Troubleshooting

Symptoms you may see while bringing up a MAX22200, their causes and the fix for each.

Start with the cheapest checks: the return value of the failing call, GetLastFaultByte() and a raw STATUS read. They tell bus problems (no reply, COMER) apart from chip problems (the chip answers but reports a fault). The max22200_diagnostic example prints all three, plus a CFG_CH0 round trip, with every SPI byte.

Start-up

Symptom	Cause	Fix
Initialize() returns INITIALIZATION_ERROR	Your port's Initialize() or Configure() returned false	Check the port: SPI host, pins, mode 0 accepted
Initialize() returns COMMUNICATION_ERROR	Transfer() failed, or the chip answered COMER (0x04) three times	CMD wiring and level; Transfer() blocking until CS is high; one call, one CS frame
Every read is 0xFF or 0x00	SDO not pulled up, ENABLE low, VM off or below 4.5 V	Pull-up on SDO, ENABLE trace, VM at the chip pins
Initialize() is OK but STATUS reads ACTIVE = 0 or UVM = 1	VM sagging below UVLO, or V18 not settled	Check VM under load, the 2.2 µF on V18, the charge-pump parts; write ACTIVE again after the supply is stable
STATUS has impossible values (reserved CMxy = 11, ONCH set at boot)	The chip kept its registers through an MCU reset	Registers reset only when V18 falls below about 1 V: power-cycle VM

Channels

Symptom	Cause	Fix
Channel is on in ONCH but the output does not switch	ACTIVE = 0, or less than 2.5 ms (tWU) since you set ACTIVE = 1 with WriteStatus()	Check StatusConfig::active; wait tWU after a raw ACTIVE write (Initialize() already waits); check your port's DelayUs()
Low-side channel does nothing, high side works (or the reverse)	The load is wired for the other side	Low side: load to VM. High side: load to GND, VDR only
ConfigureChannel() returns INVALID_PARAMETER	IFS is 0, SRC at 50 kHz or more, or channel above 7	SetBoardConfig(), a lower chop_freq, the channel number
A current or duty setter returns INVALID_PARAMETER	The channel is in the other drive mode: current setters need CDR, duty setters VDR	Configure the mode first with ConfigureChannelCdr() / ConfigureChannelVdr()
Drive mode or side does not change	A channel is on, or TRIGA / TRIGB is high	Switch every channel off and hold both TRIG pins low first
Current is off by a constant factor	IFS does not match RREF, or BoardConfig(r, true) and half_full_scale both set	BoardConfig(rref_kohm, false) with the real RREF
VDR duty is higher than requested	Driver 1.0.0, which scaled VDR percent by 127 / 100	Update the driver; see Drive modes
HIT time reads back shorter or longer	HIT_T is an 8-bit count of 40 chopping periods	Expected; compare raw register values
Another channel turns off when you enable one	STATUS was written outside the driver; the ONCH cache is stale	ReadStatus() after any raw STATUS write

Faults

Symptom	Cause	Fix
FAULT low right after power-up	UVM is set at every power-up	Initialize() reads STATUS, which clears it
OCP on a channel	Output above 1.1 A: a short, or a coil with too low resistance in VDR	Clear ONCH, read FAULT to clear the flag, fix the load
HHF on a channel	HIT current not reached within the HIT time	VM too low for the coil resistance, HIT too high, HIT time too short
OLF on a channel	Load disconnected, with open_load_detection_enabled set	Wiring; the flag clears when the output turns on
DPM on every cycle	No dip seen: the plunger is stuck, or the thresholds are too strict	Hold the plunger to compare; tune with c21_dpm_tuning_test
DPM never fires, even with the plunger held	Current slope above 700 mA/ms above the start current, or the start current above the dip	Set the start current just below the expected dip; check the slope on a scope
ClearChannelFaults() clears other channels too	The part is a MAX22200, not a MAX22200A	Expected: the MAX22200 clears every flag on any FAULT read

A DPM flag means no dip was revealed, so a healthy valve produces no DPM flags. To check that detection works, hold the plunger still during one cycle: the flag should set for that cycle.