Two SIPOS actuators at one company developed the same fault: a constant deviation of about 10% between command and feedback — with the command at 0%, feedback read about -9.7%; at 100% command, feedback reached only 86.7%. Instead of blindly swapping parts, the reporter took a full set of voltage measurements that documented the diagnosis beautifully — and the final culprit was a surprise.
Measured directly across terminals 1 and 2 of X3.1 (the position feedback output):
| State | Faulty actuator | Healthy actuator |
|---|---|---|
| No load, open circuit | ~27 V DC | 15.8 V DC |
| Full close | 0.7–0.8 V DC at only 1.8 mA | — |
| Full open | only 4.5 V DC | — |
| Connected to the DCS card | cannot reach 1–5 V | 1–5 V DC, normal |
The picture is clear: the faulty actuator's position-feedback voltage reads abnormally high unloaded, but collapses under load — connected to the DCS card it can never deliver the standard 1–5 V (corresponding to 4–20 mA). That is the direct cause of the fixed-ratio deviation between command and feedback.
The reporter's path was typical:
1. First suspected the feedback board — replaced it with a new one, no change;
2. Then suspected the gear-train potentiometer — replaced the whole assembly, no change;
3. Finally replaced the power supply board — and it was cured.
Both the generation and the output of the position feedback signal depend on the actuator's internal supply. With a degraded power supply board, load capability drops and the feedback signal collapses the moment a load is attached. Unloaded measurement looks "healthy enough," but under load the true colors show. The value of this case is precisely that it documents an easily missed failure mode: the feedback board and the potentiometer are fine — the culprit is the power board feeding them.
Peer consensus agrees: SIPOS faults are essentially "control board or power board" — "you either replace the control board or the power board." The suggested order: measure toward the cheaper, higher-failure-rate power board first; do not start by replacing signal-chain boards.
1. For a "fixed-ratio command/feedback deviation," take unloaded vs. loaded comparison measurements first — collapse under load is the fingerprint of a supply problem;
2. Component-swap order should go from power to signal: power board → feedback board → gear-train potentiometer;
3. Measurement data (the four sets: unloaded / full-close / full-open / loaded) are the eyes of diagnosis — record them completely and the direction will never be far off.
