Shop OBEX P1 Docs P2 Docs Learn Events
Help with the voltage regulator for propeller... — Parallax Forums

Help with the voltage regulator for propeller...

RottenJalapenoRottenJalapeno Posts: 27
edited 2007-03-23 17:16 in Propeller 1
I just had a pcb made that used the LM2937 -3.3·(bought from parrallax's website) to regulate the power going to the prop chip.· I assembled everything and then I noticed it was getting 14v at the output of the regulator.· I did not have a 10uf cap on the output side of it which might be part of the problem but I threw a 10uf cap on the output to gnd and it was still getting a voltage of around 4.2v with all the components on the board but 3.3v if nothing was connected?· I guess this probably doesnt belong on the prop forum but I thought that all of you have used the regulator before to power the prop chips.· What I am wondering is if anyone else had this problem before, and if I'm doing anything wrong.· I've tried like 4 different regulators to see if it was just a bad reg, but all of them did this.· and there is a 5v regulator on it LM2940 which is working fine (I do have caps on the output side of this but it works without them).·

Any suggestions, hints, or mocking would be appreciated!

Comments

  • Mike CookMike Cook Posts: 829
    edited 2007-03-22 23:57

    I'm using this regulator on my Home Brew Propeller Demo Board:

    http://forums.parallax.com/showthread.php?p=579595

    I did notice that it will float above 3.3 if there is not a capacitor on the output, I think the data sheet suggests 10uF. Attached is a diagram of how I have it connected.


    attachment.php?attachmentid=46106

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Mike
    LM2937_3_3.JPG
    735 x 399 - 20K
  • RottenJalapenoRottenJalapeno Posts: 27
    edited 2007-03-23 00:03
    But w/o a cap should it float to 14.5v? and even with a cap, first I tried 10uf, then I put a 100uf cap on and that still made it float to over 4.3v (I'm using electrolitic, will that effect anything?)
  • Mike CookMike Cook Posts: 829
    edited 2007-03-23 00:08

    From what I remember without the capacitor, on the output, I was getting around +4VDC out of it,·input power to the voltage regulator was +9VDC or so.·Also I may have had a LED and resistor connected to the output while I was measuring it so there would be a 'load' connected. When I added the electrolytic capacitor to the output, as the data sheet suggests, the output was a rock solid 3.3.

    What's your input power?


    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Mike
  • RottenJalapenoRottenJalapeno Posts: 27
    edited 2007-03-23 00:12
    my input is 15vdc, and theres a 100uf cap to ground on the input side of it, I dont have a scope available but I dont think the input·is excessively noisy.·
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2007-03-23 00:16
    To meet the LM2937's minimum ESR requirements, you ought to be using a tantalum cap on the output.

    Another question comes to mind, however: Is there anything on your board that uses the 5V output from the LM2940? If so, and depending on how it interfaces to the 3.3V circuitry, you may be getting 5V backfed into the 3.3V supply through an I/O pin.

    -Phil
  • RottenJalapenoRottenJalapeno Posts: 27
    edited 2007-03-23 00:55
    Yes, I do have a few 10k pullup resistors connected to the I/O pins on the prop chip. I guess its just in my design that I'm getting these varying voltages. The board that I am using the regulator on worked fine using the lm1117 without any caps on the output side. But looking at the Data Sheet I guess I still should have had a cap in there. So I realize my mistake and now I'm just trying to find an easy fix, since I have around $400 invested into 20 of these boards. I cannot easily go back to using the lm1117 due to the different pin configuration on the regulator. But maybe someone reading this knows of a little more forgiving regulator that I can use with the standard (in--gnd--out) pin config?
  • TransistorToasterTransistorToaster Posts: 149
    edited 2007-03-23 15:25
    >To meet the LM2937's minimum ESR requirements, you ought to be using a tantalum cap on the output.

    I've been looking at LDO regulators and I am concerned with the criticalness of the ESR in capacitors and risk of oscillation. I saw magnitude and phase response curves supplied with the IC makers and I mostly understand how it is supposed to work when the regulator is unloaded. Do I need to be concerned at all with loads such as clocks pulling current at a periodic rate through the regulator?
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2007-03-23 15:58
    RottenJalapeno,

    To see the effects of using vs. not using a capacitor on the regulator, check out the images in this thread...

    http://forums.parallax.com/showthread.php?p=637871

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • parts-man73parts-man73 Posts: 830
    edited 2007-03-23 17:16
    Here's good reading on a similar subject. About power supplies and their characteristics under load vs open circuit.

    www.glitchbuster.com/wallwart.htm

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Brian Meade

    "They who dream by day are cognizant of many things which escape those who dream only by night" - Edgar Poe
Sign In or Register to comment.