Forum Rule: Always post complete source code & details to reproduce any issue!
Results 1 to 3 of 3

Thread: AREF and the 470ohm resistor. I did a dumb!

  1. #1
    Senior Member
    Join Date
    Feb 2013
    Posts
    179

    AREF and the 470ohm resistor. I did a dumb!

    So, I guess I did a dumb - my custom PCB has some potentiometers (from joysticks) running between GND and AREF, and other parts (motorized faders) running between 0 and 3.3V.

    This seems to be a problem, as the faders will hit a certain point (right at 2.65v) where they 'max out' the ADC and just output 4095 (running at 12 bits). Joysticks work great, but are going between 0 and 2.65v.

    As a possibly temporary solution, I yanked the 470 ohm resistor in the AREF section on the schematic, and replaced it with a solder bridge - and told analogReference to be EXTERNAL. This seems to have the effect of making my reference voltage 3.3v, the joysticks float between 0 and 3.3v, as do the faders.

    So, my question is two fold:

    1) Is this EVEN MORE STUPID than the initial error? Am I going to break a bunch of stuff by feeding 3.3v to the VREFH pin?
    2) Is there some way to 'fix' this problem I've made in software? Ideally, something that I could do just before reading the faders, then turning off just before reading the joysticks.

    Thanks!

  2. #2
    Senior Member PaulStoffregen's Avatar
    Join Date
    Nov 2012
    Posts
    20,758
    1) Is this EVEN MORE STUPID than the initial error? Am I going to break a bunch of stuff by feeding 3.3v to the VREFH pin?
    Feeding the same 3.3V as the VDD & VDDA pins directly to the VREFH is perfectly fine.

    The one things that's not allowed is driving VREFH more than about 0.2V to 0.3V higher than VDD or VDDA. So if the 3.3V for those is separate (like from completely different power sources), you'd need to be careful to make sure they stay about the same, even during powerup and shut down. If they're just physically connected by a wire or inductor or resistor, everything should be fine.

    2) Is there some way to 'fix' this problem I've made in software?
    Probably not. But maybe? Really, to even begin to offer a suggestion on how to fix this without the hardware mod, I'd need to understand why the faders behave so strangely. Obviously they're not the same as the regular pots. But I don't know anything about them, not even a part number.

    The proper hardware connection is probably a lot simpler.

  3. #3
    Senior Member
    Join Date
    Feb 2013
    Posts
    179
    Aha - it seems like it should be fine, then. Essentially, from the Teensy 3.1 schematic on the site, I've 'removed' the 470ohm resistor between VDDA and VREFH (I'm replacing it with a 0 ohm resistor). Since VREFH is now just linked to VDDA, it shouldn't ever go 'above' VDDA (they'll be linked).

    The hardware fader I'm using is a Bourns PSM01-082A ( http://www.mouser.com/ProductDetail/...Ohcgr21wIo31Wl ). It essentially is just a pot, at it's most basic. The difference between the faders and my other pots is that my faders are going between Ground and 3.3V (I have a ground plane and a 3.3V plane in my board - it's a 6 layer monstrosity), while the other pots go between Ground (plane) and AREF (traces) - with that 470ohm resistor, the high range on the pots is 2.65 (2.56?) volts. I see the faders, when they go above that 2.xx voltage, just 'top out' in terms of analog readings (which is about 5/6ths of the way to no-resistance on the fader). Does that make sense? It's hard to explain.

    But, it sounds like dropping that 470 ohm resistor in this configuration won't cause any problems, and everything is working correctly without it - so unless there's something glaring that I'm missing I'll just leave this as though it were by 'design.'

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •