Need help running i2c on attiny85t

[ZiZizou]

I am trying to make i2c with Wire.h using the ATTinyCore work on the ATTiny85v 8 pin dip chip. AFAIK, the difference of the "v" means that it can be operated on lower voltages than the traditional model.

I am using the example slave_receiver for testing purposes to upload to the ATTiny85v chip. I then use the i2c_scanner sketch on the Arduino to verify if the i2c operation is successful, but I haven't gotten any luck so far.

The chip is being run at 8MHz and I make sure to burn bootloader prior to uploading via Arduino as ISP. The board option I select is "ATTiny25/45/85(no bootloader)" - to program the same.

I make sure to use 10KOhm resistors as pullups on the pins PB0 and PB2 of the ATTiny85 and connect SDA on the Arduino UNO to PB0 and SCL on Arduino UNO to PB2. I am also using a 100nF cap b/w vcc and gnd of the chip.

I am at a loss at how to troubleshoot further. I know the board is being programmed because a blink sketch that I uploaded worked fine. Kindly let me know if any further details are required and I will provide them ASAP. Are there any alternatives for chips (preferably DIP) that are as space conscious as the ATTiny85 (in case the i2c doesn't work out on this)?

Thank you for your thoughts, guidance and criticisms.

[hmeijdam]

If testing on a breadboard I would try swapping out wires and confirm with some I2C slave device that the I2C scanner is performing ok. I have been bitten by bad connections on my breadboards more than once.

[SpenceKonde]

Why are we talking about the bootloader when you say that the sketch uploads successfully? And why is the uploading related to I2C? I am aware of no I2C bootloaders, amd based on the specs and foibles of I2C this should surprise nobody.
.
If the sketch uploads successfully with verify on there is no problem with uploading (on my other two cores, I've rigged up the platform.txt to force upload output to be verbose, and to always verify the contents of the flash no matter what the user asks for. There is no CRC or checksum or any other sort of error checking, and allowing those to be turned off is doing them a disservice, as is hiding the information only shown on verbose upload. ATTinyCore will be getting that in the 2.0.0 release

The problem Is I2C. Do you have an oscilloscope? If so, monitor the two I2C lines during an write or read to/from. From a photo of that, I could spot several hardware/wiring issues.

Speaking of wiring- you say space constraints, but yo also mention an arduino Uno. If you're using that gigantic uno, the only way thee could be a space constraint is if you are using the wrong main board ( a gigantic Uno instead of the identical processor on a much smaller and less expensive nano or pro mini (the clones are around $3) - OR you have the t85 located somewhere relatively far from the device - on the end of long wires. I2C was designed for communication between multiple IC's on the same circuit board. It works with short wires too. But not long ones. Weak pullups will also not help if there are long wires. If the wires are long, I would put the strongest pullups I could which would not force the current needed to drive them to approx 0V did not exceed 3mA - on both standard and fast mode the current is throttled to that.

In either event, please let me know whether you are using the 1.5.2 release of thej 2.0.0-dev github latest version.
Whichever one you are using, download the other one and try to use that version in it's entirety (and whether or not that works, so I know how to fix the bug (it will never be fixed without this information), please then try taking the Wire library from the version that works and replacing the library version for the one that does not work (this is more likely to work taking the 1.5.2 library to 2.,0., as the newer library may require the newer core

[hmeijdam]

Works fine for me on IDE 1.8.13 with core 1.5.2

20:01:48.367 -> 
20:01:48.367 -> I2C Scanner
20:01:48.367 -> Scanning...
20:01:48.367 -> I2C device found at address 0x08  !
20:01:48.414 -> done
20:01:48.414 -> 

Also on IDE 1.8.13 with Github version 2.0.0 of Attinycore

I2C Scanner
Scanning...
I2C device found at address 0x08  !
done

[SpenceKonde]

The github versions if sync'ed inthe last couple of days have significantl fixes particularly for the x5-series. There were a LOT of issues with it (afaict, it had never been worked over to the ATTC 2.0.0 standard)