Machine Design #72: Timing Budgets and Timeouts — Do Not Pick Five Seconds Just Because It Feels Long
1. Start with the function and the evidence
Before choosing a setting, dimension, sensor, or component, write what the machine timing must do, where its input comes from, how it behaves at the boundary, and what counts as failure. A phrase such as “make it like the old machine” hides assumptions about timing, ownership, access, maintenance, and safety. A testable requirement states the input condition, operating condition, expected result, acceptance limit, and measurement method. Q-up guidance is a starting point; the final decision must be checked against the real model, process, operator, and installation.
2. Four layers must be reviewed together
Function and load
Separate nominal, start-up, impact, misalignment, fault, and recovery conditions. Trace the result through the complete system, not only one block on a drawing. Include inertia, temperature, network delay, pressure, cleaning, operator action, and the cycles that actually matter. State the boundary and the reason for the selected margin.
Design and implementation
The same requirement behaves differently when it is implemented in mechanics, controls, software, data, or a work instruction. The drawing and the program must describe what the workshop and the operator can make, verify, and restore. A tight number is not a substitute for an understood process.
Interfaces and ownership
Define the datum, signal, identifier, direction, access, and handoff at every interface. If model variants exist, use geometry, permissions, or explicit selection to prevent a wrong part or recipe. Record who owns the value and who can approve a change.
Operation and maintenance
Check access, replacement time, recovery position, cleaning, adjustment, and the evidence that returns the machine to baseline. A design that works only with one experienced person is not repeatable.
3. Core checks for this topic
- Cycle steps, nominal time, and worst-case time: record the assumption, design value, source, measurement method, and pass/fail criterion.
- Sensor, actuator, network, and controller response: record the assumption, design value, source, measurement method, and pass/fail criterion.
- Timeout boundary linked to a physical failure mode: record the assumption, design value, source, measurement method, and pass/fail criterion.
- Retry limit, alarm, and retained diagnostic context: record the assumption, design value, source, measurement method, and pass/fail criterion.
- Safe stop and restart after a late response: record the assumption, design value, source, measurement method, and pass/fail criterion.
- Measurement at min/max load, temperature, and pressure: record the assumption, design value, source, measurement method, and pass/fail criterion.
The checklist is useful only when every line has evidence. “Reviewed” is not “calculated”, and “calculated” is not “tested at the boundary”. Keep an evidence ID next to each requirement so an engineering change can be audited.
4. Variation, limits, and failure modes
Build the chain from the functional input to the characteristic that must be protected. Separate nominal, worst-case, and statistical values only when the process is stable enough to justify it. Include human action, delay, data loss, wear, contamination, and recovery time. If the system succeeds only because someone nudges it, the design has no reliable capability.
Ask: “If this is wrong, what will the machine show, and what evidence will reveal the cause?”
| Failure mode | Machine symptom | Verification direction |
|---|
| Boundary selected too optimistically | Passes a demo, fails at speed, load, or a late response | Test min/max and the defined fault case |
| Requirement not tied to function | Play, bind, wrong state, or wrong record | Rebuild the chain from the functional datum |
| Process or data step omitted | Drift, missing evidence, or unrecoverable stop | Review the route and measure each critical step |
| Recovery not designed | Long stop or unsafe restart | Run an interruption and recovery trial |
| Documents out of sync | Correct name but wrong revision or setting | Baseline drawing, software, recipe, and work instruction |
For this article, challenge the title risk directly: cycle-time budget, response variation, timeout evidence, retries, and safe stop. A sentence in a report does not close a failure mode; a measured result, an owner, and a re-test condition do.
5. Release and supplier discipline
Specify functional requirements that can be inspected. For a special requirement state its scope, datum or data source, measurement condition, and acceptance limit. Notes such as “handle carefully” do not tell a supplier or operator what to do. Keep the rationale, revision, assumptions, review comments, prototype or FAT result, and the point to recheck after change.
On the real machine, run start-up, normal duty, stop, restart, interruption, and a controlled fault. Record force or torque, temperature, timing, alarms, identifiers, visual marks, and replacement or recovery time as relevant. These observations reveal assumptions that a drawing review missed.
6. MINATA release checklist
- [ ] Function, input, duty cycle, and fault or environment boundary are written.
- [ ] Material, process, control, data, and inspection method are agreed.
- [ ] Functional interfaces and ownership are visible.
- [ ] The six topic checks above have evidence and pass/fail criteria.
- [ ] Assembly, selection, recovery, and mistake-proofing were tried on hardware.
- [ ] Failure modes have an owner, evidence, and re-test condition.
- [ ] Supplier and incoming records link to the revision and lot.
- [ ] Commissioning baseline and maintenance response are defined.
7. Practical review passes
Review pass 1: machine timing
Input and boundary: list every external condition, lower limit, upper limit, and simultaneous worst combination. Write the physical reason for each limit and the instrument or log that proves it. For machine timing, keep the observation concrete: cycle-time budget, response variation, timeout evidence, retries, and safe stop. Record the value, unit, condition, owner, and pass/fail decision.
Review pass 2: machine timing
State and transition: draw the normal path, the interruption path, the timeout path, and the recovery path. For every transition name the permissive, the stop action, and the evidence retained for diagnosis. For machine timing, keep the observation concrete: cycle-time budget, response variation, timeout evidence, retries, and safe stop. Record the value, unit, condition, owner, and pass/fail decision.
Review pass 3: machine timing
Manufacturing and installation: ask how the feature, value, identifier, or record is created, measured, labeled, and changed. Compare capability with the tolerance or timing budget before release. For machine timing, keep the observation concrete: cycle-time budget, response variation, timeout evidence, retries, and safe stop. Record the value, unit, condition, owner, and pass/fail decision.
Review pass 4: machine timing
Human and maintenance: observe a first-time operator or technician. Measure access, tool use, selection, cleaning, replacement, recovery, and the time needed to return to baseline. For machine timing, keep the observation concrete: cycle-time budget, response variation, timeout evidence, retries, and safe stop. Record the value, unit, condition, owner, and pass/fail decision.
Review pass 5: machine timing
Data and audit: search for a real lot or serial and confirm that the answer is complete, ordered, protected, and understandable to someone who did not build the machine. For machine timing, keep the observation concrete: cycle-time budget, response variation, timeout evidence, retries, and safe stop. Record the value, unit, condition, owner, and pass/fail decision.
Review pass 6: machine timing
Change and learning: record what changed, which assumption it touches, which evidence must be repeated, who approves it, and how the result is fed into the next revision. For machine timing, keep the observation concrete: cycle-time budget, response variation, timeout evidence, retries, and safe stop. Record the value, unit, condition, owner, and pass/fail decision.
Review pass 7: machine timing
Boundary trial: combine the most difficult load, temperature, pressure, model, delay, contamination, and operator sequence that can occur together. Do not close the review until the measured result is attached. For machine timing, keep the observation concrete: cycle-time budget, response variation, timeout evidence, retries, and safe stop. Record the value, unit, condition, owner, and pass/fail decision.
8. Closing note
Good engineering is a chain of explicit assumptions that survives manufacturing, operation, maintenance, and change. For machine timing, the right question is not only “will it work?” but “what evidence will show it continues to work when the boundary moves?” That is the MINATA standard for turning a drawing, sequence, or record into a dependable machine.
Public references
- MINATA Q-up design practice and internal review checklist.
- Applicable mechanical, controls, data, or safety standard selected by the project owner.
View all MINATA technical articles