LoPy pinout (work in progress)

[Xykon]

@LoneTech said in LoPy pinout (work in progress):

Here's a little surprise: GPIO 34 and up don't have output capability, according to the technical reference manual. That's P13 to P18 on the LoPy. Initial experiments seem to support this; I'm not getting them to output anything. This is because they have dedicated analog input functions, apparently. Perhaps this is why they are blank in your table?

Sorry took me a while to see your latest update. I will update my image with your findings tomorrow.

I first started using the blink example from esp-idf to find the JTAG pins using a logic analyser to see which pin would change states. I then used openocd-esp32 to write to GPIO_OUT_REG and GPIO_OUT1_REG to try and identify the rest of the pins.

Since P13-18 are input pins (a fact that I missed in the reference manual) it makes sense that I couldn't identify them.

[mfgeorge]

Hello all,

I have found the i2c pins for bus 0 (currently the only one that does not throw an OSError during instantiation).

P9 - SDA Bus 0
P10- SCL Bus 0

Again this is presumably for i2c bus 0. ie i2c_object = I2C(0, I2C.MASTER)

Method for finding these pins:

These pins were found by looping a "while True: i2c.scan()" in the REPL prompt and scoping around for a clock signal. After the clock signal was found I continued the i2c.scan() and plugged in an i2c sensors SCL to the P10, and checked to either side with the sensors SDA pin and found the address for the sensor was reported back while SDA was connected on P9.

Hope this helps and happy hacking everyone!

[Xykon]

Thanks... it looks like you've done the same thing I did tonight. Currently doing some tests with a simple 24LC64 eeprom.

[mfgeorge]

No Problem...

Another live update. For UART Bus 1 (ie. uart = UART(1,baudrate=9600)) the pins are:

P3 - UART TX Bus 1
P4 - UART RX Bus 1

[Xykon]

@mfgeorge Updated image with additional information provided.

[Xykon]

It's late and I might be wrong but I think UART1 should be P3 RX1 and P4 TX1.

[mfgeorge]

@Xykon Hmmm.... I can double check what I did tomorrow. But also remember that Tx on peripheral device goes to Rx on master, and vice-versa for the other combo. They need to criss-cross so that one is transmitting to another's receiving pin. A little confusing, I know.

[Xykon]

I connected an FTDI breakout board to UART1 and wired TX on the FTDI to P3 and RX to P4. I disconnected it for the night will check again tomorrow.

[livius]

i suppose that this should be compatibile with wipy yes or no?

[littlebat]

@livius said in LoPy pinout (work in progress):

i suppose that this should be compatibile with wipy yes or no

That is for wipy 1.0 not wipy 2.0

[Xykon]

@littlebat said in LoPy pinout (work in progress):

i suppose that this should be compatibile with wipy yes or no

No as they use different chips (cc3200 in the WiPy1 and ESP32 in the LoPy / WiPy2).

For example WiPy1 doesn't have JTAG on the expansion header while the LoPy does.

[Xykon]

I've uploaded the Excel file I used to make the image in case anyone wants to make modifications.

[LoneTech]

You may want to take a look at the ESP32 technical reference manual. Section 4.10 IO_mux and 4.11 RTC_mux list the dedicated alternate functions for each pin, but more importantly, section 4.9 GPIO matrix lists all the functions that can be mapped anywhere, which includes those UART and I2C functions (there are over a hundred each direction). In short, unless the software does something silly, there's no particular pin for RX1/TX1 et cetera (micropython should use the GPIO matrix). UART 0 only has them because the bootloader ROM selects those pins, though I'm not certain if it does it via the IO mux or GPIO matrix.

[Xykon]

Yes I'm aware of that... I was just putting in the defaults for SPI0, I2C0 and UART0/1 as the examples in the quick reference do not specify any custom pins.

[LoneTech]

I see. It's the default arguments of the micropython port we're looking for then. The chosen pins don't match the IO mux. A comment in https://docs.pycom.io/lopy/lopy/quickref.html#pins-and-gpio seems to be the documentation we get.

UART 1 pins are most likely chosen to match the WiPy 1.0 (CC3200) layout documented in https://docs.pycom.io/wipy/wipy/quickref.html but the I2C looks different. Hopefully we can clean the documentation up once the project opens up a bit.

[mfgeorge]

@Xykon OK, so I again verified my results.
Connecting a slaves RX pin to P3 and the slaves TX pin to P4 yielded results. I tried it the other way, and nothing.

So again, I stand behind what I stated before.

Slaves RX ---> LoPy Tx - P3
Slaves Tx ---> LoPy Rx - P4

The criss-cross is confusing I know, but this is the standard way of wiring UART and naming the pins. The slave I am using is well labeled, and I have tested the same exact wiring scheme named above using other microcontrollers. I am now pretty positive of those two pins.

[Xykon]

@mfgeorge It looks like the DFRobot FTDI breakout board I was using to test UART1 has RXI and TXD inverted. I compared it to the Sparkfun FTDI breakout I was using for UART0 and it has the RXI and TXD the other way around.

Lesson learned: Never trust the silkscreen on cheap breakout boards. I will update the graphic tomorrow.

[pighixxx]

Hi all,

Today i released the official LoPy pinout:

Download the PDF here

Have a noce day!

[LoneTech]

@pighixxx Thanks! I'd like some small additions if possible.

1. Which level to use to select internal antenna
2. Bootstrapping pins:
   2a. P2 for the low level bootloader, must not be grounded for normal bootup
   2b. P12 for the multiboot / safemode, which should not be kept high for normal MicroPython use

I haven't dug around for the capacitive touch features, but I'd have expected that SensVN and SensVP formed one pair, and SensCN and SensCP would form another. Still much to learn.

[pighixxx]

Hi,

Today i release the WiPy and Pymakr pinouts then I make the changes to WyPy pinout in the next revision.