isConnected test fails on unbranded micro OLED display

[PaulZC]

Example6_DrawIcon uses the standard isConnected test to discover a display with address 0x3D or 0x3C.
The unbranded OLED display I have fails this test:

If the SparkFun displays pass this test, then please close with extreme prejudice!
Paul

[nseidle]

Qwiic OLED happily acks at 0x3D (default address).

Don't mind the 0x42 u-blox GNSS and 0x36 MAX17048 fuel gauge, I tested on the ESP32 RTK Surveyor that has these built in.

What address does your unbranded display ack to?

[PaulZC]

It's really very strange... It defaults to address 0x3C and ACKs OK when I write data to it:

But doesn't respond to isConnected either way round:

I suggest we don't lose any sleep over this as it is an unbranded display, but maybe we should leave this issue open (and make it sticky?) just in case any users have the same issue?

[lewispg228]

Hey Paul and Nathan,
Not sure if this is helpful, but this reminds me of an issue I ran into with the cryptographic IC I worked on a while ago. It required a "wake" command (which I think is actually kind of an old standard to wake up an entire I2C bus?). It basically just means holding the SDA line low for a specific amount of time. I ended up having to do something different than our usual isConnected(). Maybe this unbranded OLED would like to see a wake command first?

Here is where I implemented it in the ATECCX08a library.

From the ATECC508A datasheet:

"
The Wake condition requires that either the system processor manually drives the SDA pin low for tWLO,
or a data byte of 0x00 be transmitted at a clock rate sufficiently slow so that SDA is low for a minimum
period of tWLO

"

For the ATECC508a the required time was 60uS, so writing 0x00 at 100kHz was sufficient.

Hope this is helpful. Might be worth a shot?

[PaulZC]

Thanks Pete - I'll give it a try!
All the best,
Paul

[PaulZC]

So, pulling SDA low didn't seem to do the trick. But that got me thinking... And if I give the display a manual reset before trying isConnected then it does work! Cheers Pete - you're my inspiration! ;-D

@nseidle: should we update the example? I'm not seeing any downside?

[nseidle]

Sounds good to me!

[PaulZC]

Done. The new version is in the release_candidate branch for now...
And that makes it beer-o'-clock!
Closing...