Shop OBEX P1 Docs P2 Docs Learn Events
New design has bad voltage. — Parallax Forums

New design has bad voltage.

PaulPaul Posts: 263
edited 2010-10-12 13:03 in Propeller 1
I just built a new Prop design using a LQFP package and the 3.3v LM2937 regulator is putting out 4.6v. (+5V in) This is probably my fifth LQFP design and I've never had problems with them before.
So far I have replaced the regulator, the capacitors on both the regulator's input and output, and the Propeller chip. Without the Propeller I have VDD = 3.3v.
Watching on a scope I found I get 3.3v until the EEPROM is read then VDD jumps up to 4.6v. Also I never see anything on the Xtal leads. I haven't been able to program it so all I/O should be floating.
Unless someone has ran into this before I'm at the toss it and start over stage I think. Any ideas?

Paul

Comments

  • RaymanRayman Posts: 14,889
    edited 2010-10-12 08:14
    I vaguelly recall seeing similar behavior when something was wrong with the circuit...

    My guess would be that you're trying to draw too much current through the regulator... Perhaps you have something shorted to ground somewhere...
  • RobotWorkshopRobotWorkshop Posts: 2,307
    edited 2010-10-12 08:21
    You mentioned this is a new design. I would start double checking the layout, etc to make sure there aren't any mistakes or un-intentional lines shorted together. Also, check to see if there were any problems cause when the PCB was made. Several years ago I had a problem with a local board house when they changed the process used to tin the boards. As a result I endedup with some bad boards with solder bridges on the PCB itself. They replaced them all and I haven't had problems yet.

    Robert
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2010-10-12 08:26
    With so so little information to go on I will just guess.

    GUESS #1: You have 5V levels running into an input somewhere. This finds it's way through the substrate and onto VDD pulling it up and the regulator is not designed to pull down a voltage so it sits there at 4.6V.

    If you have 5V inputs then you must not only current limit them but you must make sure they do not affect VDD. You could try a zener but at those low voltages they aren't much chop. I use 2 red leds in series with a 47R across VDD and GND to help shunt those stray currents. When the Propeller is drawing more power then that excess current is soaked up (technical term) but when it sits back and idles you see the voltage shoot up.

    Quote "any ideas?"
    SUGGESTION: Show us your circuit
  • LeonLeon Posts: 7,620
    edited 2010-10-12 09:16
    Are you using a low ESR 10 uF capacitor, and is it close to the regulator leads? It might be oscillating when it has a load on it.
  • CannibalRoboticsCannibalRobotics Posts: 535
    edited 2010-10-12 09:24
    I've seen this a few times and traceable to two causes.
    1) On a new design the first thing we do is install the regulator(s) and check for the right voltages at the pads. If the regulator is not 'loaded', it will float up to the supply level. Where are you checking our voltage and have you done continuity checks to the pads on the prop?
    2) The regulator has a different pinout than assumed in the design they will behave like this.
  • PaulPaul Posts: 263
    edited 2010-10-12 09:50
    Thanks for all the suggestions. I have 10uf ceramic monolithic caps about .1" from the regulator. I have tried Electrolytic and Tantalums with the same results. I don't see any "ringing" on the scope. I'll check again for the +5V feeding into the 3.3V but I haven't found it yet. The regulator is a standard LM2937-3.3 as I said before. (Pinout: IN-GND-OUT)

    Looks like I have a short under some component somewhere. Time to drag out the SMD rework station. Now where'd I put that manual?

    Thanks forum, I appreciate the help.

    Paul
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2010-10-12 10:22
    The LM2937 requires an output cap within certain ESR limits to guarantee stability. The ESR of most ceramic caps is too low. You said you tried a tantalum instead with the same results, so the cap may not be contributing to your current problem. Nonetheless, if you end up modifying your layout, I would still switch to tantalum, just to eliminate a potential source of trouble in the future. Either that, or switch to a regulator that's designed to work with ceramic output caps.

    -Phil
  • PaulPaul Posts: 263
    edited 2010-10-12 12:57
    Thanks Phil. The ceramic caps were just a shot in the dark anyway.

    The problem was (drum roll) I designed for a MAX232 and put in a MAX3232 instead. The 5V must have been feeding into the Propeller pins and screwing things up. I should have known better.

    Paul
  • RobotWorkshopRobotWorkshop Posts: 2,307
    edited 2010-10-12 13:03
    Hello Paul,

    I'm glad you found the problem and thanks for posting the resolution. I've been using the MAX3232 with all my Propeller projects since it runs off the 3.3V supply. Just makes things easier to deal with.

    Robert
Sign In or Register to comment.