Skip to content

Changelog

What changed in each release of the driver. Entries marked BREAKING need source changes in code that used the affected API.

Unreleased

Documentation

  • The documentation site moved to the HardFOC MkDocs platform: new page set (quickstart, hardware, drive modes, SPI protocol, architecture, API guide, configuration, porting, examples, troubleshooting), the Doxygen API reference under /api/, and every code sample compiled by the host build from tests/docs/.
  • Corrected the hardware facts: the chip has no logic supply pin (VM 4.5 V to 36 V only), the outputs are eight single half-bridges OUT0 to OUT7, SPI is mode 0 only, and there is no RESET pin; ENABLE, CMD and FAULT are the control pins.

Breaking changes

  • BREAKING (behaviour, no signature change): SetHitCurrentMa(), SetHoldCurrentMa(), their ...A() and ...CurrentPercent() variants, SetHitDutyPercent() and SetHoldDutyPercent() no longer change the channel's drive mode. A current setter on a VDR channel, or a duty setter on a CDR channel, now returns INVALID_PARAMETER and writes nothing. 1.0.0 silently flipped VDRnCDR and reinterpreted the other setpoint in the new unit (and the chip ignores a mode change while any channel is on or a TRIG pin is high). Select the mode with ConfigureChannel(), ConfigureChannelCdr() or ConfigureChannelVdr() before using the setters.
  • BREAKING (behaviour, no signature change): ConfigureDpm(start_ma, threshold_ma, debounce_ms) converts the debounce with the chopping frequency of the channels that have DPM enabled instead of always FMAIN_DIV4, so the same debounce_ms gives a different DPM_TDEB code on channels not at FMAIN_DIV4. It reads CFG_CH0..7 to find that frequency, and returns INVALID_PARAMETER (writing nothing) when DPM channels use different frequencies or an argument is negative or NaN. Enable DPM on the channels before calling it, or use the new overload that takes the frequency.
  • BREAKING (behaviour, no signature change): the callback registered with SetFaultCallback() is now called by the driver. Code that called it itself after ReadFaultRegister() or ReadStatus() now sees each fault twice; remove those calls.

Fixed

  • VDR duty cycle is written as percent, as datasheet Table 3 defines HIT[6:0] / HOLD[6:0] in VDR: 40 % is now HIT[6:0] = 40 (1.0.0 wrote round(40 × 127 / 100) = 51, so the output ran at 51 %). Reading back returns the field as percent (100 and above read as 100 %). If you compensated with the 100 / 127 workaround, remove it.
  • Initialize() waits the 2.5 ms wake-up time tWU after writing ACTIVE = 1 (datasheet Electrical Characteristics, "wake-up completed?" in Figure 6), so a channel switched on right after it drives its output. Initialize() now takes at least 3 ms.
  • ConfigureDpm() converted the DPM debounce with the FMAIN_DIV4 chopping frequency whatever the channels used; DPM_TDEB counts periods of the monitored channel's fCHOP (datasheet CFG_DPM, TDEB = DPM_TDEB / fCHOP). It now uses the chop_freq of the channels with DPM enabled (exact fCHOP, e.g. 33.33 kHz), returns INVALID_PARAMETER if they differ, and rejects negative or NaN arguments.

  • The fault callback set with SetFaultCallback() was stored but never called. The driver now calls it for each OCP / HHF / OLF / DPM bit it reads from FAULT, and for OVT / UVM / COMER read from STATUS after Initialize().

  • cmake --install left out src/max22200.ipp, which max22200.hpp includes, so an installed driver did not compile. The .ipp is now installed next to the headers (the header includes it from there when ../src/ is absent) and the package exports the target as hf::max22200, the name add_subdirectory() users link (1.0.0 exported hf::hf_max22200, which could not be used).

  • The ESP32 example bus started TRIGA and TRIGB high. They are active-high inputs with internal pull-downs (datasheet Pin Description), and the chip ignores VDRnCDR / HSnLS changes while either is high (Programming Procedure step 5c), so the examples' drive mode and side settings could be ignored. They now start low, and the trigger-pin test pulses low -> high -> low. SetTrigA() / SetTrigB() documentation now says true = high = trigger asserted.

Added

  • ConfigureDpm(start_ma, threshold_ma, debounce_ms, chop_freq) overload that takes the chopping frequency.
  • Host unit tests (HF_MAX22200_BUILD_TESTS) that run the driver against a model of the MAX22200 SPI protocol.

1.0.0

  • First release: MAX22200<SpiType> with CDR and VDR channel configuration in real units, channel switching, full-bridge states, STATUS and FAULT access, selective fault clear, DPM configuration and a CRTP SPI interface.