I’ve been messing with oleds lately and I’ve found I’m getting a higher failure rate than I would expect. So I did some digging, and if what I’ve found is correct, then it suggests that the prevailing advice to power oleds from the 3.3 volt pad could be wrong.
From my own tests I’ve found that powering from 3.3 volts, an oled might work for a few minutes, but then simply stop. Sometimes after powering down for 10 minutes and trying again it would work again.
However if I power the same oled from batt+, it’s flawless and completely reliable.
Looking at the back of the oled I can see it has a 662k chip. So I looked up the specs and what it does. Here they are:
- Output Voltage: 3.3V (Fixed)
- Input Voltage Range: 1.8V to 6.0V (Max)
- Maximum Output Current: Typically 200mA to 300mA
- Dropout Voltage (V_{do}): ~ 250mV (at 100mA load)
- Quiescent Current (I_{q}): Very low, typically ~ 3μA to 25μA
- Package: SOT-23 (3-pin surface-mount)
- Operating Temperature: -40°C to +85°C [[1]
So that chip is a voltage regulator designed to output a solid 3.3 volts to the screen. But crucually it looks as though that chip has its own voltage drop of around 250 mV. That means if we connect it to a Proffie 3.3 volt rail, which according to my meter outputs a rock steady 3.28 volts, then the oled screen itself only gets just over 3 volts by the time it’s been re-regulated by the 662K chip. Since the screen wants 3.3 volts, it’s therefore right on the edge of what it can function with, which is why we see weird behaviour or erratic failures.
Using Batt+ solves all of this, and since the oled’s 662k chip can cope with up to 6 volts, there’s no risk of it getting too much power.
This all seems rational to me, but the fact that for so long we’ve old everyone to use 3.3 volts for oleds makes me think there must be some detail I’m missing.
Would be interested to hear people’s thoughts on it.
