A hashboard is a serial chain of ASIC chips. Here's the physical principle behind how the Brit Tester talks to that chain, chip by chip, without needing a complete miner.
Every ASIC on an Antminer hashboard is wired into the same serial daisy-chain: command and clock signals travel from chip 1 down to chip N, and responses travel back up the same path. The tester connects directly to this chain — the same interface the miner's own control board would use — plus a separate I2C bus to the board's PIC and EEPROM.
Because it's a chain, a single break, weak solder joint, or dead chip has a predictable effect: every chip downstream of the fault stops responding. That's the whole basis of the diagnosis — the tester doesn't guess, it reads exactly where the chain of responses stops.
Simplified signal path: the tester drives the same command chain the miner's control board would, and reads a separate I2C bus for board identity and temperature.
Since every chip forwards the signal to the next one, a break at chip 12 means chips 13 through N stay silent — regardless of their own health. Reading how many chips respond, and where the responses stop, points straight at the fault without disassembling anything. The full manual documents the dichotomic method used to pin down the exact chip.
Each test asks a progressively harder question of the same chain. A board only needs to go as deep as the situation requires.
A lightweight, continuous read-id sweep with no load pattern. Answers a single question fast: how many chips respond right now? Ideal for a first look at an unknown board or a quick sanity check after a repair.
Runs real verification patterns through the chain and requires a minimum number of valid responses per chip. This is the pass/fail gate for physical integrity: soldering, continuity, power delivery.
Repeats the pattern test across independent cycles. A chip that passes once but fails on a later cycle is exactly what this step is built to catch — the kind of intermittent fault a single quick test misses entirely.
Real jobs at the chip's own operating frequency, read straight from the EEPROM. Every chip must respond in every cycle — the closest thing to production conditions without plugging the board into a live miner.
See exact thresholds and timing in the manual →
A full miner tells you pass or fail at the board level — it won't tell you which of 100+ chips is the problem, and it needs a complete, working chassis, PSU and network connection just to power on. The tester talks to the chain directly on an open bench: no miner assembly required, and results are per-chip, not per-board.
A multimeter reads voltage levels — useful, but it can't tell you whether a chip is actually answering commands correctly. LiveScan reads the real protocol response from every chip in the chain, live, which is the only way to know how many chips are actually communicating right now.
ChainCheck proves a chip can answer once, briefly. Some chips are marginal: they respond fine under a short test but drop out under sustained load or heat. That's precisely the gap CoreCheck and FullLoad are designed to close, by repeating the test across independent cycles instead of relying on a single pass.
Yes — that's the point. You supply GND/+V from a bench power supply, connect the tester's signal cable, and Brit handles the rest: chain reset, chip discovery, and every test in the ladder. No miner control board is involved at all.
LiveScan and ChainCheck cover the physical integrity question every shop needs day to day. CoreCheck, FullLoad and the EEPROM editor are the tools for deeper diagnostics and board recovery — higher-value work that a repair shop bills for differently. The split mirrors that, not an artificial paywall on features the hardware already runs.