Wiring + voltage

The CH341A black module connects to a chip in one of three ways:

1. ZIF socket — insert a loose 8-pin chip directly. Most reliable; no clip-orientation guessing.
2. SOIC-8 clip via ribbon cable + breakout — clip onto a chip that’s still soldered to a board, ribbon cable carries the signals to a small adapter PCB that plugs into the ZIF socket.
3. Direct flying-wire breadboard — wire each CH341A pin header to the chip with individual jumpers. Slow; useful when nothing else fits.

SOIC-8 pinout

Looking down at the chip with the dot/notch (pin 1 marker) at the top-left:

Pin	SPI (25xxx)	I²C (24Cxx)
1	CS#	A0
2	DO / MISO	A1
3	WP# (write protect)	A2
4	GND	GND
5	DI / MOSI	SDA
6	CLK	SCL
7	HOLD# / RESET#	WP# (write protect)
8	VCC	VCC

The red wire of the SOIC clip marks pin 1. About half of first-try failures are clip-reversed — always double-check orientation against the package marker.

CH341A pinout

The same CH341A board serves both SPI and I²C, but the signals land on different pins per mode:

Pin	SPI	I²C
D0	SCS (chip select)	SCL
D1	—	SDA
D3	SCK	—
D5	MOSI	—
D7	MISO	—

The ZIF socket on the black module is internally wired to route the right CH341A pins to the right socket positions for each chip family — that’s why a 24XX chip placed in the 24XX position works even though SPI and I²C use different CH341 pins.

When using a SOIC clip + breakout, the breakout is the bit that maps clip wires to ZIF positions. Make sure it’s plugged into the ZIF half that matches your chip family:

• SPI flash (25xxx) → breakout in the 25XX position
• I²C EEPROM (24Cxx) → breakout in the 24XX position

If you’re moving between an SPI chip and an I²C chip without moving the breakout, etch341 won’t be able to talk to the new chip — see the I²C troubleshooting section.

Voltage

The CH341A black module has a 3.3V/5V jumper near the USB end.

• 3.3V — correct for every 3.3V chip in the database (W25Q, W25X, MX25L, GD25Q, SST25VF, AT25SF, EN25QH, P25Q, IS25LP).
• 5V — will damage every chip in the database. Do not flip the jumper to 5V unless you know exactly why.

For the 1.8V chips in the database (W25Q*JW, MX25U, GD25LQ), the standard 3.3V/5V jumper isn’t enough — these need a real 1.8V rail. Two options:

1. CH341A V1.7 module — the V1.7 has a separate switch alongside the 3.3V/5V jumper that selects 1.8V. Flip the 3.3V/5V jumper to 3.3V and flip the V1.7’s 1.8V switch on, and the output rail becomes 1.8V.
2. Level-shifter adapter — a separate breakout PCB that sits between the CH341A and the 1.8V chip, regulating the supply rail to 1.8V and translating logic levels accordingly. Several vendors sell these; the typical V1.7 module obsoletes the need for most users.

Plugging a 1.8V-only chip into a 3.3V programmer over-volts every input pin and can permanently damage the chip in seconds. Always check the chip’s datasheet for its actual supply range before hooking anything up.

Sanity check: VCC reaches the chip

Before running any operation, especially in-circuit, take 10 seconds and probe pin 8 of the chip with a multimeter (red probe to pin 8, black probe to any GND on the board, with the CH341A plugged in to USB):

• ~3.3V — power is good (or ~1.8V for U-series chips in 1.8V mode).
• 0V — clip’s not seated, the wire to pin 8 broke, or in-circuit there’s a power switch in series with the chip’s supply.
• Anything between 0V and 3.3V — the in-circuit host is partially powering the rail through some other path. Fighting the programmer; lift pin 8 from the PCB to isolate.

In-circuit programming

Clipping a chip that’s still soldered to a board is convenient but runs into the host’s other power and signal paths.

MISO stuck low (0x000000)

The host’s SPI controller actively drives MISO low even when the board is “powered off”. etch341 detect returns JEDEC ID : 0x000000 and verbose mode shows clean command bytes going out but nothing meaningful coming back.

Diagnosis:

etch341 detect -v        # with the clip OFF the chip — nothing else changed

• <- IN [4]: ffffffff → the CH341A is fine; the target board is the culprit. Move to the recovery steps below.
• <- IN [4]: 00000000 → CH341A or wiring issue, not the target.

Recovery, in order of effort:

1. Disconnect the host’s power completely. Pull the AC plug and any battery; let any standby caps bleed for 30 seconds. Some boards still drive MISO from a 3.3V_AUX rail.
2. Lift pin 7 (HOLD# / RESET#) of the chip from the PCB pad with a hobby knife — the chip’s internal pull-up then keeps HOLD# high. Reversible.
3. Lift pin 8 (VCC) and inject 3.3V externally so the chip is fully isolated from the board’s supply path. Reversible.
4. Desolder the chip entirely and program it in the ZIF socket. Most reliable; needs a hot-air station.

HOLD# being held low

Some game cams / IoT devices park their flash by driving the chip’s HOLD# pin (pin 7) low — the chip ignores all SPI commands until HOLD# goes high. Symptom is the same as MISO stuck low; the recovery is the same (lift pin 7).

Is the chip really an I²C EEPROM?

Not every 8-pin SOIC on a board is an EEPROM. PCIe risers, mining gear, and many cheap consumer boards have SOIC-8 buck regulator controllers that look identical from the outside. If i2c scan returns empty and the chip has a marking like MT…, MP…, or AOZ…, it’s probably a switching regulator. Tell-tale signs nearby:

• A large cylindrical inductor (matching the chip’s pin spacing for the SW node)
• Tantalum or large-value ceramic input/output capacitors
• A small Schottky diode

EEPROMs typically sit alone, with only small bypass caps. If the chip is between two big caps and an inductor, it’s almost certainly a regulator, not memory.