Troubleshooting¶
Causes of an error or a wrong picture, by symptom. The bus is write only, so most hardware faults show as a panel that
is dark, white, tinted or shifted, with every call returning Ok.
Datasheet sections this page relies on
From Solomon Systech (copy from the Adafruit mirror) SSD1331 1.2:
- Continuous scrolling: 9.2.7 to 9.2.9 Scrolling, p. 51
All of them, with the driver API for each: Datasheet map.
EnsureInitialized returns Bus¶
The port reported a failure: its EnsureInitialized() returned false, or a transfer or a pin write did.
- Check that your port's
EnsureInitialized()returnstrueon its own (log it). - Check the pins in your port against the board: SCLK, MOSI, SS, D/C, RES and the supply enable.
- With a logic analyser, check that RES# pulses low for a few microseconds, then SS# frames appear with D/C# low.
EnsureInitialized returns InvalidArgument¶
The Config fails Validate(); nothing was sent and no pin changed. Call isf15acp4::Validate(config) on your
configuration and look at the fields in Configuration.
The panel stays dark, no error¶
Every call succeeds because the bus cannot say otherwise. Work down this list:
| Check | How |
|---|---|
| The 15 V rail | Measure it after DisplayOn(); it must be 14 V to 16 V (ISC15ANP4, ISC01P: 15 V to 17 V) at the module's pin 9 |
| The supply enable | Does the pin go high in DisplayOn(), and is it the pin your regulator listens to? A P-channel switch needs the port to invert it |
| D/C# | A command with D/C# high is taken as data: the controller is never configured. Scope it |
| RES# | Held low keeps the controller in reset; it must be high after EnsureInitialized() |
| SS# and the clock | Mode 0, at most 6.6 MHz, SS# low for each transfer |
| The controller after a dip | A brown-out of VDD resets it silently: call Reinitialize() |
| Contrast | master_current 0 or contrast 0 is nearly black; try the all-on mode: SetDisplayMode(DisplayMode::AllOn) |
Colours are wrong¶
Red and blue swapped: set Config::red_on_segment_a = false, or Remap::reverse_colour_order
(Ambiguities). Washed out or too dark: master_current, the contrast fields, or the
gray scale table (UseLinearGrayScale()). Everything tinted by one colour: one contrast_* value is off.
The picture is shifted, cropped or mirrored¶
Shifted by a few columns: the variant's column offset (PresetFor() for the ISC15ANP4 and ISC01P). Mirrored:
Remap::column_reverse or com_scan_reverse. Upside down: com_scan_reverse and column_reverse together. The
initialisation values are carried from the vendor's notes (Ambiguities).
A line or rectangle is missing or wrong¶
The engine was still busy when the next command arrived: raise Config::engine_settle_us
(Ambiguities).
Calls return ScrollActive¶
A RAM write, drawing call, SetupScroll() or ApplyConfig() while scrolling. Call DeactivateScroll(), then
rewrite the screen (RamNeedsRewrite()).
Calls return Locked¶
SetCommandLock(true) was called. SetCommandLock(false) unlocks; the power-down unlocks by itself.
Poll returns Unsupported¶
The port has no ReadSwitch() or no NowMs() hook, or the module is the display-only ISC01P.
The button reports several presses¶
Contact bounce longer than ButtonConfig::debounce_ms: raise it (scope the switch), and poll faster than half the
debounce time. Presses shorter than the poll period are missed.
The button seems inverted¶
ButtonConfig::active_low is true for a switch to ground with a pull-up (closed reads low). Set it false for a switch
to VDD with a pull-down.
The display wakes with a press but the press does nothing¶
By design: a press that wakes a dimmed or sleeping display has woke_display set. Treat it as a wake, not a command.
The display went white or dark after weeks of running¶
A brown-out or a noise glitch reset the controller; the driver cannot know (Limitations).
Reinitialize() repairs it. Check the VDD supply and its decoupling.