You walk in at 8am, the autoclave throws a fault on the helix test, and there’s a full appointment book waiting outside. Before you panic and ring an engineer, take a minute. Most of the test failures we get called out for turn out to be five-minute fixes the practice could have done themselves, and the ones that aren’t are usually easy to diagnose once you know what each test is actually telling you.
This is a working-through of what to check first, what the different test failures mean, and the point at which you need to stop using the machine and pick up the phone.
Before anything else, take the autoclave out of service for sterilising real instruments until you’ve worked out what’s gone wrong. That isn’t a CYA suggestion, it’s the regulatory position. HTM 01-05 paragraph 4.21 is explicit: a sterilizer that fails any test “should be withdrawn from service and advice should be sought from the manufacturer and/or maintenance contractor.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.” CQC’s dental mythbuster 12 reinforces the validation-records expectation, and the PSSR 2000 Written Scheme requires a competent person to confirm the system is safe to operate.
Any loads you’ve processed since the last passing test should be re-processed once the machine is back. Put a note in the logbook now while it’s fresh.
That doesn’t mean every test failure is a disaster. It means the next five minutes you spend figuring this out are the ones that matter.
Practices run several tests on an autoclave and people use “weekly test” loosely to mean any of them. The diagnosis is different in each case.
Daily Automatic Control Test (ACT). Runs on every cycle. If this is what failed, the machine itself is telling you the cycle didn’t reach validated parameters. Pressure, temperature or hold time didn’t land in spec. Could be a one-off, could be a sensor issue, could be a steam supply problem.
Daily steam penetration test. On a vacuum (Type B or Type S) machine you run either a Bowie-Dick (porous load) or a Helix (hollow load) test daily before clinical use. If this is what failed, the failure is almost always about air. Air not getting out, or air getting back in.
Weekly air-leakage test. Also called the vacuum leak test. This is the one most practices actually mean when they say “the weekly test failed”. The chamber pulls a vacuum, holds, and the machine measures whether pressure rises during the hold. A rising pressure means air is leaking in. By a long way the most common cause is the door seal.
Quarterly periodic tests and annual revalidation. These are done by a competent test person, usually the service engineer. If one of those failed during a planned visit, the engineer will already be telling you what’s wrong; you’re not in the panic-Google scenario this article is for.
So which test failed? That’s where you start.
The single most common reason an autoclave test fails is something the operator did, not something the autoclave did. Worth working through these in order before you escalate.
Door seal. Open the chamber, look at the seal. Is there a hair, a glove tip, a piece of pouch material caught in it? Is it sitting evenly in its groove? Is it visibly cracked, flattened, or shiny? Wipe it with a damp lint-free cloth, run a finger round the inside of the door rim looking for anything caught. Reseat the seal properly if it’s slipped. Re-run the test. If the seal cleans up and reseats, you’ve fixed it.
Test pack assembly. Bowie-Dick and Helix tests only work if the test device itself is assembled correctly. The capsule on the end of a helix tube screws on. If it isn’t tight, the test fails. If the Bowie-Dick sheet is upside down or folded the wrong way in its pack, it fails. If the indicator strip is expired (look at the date on the box) or has been stored badly (heat, sunlight), it fails.
Cycle selection. You’d be amazed how often the wrong cycle’s been picked. The Bowie-Dick test cycle and the Helix test cycle are different cycles on the autoclave’s menu. So is the routine B-cycle. If someone’s switched the menu to a Class S cycle on a unit that supports both, the test result is meaningless.
Chamber temperature. Some test packs require the chamber to be warm before the test cycle runs, pop a routine warm-up cycle through it first. Running a Bowie-Dick into a cold machine is a known false-fail.
Water supply. If the unit’s reservoir is low, the cycle can fail to reach pressure. Check it.
If you’ve worked through all five and the test still fails on a re-run, the next paragraph is for you.
Stop here and call an engineer if:
HTM 01-05 also requires that anything more than user-level maintenance is done by a competent person. Practically, that means: yes to wiping the seal, checking the test pack, refilling the reservoir, looking for the obvious. No to opening the case, swapping pump components, recalibrating sensors, or anything else that requires tools.
If you’re in any doubt about whether what you’re looking at is operator-fixable, default to calling. A wasted half-hour engineer visit is cheaper than a CQC inspection where the logbook shows the machine has been failing tests on and off for months.
This part matters more than people think. The first thing a CQC inspector asks when they look at the validation records is “show me what happened the last time a test failed”. Your logbook needs to answer that question on its own, without you in the room.
Each failure entry should include: the date and time, which test failed, what you found when you investigated, what you did about it (cleaned the seal, replaced the strip, called the engineer), and the date and result of the re-test that passed. If an engineer attended, their report goes in the same record.
If the machine sat unused while you waited for an engineer, note that too. The point is to demonstrate that the practice spotted the failure, acted on it appropriately, and didn’t process clinical loads while it was outside spec. That’s the standard you’re being measured against.
We get the panicked-8am phone call most weeks. About half the time the practice has worked through the obvious checks above and the seal or test pack was the answer, and we talk them through it on the phone. The other half the time the machine needs an engineer.
The repeat-offender failures we see most often:
Most of these are an engineer visit, a part swap, and a re-validated machine the same afternoon. The autoclaves we sell, and the ones we service, are mostly built to be opened up and worked on by a competent engineer, not thrown away.
If you’ve worked through the checks and the machine still won’t pass, give us a ring on 01274 265566 or contact us through the Service and Support form. We cover dental, podiatry, beauty and tattoo settings across the UK and our engineers carry the common parts on the van.
If you’d like a steadier footing than the panic call, most practices we look after are on a service contract that covers the quarterly periodic tests, the annual revalidation, and the PSSR 2000 written-scheme examination in one go. Have a look at the Service & Maintenance page for what that looks like, and our walk-through of HTM 01-05 validation for the longer answer on what compliance actually requires.
The shortest version of this whole article: most failed tests aren’t broken machines. They’re door seals, test packs, and chamber temperatures. Work through the five checks above before you panic, and when none of them are the answer, that’s when we come in.
