Shop OBEX P1 Docs P2 Docs Learn Events
Having problems with LM317 Adjustable Voltage regulators — Parallax Forums

Having problems with LM317 Adjustable Voltage regulators

RagtopRagtop Posts: 406
edited 2011-08-03 17:48 in General Discussion
I was hoping to use a LM317 Adjustable Voltage regulator to create a 6V line for servos and as input to a 3.3v regulator to run the prop.

I can't seem to get close to the voltages according to the resister values.
http://www.reuk.co.uk/LM317-Voltage-Calculator.htm

So I went to trim pots, the only ones I could find at Radio Shack where 10K. They are a bit hair triggered as only a small part of the range seems usable. I am assuming 2K would have been better.

I dialed in my voltage using a multimeter before connecting 3.3v reg. Great. But when I connect the 3.3V reg, no prop and the outgoing voltage to the 3.3V has dropped to 2.2V with nothing coming out of the 3.3v reg.

Has anybody used these?
«1

Comments

  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 17:10
    I have used them, a lot. They are very reliable, solid performers.
    Their tab is NOT Ground/Vss/Return (the tab is Output). I mention this because lots of people assume tab is Ground and fasten it to Ground.
    They need that 1uF output cap.

    The input voltage needs to be at least 3V more than the output.
  • RDL2004RDL2004 Posts: 2,554
    edited 2011-08-02 17:11
    Which 3.3v regulator are you using? It could be a drop out problem (input voltage too low) or incorrect output capacitor (ESR too high). Really, more info is needed, like a schematic with part numbers.
  • ercoerco Posts: 20,259
    edited 2011-08-02 17:17
    Yep, they're pretty bulletproof. Check & recheck your wiring, your 10K pot should work fine. As long as you're using all 3 pins on the pot, it's just a voltage divider; it will replace R1 and R2 in your schematic. A 2K pot would work exactly the same.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 17:20
    You can't use a/the pot for the 240Ω "R1" (no).
    The 240Ω is fixed, a must.
    "R2" can be a pot (rheostat).
  • ElectricAyeElectricAye Posts: 4,561
    edited 2011-08-02 17:51
    PJ Allen wrote: »
    ...

    The input voltage needs to be at least 3V more than the output.

    PJ has probably hit the nail on the head. Another thing to consider for down the road: the input power supply must be able to give you enough amps, too. You can't put a weenie power supply on the input side and expect to drive an amp or more at the output.
  • RDL2004RDL2004 Posts: 2,554
    edited 2011-08-02 18:01
    You guys did read the part where they state that they are using the LM317 "as input to a 3.3v regulator" ? If this is a low dropout type, it may have a missing or incorrect output capacitor. We have seen this problem before.

    Check and re-check all the wiring, then turn up the LM317 a couple of volts and see if the 3.3v starts working. If not, check the data sheet for the 3.3v regulator, it may require a specific value, low ESR cap on the output.
  • RagtopRagtop Posts: 406
    edited 2011-08-02 18:07
    Thank you all for the replies.

    I am using a:
    12v 200ma walwart
    .1 uf disc cap on the input pin (one end to ground)
    A 220 resister between the adjust pin and the output pin
    1 10K trimpot with wiper to adjust pin and a outer pin to ground
    A 1000uf cap on the output pin (negative side to ground)
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 18:12
    Is the 3V3 supply output disconnected from everything?
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 18:14
    1000uF is a big cap, "probably" OK, but still.

    PE - Push that 'F5' button once in a while or we'll be here all night.
  • RagtopRagtop Posts: 406
    edited 2011-08-02 18:17
    PJ Allen wrote: »
    Is the 3V3 supply output disconnected from everything?

    Yes. Nothing is connected to the 3.3v reg output currently.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 18:17
    So, when the 317 is connected to the 3V3 supply, what happens to the 12V supply?
  • RagtopRagtop Posts: 406
    edited 2011-08-02 18:21
    I haven't noticed an input change. The walwart seems to spit out 14-15v. Just the output pin goes from 6v to 2.2v when connected to the 3v3.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 18:24
    Can you run some LEDs from the 317, draw some current from it, and keep it in regulation?
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 18:26
    If you're not running anything from the 3V3, it should be OK with the 12V supply as its input.


    PE -- "01:38 AM" Any progress?
  • RagtopRagtop Posts: 406
    edited 2011-08-02 18:40
    With just led attached the 317 output pin drops 1v.

    Could it be the 3.3v is defective?
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 18:41
    How about with LEDs (and resistors, OK) connected to the 317 output (no 3V3 connected)?
  • RagtopRagtop Posts: 406
    edited 2011-08-02 18:46
    PJ Allen wrote: »
    How about with LEDs (and resistors, OK) connected to the 317 output (no 3V3 connected)?

    Yes. That is what I did. The led lite and the 317 output pin dropped 1v from when it had nothing attached.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 18:48
    ???
    You said, "With just led attached the 317 output pin drops 1v. Could it be the 3.3v is defective? "

    The 3V3 is not the 317, the 317 is the "6V".
    Get rid off the 1000uF
  • RagtopRagtop Posts: 406
    edited 2011-08-02 18:51
    PJ Allen wrote: »
    ???
    You said, "With just led attached the 317 output pin drops 1v. Could it be the 3.3v is defective? "

    The 3V3 is not the 317, the 317 is the "6V".
    Get rid off the 1000uF

    No, I meant could the 3.3v be my problem and not the 317.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 18:53
    "We" need to know if the 317 is a good circuit -- all by itself, divorced from the 3V3 situation.
    Not just with a no-load output, but when it draws some current, does some work, with no 3V3 attached.
    Does that make sense?
  • RagtopRagtop Posts: 406
    edited 2011-08-02 19:02
    I think I am just going to have to start from scratch in the morning. Now it is outputing 7v no matter what I set the trimpot to.
  • RagtopRagtop Posts: 406
    edited 2011-08-02 19:04
    PJ Allen wrote: »
    "We" need to know if the 317 is a good circuit -- all by itself, divorced from the 3V3 situation.
    Not just with a no-load output, but when it draws some current, does some work, with no 3V3 attached.
    Does that make sense?

    Yes. That is what I did. 317 output pin to 330 resister to LED to ground. Dropped 1v from when nothing attached.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 19:06
    Get the 317 working by itself, test it with some load on the output, a few LEDs works for me.
    The output cap for it should be 1uF, something low like that.
    It should stay in regulation, only change by millivolts (not volts.)

    Then you can try to run the 3V3 circuit.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 19:08
    The 317 output dropping to 1V is no good.
    The output dropping by 1V (from 6V to 5V) is no good.

    A-O-I (adjust, output, input), looking at the face.

    Measure the input voltage, too, make sure it stays up.
    Take data.
  • RagtopRagtop Posts: 406
    edited 2011-08-02 19:09
    Thank you, I will try it with 1 uf in the morning.
  • RagtopRagtop Posts: 406
    edited 2011-08-02 19:11
    PJ Allen wrote: »
    A-O-I (adjust, output, input), looking at the face.

    ARGH!!!!!!!!! The adjust pin is on the outside. I can't believe I missed that.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 19:12
    And what were you doing?
  • RagtopRagtop Posts: 406
    edited 2011-08-02 19:15
    I guess I was thinking it had the same layout as the 3.3v
    Input, Ground or adjust, and output.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-02 19:17
    No - you look at the datasheet!
    It's A-O-I, and like I posted early on, the tab on it = the output.
  • kwinnkwinn Posts: 8,697
    edited 2011-08-02 21:33
    Ragtop, with a 220 ohm R1 you need an 880 ohm (not a standard value) R2 for a 6.0V output. Using a 1K trimpot instead of 10K would make adjusting the voltage much easier. Using a 120 ohm R1 and a 470 ohm or 499 ohm R2 would give you 5.9V or 6.2V respectively. The 470 ohm resistor is a standard value and quite easy to find, and although not a standard value the 499 ohm resistor is also easy to find. I use the 120/499 combo to provide 6.2V to several motors and other circuits that require 6.0V and have had no problems to date.
Sign In or Register to comment.