MR-J4-A: why INP can turn on while the axis is still moving
Use PA10, PC24 and PD31 to interpret MR-J4-A in-position status, avoid comparing command pulses with encoder pulses, and distinguish amplifier INP from the PLC’s machine-completion logic.
Contents
- 1. INP evaluates a deviation window, not standstill
- 2. Read three parameters together
- 3. Check units before comparing E with PA10
- 4. Hypothetical records: why 80 pulses can be “in position”
- 5. Trace the signal chain without widening the window to hide a problem
- 6. Preserve evidence that another technician can check
If INP turns on before an axis stops, is the servo output faulty? In MR-J4-A position control, first establish what the signal actually evaluates. This article uses Mitsubishi Electric manual SH(NA)030107ENG-V (November 2025) and is limited to MR-J4-_A_(-RJ) standard position control with rotary servo motors and semi-closed-loop feedback. Do not transfer it directly to MR-J4-B, MR-J5, fully closed-loop control or positioning mode. First identify the source of any similarly named “positioning complete” bit on a PLC screen.
1. INP evaluates a deviation window, not standstill
Section 3.6.1(2), printed page 3-43, explains that INP can turn on when deviation-counter droop pulses enter the PA10 in-position range. With a large range, INP may remain on during low-speed operation. INP being ON while the axis moves is therefore possible and does not, by itself, establish an output-component fault.
A sufficiently small difference between command and feedback positions does not mean the command has ended. Semi-closed-loop feedback measures the motor side. INP alone consequently cannot establish that a table, fixture or workpiece has reached its required process position. Permission for the next machine action depends on the manufacturer’s complete sequence conditions.
2. Read three parameters together
| Parameter and page | Meaning in this manual revision | Diagnostic implication |
|---|---|---|
| PA10, in-position range; 5-18 | Initial value 100; setting range 0 to 65535, with units selected by the relevant PC24 digit | 100 is an initial setting, not a universal accuracy or safety criterion |
| PC24, rightmost digit; 5-48 | ___0: command input pulses; ___1: servo motor encoder pulses | PA10=100 cannot be assumed to represent the same position window when its units differ |
| PD31, third digit from the right; 5-61 | _0__: droop pulses within the range; _1__: command pulse frequency is 0 AND droop pulses within the range | This selection applies to amplifier software version B4 or later; verify the installed version and effective setting |
Read PC24 and PD31 as four hexadecimal digits, each representing 4 binary bits. Underscores stand for the remaining hexadecimal digits; this notation is not an instruction to overwrite the entire parameter. The PD31 entry also says command pulse frequency is judged as 0 after approximately 1 ms without a position-command input. This evaluates an input gap, not mechanical settling time, a zero-speed criterion or a safe-stop confirmation.
3. Check units before comparing E with PA10
Section 4.5.3(3), page 4-27, defines E as deviation-counter droop pulses in encoder-pulse units. Display n is command input pulse frequency in kpulse/s, before electronic-gear multiplication. With PC24’s rightmost digit at 1, E and PA10 share a pulse reference. With that digit at 0, first establish the conversion from the machine’s effective command scaling. Comparing the bare numbers is insufficient.
The same table notes that E shows only the actual count’s lower five digits. Near display limits, with negative deviations or incomplete records, do not copy a number without its sign and unit. Use existing synchronized records of command frequency, speed, deviation and actual INP state to investigate timing. Screenshots taken while switching between displays do not establish simultaneous values.
4. Hypothetical records: why 80 pulses can be “in position”
This is a teaching example, not a real repair or an instruction to reproduce the condition. Assume software B4 or later, PC24’s rightmost digit confirmed as 1, PA10=100, and PD31’s relevant digit as 0. The servo is normally enabled without an alarm. Records share one time base and have no truncated-display issue:
| Hypothetical instant | Existing record | Supported interpretation |
|---|---|---|
| A: commands still arriving | n=2 kpulse/s, nonzero speed, E=80, INP ON | 80 encoder pulses are inside the 100-pulse window. The selected condition does not require zero command frequency, so the observation is consistent |
| B: after command input ends | n=0, E=140, INP OFF | Deviation remains outside the window. Ending the command alone does not turn INP on |
| C: deviation subsequently decreases | n=0, E=60, INP ON | Deviation is back inside the window. This does not establish acceptable workpiece position or completion of the machine cycle |
If the actual PD31 digit is 1, reconsider instant A: examine input-pulse timing, the approximately 1 ms criterion, sampling alignment, software version and whether the monitored signal really is amplifier INP. Resolve that evidence conflict before replacing an amplifier. If PC24’s rightmost digit is actually 0, the direct 80-versus-100 comparison also fails until the unit conversion is established.
5. Trace the signal chain without widening the window to hide a problem
INP is already on during motion: check the PD31 condition, then PA10 and its units. Only if existing records cannot satisfy that condition should you investigate timing, signal mapping and feedback evidence further. Do not diagnose from a mistaken expansion of the signal name into “completely stopped.”
Amplifier INP is ON but PLC completion remains OFF: check the input actually received by the PLC, I/O assignment and other completion conditions in the ladder logic. Separate amplifier logic, signal transmission and machine sequencing. Do not force outputs or jumper inputs to “prove” downstream logic works.
INP stays OFF after commands end: first establish, in the correct units, whether deviation has actually returned inside the window. Then use existing speed, torque and alarm records to choose the next investigation. Widening PA10 merely to obtain completion changes the position deviation accepted by the sequence; it does not establish that the original problem is gone.
6. Preserve evidence that another technician can check
Record full amplifier and motor models, software version, control mode, complete current PA10/PC24/PD31 values, command-scaling configuration, the actual signal point assigned to INP, and time-aligned command, speed and deviation records. Distinguish amplifier INP, the PLC receiving bit and the PLC’s final completion bit. Identify values available only from unsynchronized screenshots.
This article interprets existing records obtainable safely. It provides no parameter-change, forced-output, live-wiring or trial-motion procedure. INP ON does not authorize maintenance. Qualified personnel must follow the machine’s approved isolation and residual-energy verification procedure for cabinet access, wiring or disassembly, together with Safety Instructions A-2.
Source: MR-J4-_A_(-RJ)/MR-J4-03A6(-RJ) Servo Amplifier Instruction Manual, SH(NA)030107ENG-V (November 2025, English original). §3.6.1(2), 3-43; §4.5.3(3), 4-27; §5.2.1, PA10, 5-18; §5.2.3, PC24, 5-48; §5.2.4, PD31, 5-61; Safety Instructions A-2. All references use printed page numbers. The hypothetical records and signal-chain analysis are editorial interpretations, not manufacturer field cases or prescribed tests.