One power chain, three documentation surfaces
The SO-101 follower arm is a six-axis desktop arm published by Hugging Face LeRobot and TheRobotStudio. At the source revisions RCI audited — LeRobot commit d23b654 and SO-ARM100 commit 7629d2a — the follower configuration uses six FEETECH ST-3215-C001 servos (7.4 V variant, 1:345 gear ratio) on a serial bus fed by one motor controller board. The procurement row for that board is generic, but the pinned retailer links and LeRobot's pinned setup instructions identify a Waveshare bus-servo board; Waveshare's current page identifies it as Bus Servo Adapter (A), SKU 25514.
Every electrical element of this chain therefore has a public documentation surface: the responsibility party's build documentation, the servo manufacturer's product page and the controller manufacturer's product page. On the single question of supply voltage, those three surfaces do not agree.
RCI's editorial policy retains conflicting evidence instead of quietly picking a winner, and RCI 001 defines Conflicting as its own evidence state: both values and their conditions are preserved. This note applies that state to a whole power chain.
The three statements, side by side
| Source | Statement | Scope / revision noted | What it covers |
|---|---|---|---|
| TheRobotStudio SO-ARM100 README (responsibility party) | Uses the 7.4 V C001 follower path and links a 5 V power supply, noting the servos may produce less torque at 5 V. The pinned Japan regional link resolves to the ATS024T-W050U: 5 V / 4 A, 5.5 × 2.1 mm barrel, centre-positive. | Commit 7629d2a (2026-08-05); regional retailer links, no supply revision binding. | The documented build's supply path for the whole arm. |
| FEETECH ST-3215-C001 product page (servo manufacturer; manufacturer listing as recorded by RCI on 2026-08-14) | Identifies the ST-3215-C001 as a 7.4 V model and publishes a 6–7.4 V operating range. | Current product page; no pinned revision identifier. | The electrical boundary of the six bus servos. |
| Waveshare Bus Servo Adapter (A), SKU 25514 (controller manufacturer) | Publishes a DC 9–12.6 V input range for ST-series servos. | Current product page; the physical PCB revision behind any given retail unit is not pinned. | The input the controller board is stated to accept and pass to the servo bus. |
Taken literally, a 5 V supply sits below the servo maker's 6 V floor and well below the controller maker's 9 V floor, while the servo's 7.4 V rating sits below that same 9 V floor. The three statements cannot all describe one safe, consistent power chain unless they refer to different hardware revisions, variants or conditions — and none of the surfaces says which.
Three credible sources, one chain, no single voltage that satisfies all of them as written.
What this does and does not mean
It does not prove any document is wrong. Manufacturer pages describe current products without pinned revision identifiers; the build documentation is pinned to a commit. A board revision change, a servo variant boundary or an unstated tolerance could explain part of the spread. RCI has not established which, and will not guess.
It does mean the combination cannot be assumed. Until the responsibility party or the manufacturers confirm in writing which exact controller revision, supply and servo combination they stand behind, no supply voltage for this exact chain is supported by all of its own documentation at once.
Retail links are not compatibility evidence. A retailer page linking a board next to a supply shows what is sold, not what is specified. One retailer catalog surface has even shown inconsistent 5–12 V and 9–12.6 V entries for the same board sales code, which is exactly why RCI resolves electrical boundaries against manufacturer and responsibility-party statements rather than storefronts.
RCI holds the conflict open
This conflict is recorded as a documented conflict, not resolved by interpretation. While it stays open, RCI proceeds to no hardware acquisition, no wiring and no power application for this exact chain. RCI has run zero physical tests of any element named on this page; every statement above is a documentation claim attributable to its source.
If the responsibility party or either manufacturer publishes or provides a written clarification — an exact board revision, an input-voltage notice, a revised BOM row — RCI will record it here with its date and scope, and the prior state of this note will remain visible.
This note is not wiring advice and not a safety verdict. It states only that three public voltage statements about one power chain are currently incompatible as written, so the combination should not be assumed compatible.
Seen the missing piece?
If you can point to a revision-controlled schematic, datasheet or responsibility-party statement that closes any part of this conflict, submit it through RCI corrections. Structured evidence with a source URL and date is enough; the correction trail stays public.