  • Contributor
  • Posts: 8
  • Country: dk
  • I write 02 05 0E 01 1F AF
Re: Help with lm317 power supply smoking potentiometer
« Reply #17 on: July 28, 2015, 02:54:03 am »
I don't see how you'd get 0v from that.... you'd get 1.25v - ~30v. You'd need negative voltage on the adjust pin to get less than 1.25v at the output.

And yeah, like Alana says, I don't understand what you're trying to do with that transistor there.

here's example circuits straight from the datasheet : ( http://www.ee.buffalo.edu/courses/elab/LM117.pdf , page 17-18)

This is NOT an app note on the LM317!!!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf