bolt retorque interval
Eliminating bolt retorque intervals
By ISOKLAMP Engineering, Inc. Editorial Team · Updated
A maintenance task that is really a design choice
Retorque intervals rarely appear in a design review. They appear in a maintenance manual, written after commissioning, once somebody noticed the joints were loosening. From that point the cost sits on a different budget from the one that created it, and it never gets revisited.
It is worth pricing properly.
Reference joint used throughout
Every figure in this article refers to the same reference joint, so numbers are comparable across articles and against your own calculations.
| Parameter | Value |
|---|---|
| Bolt | M16 × 2,0, property class 10.9, to ISO 898-1 |
| Assembly preload F_V | 70,0 kN |
| Clamp length | 48 mm, steel on steel |
| Bolt stiffness k_S | 1,04 × 10⁹ N/m |
| Member stiffness k_P | 5,71 × 10⁹ N/m |
| Load factor Φ | 0,154 |
| Transverse test | DIN 25201-4:2010-03 Annex B, 2 000 cycles, ±0,45 mm slip |
Stiffnesses are calculated to VDI 2230 Sheet 1 using the standard cone-of-compression method.
What a retorque actually costs
| Cost element | Typical |
|---|---|
| Technician time per accessible joint | 3 to 6 min |
| Technician time per joint behind a guard | 15 to 40 min |
| Lockout, isolation, permit where required | 20 to 90 min per area |
| Production downtime, allocated | Frequently the dominant term |
| Torque wrench calibration and records | Per-programme overhead |
| Rework where a joint is found failed | Unbudgeted, always |
Worked example. A packaging line with 2 400 bolted joints, a six-month retorque interval, four minutes per joint average, one shift of downtime per event.
2 400 joints × 4 min × 2 events/year = 320 technician-hours/year Plus 2 shifts of line downtime per year
At any realistic loaded labour rate plus downtime cost, this is a significant annual figure for a problem that was created by a specification decision costing a few euros per joint.
Retorquing restarts the mechanism it corrects
This is the part that gets missed.
When you retorque a joint you re-establish contact between surfaces that have already plastically deformed once. The asperity geometry is new, the contact pressure distribution has changed, and a second embedment cycle begins. It is smaller than the first, typically 30 to 50 percent of it, but it is not zero.
| Event | Embedment that follows |
|---|---|
| Initial assembly | 100 % of budget |
| First retorque | 30 to 50 % |
| Second retorque | 20 to 35 % |
| Third retorque | 15 to 30 % |
The series converges but never terminates. And each intervention carries the full labour cost regardless of how little embedment remains to recover.
Worse, for the mechanisms that do not self-limit, retorquing recovers nothing structural at all. Gasket creep and thermal ratcheting resume from the moment the wrench comes off, at the same rate as before.
When an interval can be eliminated
Three conditions, all necessary.
1. The total relaxation budget is bounded and known. Sum embedment, coating creep, gasket relaxation and thermal ratcheting over design life. If you cannot put a number on it, you cannot remove the interval, and the honest answer is to instrument a joint first.
2. A reserve exceeds that budget with margin. The reserve has to be at least the budget, and sensibly two to three times it, because relaxation estimates carry real uncertainty.
3. Reserve state is observable. Removing a scheduled task without a means of confirming the joint is still healthy trades one risk for another. Observation has to be cheaper than the task it replaces, or nothing has been gained.
Budget against reserve
Using the reference joint at 0,884 kN per micrometre and a 0,50 mm take-up reserve:
| Application | Life budget | Reserve used | Interval eliminated? |
|---|---|---|---|
| Machined steel, dry, isothermal | 20 µm | 4,0 % | Yes, comfortably |
| Coated steel structural | 68 µm | 13,6 % | Yes |
| Hygienic, PTFE gasket, CIP/SIP | 70 µm | 14,0 % | Yes |
| Aluminium busbar, 250 cycles/yr, 10 yr | 240 µm | 48,0 % | Yes, with inspection |
| CFRP, continuous matrix creep, 10 yr | 180 µm | 36,0 % | Yes, with inspection |
| Unbounded, unknown mechanism | Unknown | Unknown | No. Diagnose first |
The bottom row is the honest one. Where the mechanism has not been identified, the correct action is the witness mark test and a proper budget, not a different washer. See Non-rotational preload loss.
From calendar-based to condition-based
The replacement for a retorque interval is not "nothing". It is inspection, and it has to be cheap enough to be done.
| Calendar retorque | Condition-based | |
|---|---|---|
| Action per joint | Torque wrench, 3 to 6 min | Visual read, under 5 s |
| Isolation required | Usually | No |
| Downtime | Yes | No, done during normal walkdown |
| Joints touched | All of them | Only those showing depletion |
| Restarts embedment | Yes | No |
| Produces diagnostic data | No | Yes, depletion rate maps problem joints |
For the 2 400-joint line above, a visual walkdown replaces 320 technician-hours with a few hours of inspection plus a handful of targeted replacements.
The diagnostic row is the underrated one. A population of reserve indicators tells you which joints in a machine are actually working hard, which is information that does not currently exist anywhere in most plants. A joint depleting three times faster than its neighbours has a reason, and finding it before it fails is worth more than the parts.
Making the case internally
Retorque elimination is usually an easy business case and a hard organisational one, because the cost sits in maintenance and the fix sits in engineering.
- Count the joints on the interval and the minutes each takes, including isolation.
- Add allocated downtime per event. This is normally the largest term and normally omitted.
- Multiply by events per year and by remaining asset life.
- Compare against the delta part cost across those joints, once.
- Include the failures the interval does not currently prevent, since a six-month interval does nothing about a joint that fails in month two.
Point 5 is what usually decides it. Retorque intervals are set by convenience, not by the relaxation curve, so they leave real exposure between visits.
Reserve monitoring detail on Preload reserve indicator. Method comparison on Locking methods compared.
Frequently asked questions
What does a bolt retorque interval really cost?
For a line with 2 400 joints on a six-month interval at four minutes each, that is 320 technician-hours per year plus two shifts of allocated downtime. Downtime is normally the dominant term and normally omitted from the calculation. The cost sits on the maintenance budget while the decision that created it was made in engineering.
Does retorquing permanently fix preload loss?
No. Retorquing restores clamp force once and re-establishes contact on fresh asperity geometry, which starts a second embedment cycle at 30 to 50 percent of the original. The series converges but never terminates. For gasket creep and thermal ratcheting it recovers nothing structural, since both resume at the same rate as soon as the wrench comes off.
When can a retorque interval be eliminated?
When three conditions hold. The total relaxation budget over design life is bounded and known. A take-up reserve exceeds that budget with a margin of two to three times, because relaxation estimates carry real uncertainty. And the remaining reserve is observable cheaply enough that inspection costs less than the task it replaces.
What replaces a retorque schedule?
Condition-based inspection. A visual reserve read takes under five seconds per joint against three to six minutes for a torque wrench, needs no isolation or downtime, touches only the joints showing depletion, and does not restart an embedment cycle. It also produces diagnostic data, since depletion rate maps which joints in a machine are working hardest.
How much reserve does a typical application consume?
Against a 0,50 mm reserve: machined dry steel uses about 4 percent over life, coated structural steel 13,6 percent, a hygienic PTFE joint under CIP and SIP 14 percent, an aluminium busbar at 250 cycles a year over ten years 48 percent, and a CFRP joint under continuous matrix creep over ten years 36 percent.
Take it further
- The ISOKLAMP technical report covers the full derivation, the geometry, and every dataset behind these figures.
- Full residual clamp force dataset gives the residual clamp force numbers for ten securing methods, with sources.
- Design partner programme is open. Eight slots, two per sector. Bring us a joint that keeps failing and we will run a VDI 2230 Sheet 1 analysis on it.
Engineering questions go to engineering@isoklamp.com. An engineer answers, not a form.