HGV Brake Test: Roller Tests and How Often
DVSA expects brake performance to be assessed at every safety inspection, and — where EBPMS is not used — expects at least four laden brake tests spread evenly across the year, a count that can include the annual test. The assessment can be a roller brake test, an electronic braking performance monitoring system (EBPMS), or — in the Guide's words — "a decelerometer with temperature readings", but an unladen roller test tells you very little, which is where most operators lose the value.
The recurring problem is not that operators skip brake testing. It is that they pay for tests carried out on empty vehicles and file the printouts without reading them.
General guidance, not legal advice. Brake testing standards and inspection expectations are set by DVSA and change periodically. Check current GOV.UK guidance and speak to your maintenance provider about your specific vehicles.
How Often
DVSA's Guide to Maintaining Roadworthiness sets two separate expectations that are easy to conflate:
Every safety inspection should include a brake performance assessment. So if your PMI interval is 6 weeks, you are assessing brake performance roughly nine times a year.
Where EBPMS is not used, at least four of those assessments should be laden brake tests. The Guide's wording is: "Where EBPMS is not used, it is expected there will be a minimum of four laden brake tests spread evenly across the year, this can include the annual test."
That conditional matters. EBPMS is at present "only available for use on trailers", and trailers fitted with it "would only require a laden brake test if there was insufficient data to produce a report to confirm the correct operation of the system or where work has been carried out on the braking system that may have affected braking performance." So a trailer on working EBPMS is not carrying the four-test expectation; a rigid, a tractor unit, and any trailer without EBPMS is. Where a satisfactory EBPMS evaluation report is not available, or a laden roller brake test is not carried out with the safety inspection (unless a laden brake test report from within the last 14 days is produced), the Guide requires a risk assessment by a competent person instead — either gap on its own triggers it, not both together.
The Guide's inspection frequencies themselves run "between 4 to 13 weeks" depending on the work: roughly 10-13 weeks for light loads and easy conditions, 6-10 weeks for general haulage, 4-6 weeks for arduous work, 4 weeks for off-road or difficult operation, and a minimum of 6 weeks for vehicles 12 years and older. Our guide to the DVSA Guide to Maintaining Roadworthiness covers how to set and justify your interval.
"Spread evenly across the year" is doing real work in that sentence. Four laden tests in March and none after is not compliance with the expectation — it is four tests.
Why Laden Testing Matters
This is the part that changes outcomes.
The DVSA heavy vehicle brake test best practice guide is explicit: "Vehicles must be fully loaded before being roller-brake tested because the grip between the tyre and the rollers is more effective that way." The guidance sets the working target as loaded to at least 65% — and not less than 50% — of design axle weight on each axle. Separately, vehicles and trailers must be loaded when tested unless exempt by design.
The physics is straightforward. An empty axle lifts and the tyre slips on the rollers before the brake reaches its real limit. The machine records the point at which the wheel locked, not the point at which the brake ran out of capacity. An unladen test on a vehicle with a genuine braking defect can produce a comfortable-looking result.
Operators who test unladen all year and then present a laden vehicle at the annual test frequently discover a problem they have been carrying for months.
Practical ways to hit 65%: schedule the PMI when the vehicle is returning loaded rather than after tipping; use ballast at the workshop; or, for trailers, move to EBPMS, which measures performance from real laden running rather than from a bay and displaces the four-laden-test expectation while it is producing sufficient data. If your provider is testing unladen because that is what arrives, that is a scheduling problem, not a technical constraint.
Reading the Report
Imbalance
Imbalance is the difference in braking force between the two sides of an axle, and it is the figure most worth watching because it moves before efficiency does.
The DVSA guidance on understanding your HGV or trailer's brake test report states: "The vehicle will fail if the brake imbalance is more than 30%."
But it also gives an action threshold well below the failure point: "If your vehicle passes the imbalance check but the imbalance is over 25% you should investigate why this is and look to adjust your brakes to lower the imbalance."
That 25-30% band is the single most useful thing on the report. A vehicle sitting at 27% imbalance has passed and is telling you a brake is on its way out. Treating a pass as "nothing to do" is how a pass at one PMI becomes a failure at the next.
Efficiency
The pass/fail efficiency thresholds — expressed as a percentage of vehicle weight — are set out in the DVSA HGV inspection manual and underpinned by the Construction and Use Regulations. The brake test report guidance does not restate those percentage figures for the general case. The only minimum figures it gives are in its section on the unladen tri-axle semi-trailer test — a test it says "only applies to unladen tri-axle semi-trailers. It's not used for any other type of unladen trailer" — where the service brake must reach a minimum total brake force in kilograms depending on how many wheels lock, and the parking brake must reach "16% of design GVW" where any wheel on which the parking brake operates does not lock. If you run a rigid, a tractor unit, or any other trailer type, those tri-axle figures are not your thresholds. The manual sets different service brake figures for motor vehicles and for trailers, against different weights: 50% of design gross vehicle weight for a modern rigid or tractor unit, against 45% of design total axle weight for a semi-trailer manufactured from 1 October 1982. Read the band for your specific vehicle rather than carrying one number across the fleet — our DVSA Guide to Maintaining Roadworthiness guide sets out both sides.
The report will also show a pass value alongside the recorded result, so you can see the margin. A vehicle passing by a whisker is worth a conversation with the workshop.
Other checks on the report
The report covers more than headline numbers — brake bind, time lag between control application and brake response, and ovality, which the DVSA guidance describes as measuring "if there is any difference in the braking effort as each wheel rotates" (in workshop practice, that fluctuation usually points at a distorted drum or disc, though the guidance itself does not say so). Each points at a different cause, so it is worth asking your provider to walk you through a report once rather than filing them unread.
Recording It
Brake test results form part of the safety inspection record. The Guide requires safety inspection records to be kept for at least 15 months, including for vehicles later removed from the licence or sold — so the brake printout belongs with the inspection sheet, not in a separate pile.
At an operator premises visit, an examiner assessing your maintenance system will look for whether brake performance was actually measured, whether laden tests happened across the year, and whether deteriorating figures were acted on. A run of printouts showing imbalance climbing from 12% to 28% with no intervention is worse evidence than no printouts at all, because it documents a known and ignored trend.
The vehicle maintenance planner guide covers scheduling inspections so the laden tests land evenly, and the HGV MOT guide covers the annual test that can count as one of the four.
A Workable Routine
- Confirm your PMI interval is justified for your operation and vehicle age.
- Book a brake performance assessment at every safety inspection.
- Unless the trailer is on working EBPMS, plan four laden tests across the year — one per quarter, with the annual test as one of them.
- Schedule PMIs to catch vehicles loaded, or agree ballast arrangements with the workshop.
- Read every report. Watch imbalance, not just pass/fail.
- Act on anything over 25% imbalance even though it passed.
- File brake reports with the safety inspection record and keep for 15 months.
- Track the trend per vehicle — the direction matters more than any single figure.
To check how maintenance and brake testing sit against the rest of your O-licence obligations, run the free O-Licence Compliance Health Check.