Problem with OLED featherwing display 64x128

EL Wire/Tape/Panels, LEDs, pixels and strips, LCDs and TFTs, etc products from Adafruit

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
csoft
 
Posts: 3
Joined: Tue May 23, 2023 4:14 pm

Problem with OLED featherwing display 64x128

Post by csoft »

I wanted to use the OLED featherwing display for a university course with practical exercises.

I have to work with a ESP32 DevKit-C and micropython.

I got things working nicely. The configuration of the board I programmed by studying the sh1107 data sheet and looking at source code of Adafruit libraries (to verify that I do the same thing as them).

BUT!!!

After power up I need to go through the initialisation procedure between 5 and 9 times to get the display switched on. Afterwards (If I do not interrupt the power) everything works like charm and I can change and reload my program and execute it as expected.
I can also load my program as main.py in the ESP32 and execute it at power up, however, again I need to operate the devboard reset button around 8 times after the power up to get the display working. Then I can reset normally.

Anybody had a similar issue? Any idea what I could try?
(I tried the following: reduced i2c speed, inserted delays after each configuration command, double checked quality of power connection, waited some time after power up before executing the program)

(A different SH1106 based OLED display I got working without any issue, but I like the Adafruit display due to the additional buttons and would like build my course based on this if I can get rid of this problem...)

Any help appreciated! Thank you !

( I attach my test program which reads out a BME280 and displays the sensor values. Sorry, it is a horrible hack... )
Attachments
displaySensor.py
(9.85 KiB) Downloaded 1 time

User avatar
csoft
 
Posts: 3
Joined: Tue May 23, 2023 4:14 pm

Re: Problem with OLED featherwing display 64x128

Post by csoft »

I answer to my post: problem solved:

I figured out that the display seems to have a very long power up phase. If I connect the reset line of the display with the reset line of the dev board and power up the setup and wait patiently for !!! 30sec !!! things work nicely. A bit surprising this long power up phase... is this expected?

Anyway I can live with it, but if this is the normal behaviour (I almost doubt it) it should be written somewhere (apologies if it IS written and I missed it).

User avatar
csoft
 
Posts: 3
Joined: Tue May 23, 2023 4:14 pm

Re: Problem with OLED featherwing display 64x128

Post by csoft »

To leave something useful here... I have a better understanding now on what happens:

In short : not the fault of the display at all, but it needs a solid supply voltage.

The whole story:

The display has a small chip which creates a reset signal if the supply voltage drops below a certain value.
The supply voltage (3V3) comes from the esp32 Devkit-C. I now observe that when I connect the display to this supply voltage there are some spikes to low voltages every ~4 seconds (the first lasts around 150us and the subsequent spikes get shorter and shorter). These spikes on the VCC line go low enough to trigger the reset chip which then correctly creates its reset pulse. After 8 pulses the spikes have become much shorter and the voltage drops less during the spike so that the reset chip is not triggered any more and then the startup succeeds: this happens after ~8x4 = 32 sec. Since I have the reset connected to the reset of the microcontroller the program always restarts after a reset, but within 4 seconds it does not reach the point where the display should show something (the boot process of the Devboard is long). Hence the setup only starts to work after ~30 seconds.

If I do not attach the OLED display to the setup (but only a sensor) I do not see these spikes on the power line.

May be the display creates some power surge when the DC/DC converter is ramped up or when some particular configuration command is issued? Something then slowly charges up and after 8 trials the power surges are small enough to not disturb the (marginal) power supply of the development board. (Of course this is just a speculation).

So the OLED display is perfectly ok, but needs a solid power supply (which the feather proabably has... May be this is the reason why Adafruit does not recommend to use this board with other microcontrollers??? ...I do not know...)

Hope this helps to save some time for other people...

Thanks to Adafruit to make all details like circuit diagrams, data sheets, etc etc public. This helps to get things going !!!

Locked
Please be positive and constructive with your questions and comments.

Return to “Glowy things (LCD, LED, TFT, EL) purchased at Adafruit”