What to Do if Your Autoclave Fails Validation

The engineer’s just left. The validation report’s in your hand. It says fail.

Here’s what to do today, what to stop doing right now, how to work out what’s actually wrong, and the decision tree on whether to repair or replace.

First: stop running clinical loads

Before anything else, take the autoclave out of service for clinical work until the problem’s understood and fixed. That isn’t a CYA position, it’s the explicit requirement under HTM 01-05 paragraph 4.21, which states: “If the sterilizer fails to meet any of the test requirements, it should be withdrawn from service and advice should be sought from the manufacturer and/or maintenance contractor. Any instruments processed in an unsuccessful cycle should not be used.”

Any instruments processed in the last few cycles where you’re not confident the machine was hitting target parameters should be re-processed. Document the decision in the logbook now while it’s fresh.

That’s the immediate action. The rest of this article is about working out what’s going on and what to do next.

What “validation failed” actually means

HTM 01-05 paragraph 12.2 sets the validation requirement:

“Validation is needed for new equipment at installation and annually thereafter. It will also be necessary to validate equipment after any major repairs have been carried out.”

Validation is a performance qualification, independent calibrated probes go into the chamber, the engineer runs a set of cycles, and the temperatures, pressures and dwell times the cycle actually achieves get measured and compared against the manufacturer’s specification.

A pass means every parameter is within tolerance across every cycle tested.

A fail means at least one parameter is outside tolerance somewhere. That could be a single cycle at a single load configuration, or a systematic issue across the whole run. The validation report will tell you which.

The fail isn’t proof the machine is unsafe to use. It’s proof the machine isn’t currently demonstrating it’s working to specification. The two sound similar; they’re different in practice.

What’s actually wrong: diagnosing the fail

Common causes of failed validation, in roughly the order we see them at service:

Door seal degradation. The most frequent single cause. A worn or compressed seal lets steam escape during the sterilising phase, which drops chamber pressure below target. The validation print shows the temperature struggling to hold. Visible on inspection: the seal looks flattened, has set-marks where the door closes against it, sometimes a visible split.

Vacuum pump performance issues (on Class B/S vacuum machines). Pre-vacuum and post-vacuum stages don’t pull down to the required vacuum. Pre-vacuum failure shows as steam-penetration issues (the Bowie-Dick or Helix tests start failing); post-vacuum failure shows as wet loads at the end of cycle. The validation print shows the vacuum phase plateauing high.

Sensor drift. Pressure transducer or temperature probe has drifted out of calibration. The machine thinks it’s at temperature and pressure, but the calibrated validation probes say it isn’t. Symptoms vary depending on which sensor and which direction.

Steam supply / feedwater quality. Hard water leaving limescale on the heating element causes inefficient steam generation. The element runs hotter to compensate but the chamber takes longer to come up to temperature, sometimes never quite getting there at full load. The longer write-up on water quality is in our RO water article, but the short version is: distilled or RO water only, never mains.

Valve wear. Pressure relief valves, vacuum valves, water inlet valves all wear with cycle count. A leaking valve shows in different parts of the validation depending on which one.

Mechanical wear on chamber seal face. Less common but real on older units, the chamber’s sealing face has corrosion or pitting that the seal can’t compress against properly.

A competent engineer reading the validation report knows which fault pattern produced which fail. The report should tell you which parameter failed, on which cycle, and at what load, that points the engineer at the cause.

Should you repair or replace?

The decision points we’d actually use:

Repair if:

  • The unit is under 8 years old
  • The fault is a single component (seal, pump, valve, sensor) replaceable for under a third of the cost of a new unit
  • Parts are readily available in the UK for the brand
  • The machine’s service history is otherwise clean
  • A repaired unit will continue to pass validation reliably for at least another two years

Replace if:

  • The unit is over 10 years old AND failing validation isn’t a new pattern
  • Multiple components have failed in the same year
  • The chamber itself has corrosion or pitting issues
  • Parts aren’t easy to source, discontinued brands, grey imports, end-of-line models
  • The unit has had three or more breakdowns in the last 12 months
  • The capital cost of repair plus the next round of likely repairs would buy a useful chunk of a new unit’s depreciation period

The honest position is that an autoclave is a workhorse pressure vessel that does a known job for a known number of cycles. Once the cumulative repair costs start approaching 40% of replacement cost in a single year, you’re spending money to defer a decision rather than to fix a unit.

After the repair: full revalidation

Whatever the fix, the autoclave can’t go back into clinical service until it’s been revalidated. HTM 01-05 paragraph 12.2 is explicit: “It will also be necessary to validate equipment after any major repairs have been carried out.”

That’s a full calibrated-probe run, not just running the next morning’s Bowie-Dick and assuming it’s fine. The fix needs to be evidenced. The revalidation report is what closes the loop and lets the machine go back on the clinical roster.

We package the engineering fix and the post-repair revalidation as a single visit where the fault is one our engineer can address on site. For deeper issues that need parts shipped in, we schedule the repair and the validation on consecutive days so you’re back in service as fast as possible.

What goes in the logbook

The whole sequence needs documenting:

  • The original validation visit and the fail report
  • The decision to withdraw from clinical service
  • The diagnosis, what the engineer found and what they did
  • The repair details, parts replaced, dates
  • The revalidation visit and the pass report
  • The date the machine returned to clinical use

That’s what a CQC inspector wants to see when they ask “what happened the last time your autoclave failed validation?” The good answer demonstrates the practice noticed the fail, acted on it, documented the action, and returned the machine to service properly.

What to do today if you’ve just had a fail

Specifically, in order:

  1. Stop running clinical loads through that autoclave
  2. Talk to your service engineer, discuss the validation report findings
  3. Document the situation in the logbook with the date and the validation report attached
  4. Decide repair-or-replace based on the diagnosis
  5. Authorise the work
  6. Confirm post-repair revalidation is scheduled before the machine goes back into service
  7. File all the paperwork in the logbook so the audit trail is clean

If your validation engineer was us, we’ll be on the phone before you need to chase. If it was someone else and you’d like a second opinion or a competing repair quote, ring 01274 265566 and we’ll talk you through the report findings and what a sensible repair plan looks like.

If the repair-vs-replace decision is leaning toward replace, our three Class B ranges are on the practice equipment page, Castellini C Series, Stern Weber SW Series and Mocom B Classic. Quoted with installation, commissioning, validation and first-year service as a single package.

The wider compliance picture sits in our HTM 01-05 validation guide and the weekly-test-fail guide, read those alongside this one for the full compliance picture.

Further reading

Table of Contents

Featured Articles

Contact Us!