GPIO range doesn't go below 21

#1

I'm having a weird issue with an Axoloti Core that I don't believe I've used before (I have two designated for future projects and I think this is the first time I've used this one.)

On all GPIO, the range doesn't go below 21 and doesn't seem to go higher than 62. At first I thought it was the input circuit (a phototransistor), but then I discovered that if I unplugged everything from the board, the value on the floating pin didn't even skitter down below 21.

Any thoughts? Is there some sort of configuration that I've triggered or something? I'm worried that I may have somehow damaged it, but everything else seems to be working fine, and I haven't experienced anything like this on previous Axoloti projects.

#2

What if you take a GPIO and short circuit it to ground, i.e. 0V on the board. Does it still read 21 ?

#3

Can’t believe I didn’t think of that. I’ll give it a shot!

#4

Thanks for the suggestion! Even though I haven't needed one elsewhere, it looks like this one needs a pulldown resistor! Minor bummer, but not the end of the world!

#5

It's a bit hard to know without seeing the complete phototransistor circuit, but if certainly if you leave the input completely unconnected it will be sensitive to all sorts of electrical noise in the vicinity of the input circuit.

I think it's possible to enable a pullup or pulldown per pin in the STM32 chip, but I'm not sure if it has been done for the pins configured as analog inputs. Probably not if using an external pulldown actually pulls the level down to zero. It could also be that the internal pulldown is either too weak or configured to be too high an impedence to be sufficient.

At any rate, if an external pulldown fixes the problem I'd say go with that.

#6

I just haven't had to do this with other Cores, which have been maybe a little jittery without an external pulldown, but certainly not giving a consistent offset like this. And, weirdly, it's pegged at so exactly 1/3 of the total it seems like it's a resistor with the wrong value.

#7

I'm really just speculating here, but I'm wondering if the analog input pin happens to be running close to some line on the board with a duty cycle of 1/3 which when capacitively coupled would lead to an input value of 1/3 of VCC.

The behavior of the STM32 chip for the various pins is set in the firmware/board.h file. One would have to check the STM32F427 data sheet to see exactly what value the built in pull up and pull down resistors have.

#8

Hi...I have a situation where I need about 100-200 Pi's in a server room. Ideally, they do not want 5V USB chargers all over the place. There is the possibility that they can supply me with a custom connector to a 5V power rail that can do up to 3A per connection.
However, that rail runs at 5.2V not exactly 5V. Would this damage the Pi on GPIO?
Are there any other options since 2012 where mosts of the posts come from for powering a Pi without relying on Micro USB?

pcb assembly

#9

That looks like a Raspberry Pi question. I am trying to see the relevance here.