@Eric-Tremblay Ah sorry I had missed that. No, the message is correct. The address is on 16 bits, so takes two bytes to send. What happens when you try a read?

It looks like you are using signed intergers/numbers for the reading from the light sensor. When it reaches the half way value it wraps around and goes negative. If you simply define the data as an unsigned type does it fix the issue.

I think the issue was with my breadboard psu... So right now I am using the pycom 3.3V

@kjm All versions have a usable I2C. The case here is a different one, I assume. @shanefowler25 : Can you publish the full error message you got. It should be more than one line. The framebuffer module is included now. It sneaked in when Pycom moved to MicroPython core V1.11.

@Lucas_Vieira I'm missing the print statements form the script. So your test script is not executed at all.

@weingami Note that I2C level shifting is particularly easy as it uses an open drain-pullup signaling scheme. Shifting can simply be done with two FETs: https://electronics.stackexchange.com/questions/255607/i2c-level-shifting-mosfet

@dnear1 So what was the issue then with the scope diagram below: https://forum.pycom.io/topic/4263/optimal-i2c-pullup-resistors/5 That one does not look well.

@brossingo Do you use a pycom board? Then add some wait. You have to wait until the *Py and the board is initialized. Sadly there is no flag or event which tells you that init is ready. So you have to find the right time for your system by your own. It will depend on firmware version and the hardware you added - especially your display. Give 1.20 rc4 a try. This version solved my problems with i2c and pytrack

@robert-hh Works like a charm, thank you so much!

@robert-hh OK, I set up some code to try this out, several timers at different periods and several threads including one to start LTE, send an SMS and then deinit. From what I can see, the timers do NOT get interupted by the LTE device. It seems the timer counts are retained and dumped to REPL as soon as possible (buffered apparently). There is some corruption/mixup of REPL lines but all the timer flags are there I will investigate a bit further and see if I can use this to my advantage. One thread/timer I will need will be at 11uS preferably, so as it is too fast to capture by eye, I'll have to write to a file to a file to evaluate any impact by the LTE device. Biggest constraint might be the actual time to write to disk. A FIFO data buffer is probably a good solution - some work to do there. The test code was pretty rough and ready so its needs a lot of refinement to be truly useful as a valid test.

@robert-hh Thanks, I'll try that.

@robert-hh I looked at that device but I'm trying to keep my system as simple as possible so I did not pursue that option. I does, however, look like a viable alternative in the event your s/w solution was inadequate for some reason. thanks.

@sealyons I made a test with the calibrated temperature mesurement device: Calibrated device: 21.7 21.8 DS18B20: 21.9 21.9 DHT22: 21.8 21.9 SI7021: 21.7 21.8 BMP085: 22.4 22.5 BME280: 21.8 21.9 Besides the BMP085, all devices were on the spot = within 1 digit variance. That status was stable for 5 hours testing. Only the BMP085 was always ~0.6 °C above, but still that at the limits of it's spec.

@Eric73 Thank you, it works much better now!

@eric-waai Yes it was a typo I have 3 i2c devices

@robert-hh Awsome! Thank you

@robert-hh Thanks for the tips. I ended up doing below, which does save 32bytes over struct.pack. self.tempBuff = bytearray(3) def _writeWord(self,address,data): #self.i2c.writeto_mem(self.address, address, struct.pack(">H", data), addrsize=8 ) self.tempBuff[0] = address self.tempBuff[1] = data >> 8 self.tempBuff[2] = data & 0xff self.i2c.writeto(self.address, self.tempBuff)

@dr_aloui In that case it's strange. You you have an oscilloscope to look at the signal timing? The expansion board should not add any significant load to the wires, at least not to P9 and P10. The Pyboard has built-in pullup-resistors on the pins dedicated for I2C (X9/X10 and Y9/Y10). That could makes a difference.