Brake Testing for the MOT (part II)

Note to MOT Testers: The following article appeared in November 2007 and information contained in it may not therefore be current - always check the MOT Manual if in doubt about any aspect of the MOT Test.

Is the MOT brake performance Test good enough?

Does roller brake test equipment give inaccurate results?

In the preceding section we had a look at brake testing and concluded that in the interest of road safety a comprehensive review of examining a vehicle’s brakes should be initiated by VOSA. The other side of the coin, of course, is the equipment used to test brake performance, roller brake test equipment. We assume that the equipment we use accurately measures braking performance – but does it?

It is an interesting question. Does roller brake test equipment provide an accurate assessment of how effective and efficient the vehicle’s brakes will be when driven on the road? We see the wheels ‘lock up’ in the rollers and we assume that means all is well – but is it?

“No”, is the short and accurate answer. “It isn’t!” As we shall see, there are a number of reasons for this.

A ‘static’ test

When we test brakes in the MOT using roller brake test equipment, it is a static test. We cannot measure any dynamic effects. We also cannot measure any problems in the braking system should the time between applying the brake and when the pads and/or brake shoes ‘bite’, become too long – too much ‘lag’ in the system. But does this matter? What is the difference between carrying out ‘on the road’ accurate decelerometer checks and then testing the same brakes in the workshop? Well, it has been done.

A controlled experiment

In1999 an experiment comparing different methods of measuring brake performance was carried out in Australia by a company called Vipac.

First, they accurately measured vehicle brake performance using an electronic decelerometer during road tests on a dry road. This they did with a number of different vehicles. They then checked the brake performance of the vehicles using both roller brake test equipment and plate brake test equipment.  

The results were surprising. Roller brake test equipment overestimates braking performance by between 13% up to an astonishing 51%. The plate brake test equipment fared even worse with between an 18% and 67% overestimate of performance.

Take one example: An unladen SAAB 9000CS had an actual ‘on-the-road’ deceleration of 0.58G. Whereas the roller brake test equipment recorded 0.67, and the plate brake tester 0.85.

A pragmatic approach

In practice, of course, what we are really measuring in our MOT Testing bays is not braking ‘performance’, but braking ‘operation’. We are measuring the effect of the application of the shoes or pads to the drums and discs. We measure any binding and how the braking force progressively increases and decreases, the maximum force applied, together with a comparison ‘side-to-side’ to measure ‘imbalance’.

It is a pragmatic approach to do as well as we can with the kind of technology available when roller brake test equipment first became mandatory decades ago.

Of course, we shouldn’t ‘knock’ the current brake performance test too much. It is a pragmatic approach. It does detect significant malfunction. However we should also be aware of its limitations. It does not fully reflect how a vehicle’s brakes will perform when being driven on the road.

Low - tech equipment

A more bizarre aspect of the current ‘brake performance’ test is VOSA’s not having taken on board the advent of modern computerised technology.

Available roller brake test equipment is, these days, very sophisticated. Using load sensors, vehicle weight can be accurately measured and braking ‘performance’ calculated by computer with pass/fail displayed for binding, efficiency, ovality and imbalance. But even if we installed such equipment we would have to ignore the computer, laboriously look up the vehicle weight and carry out a calculation – just as we did decades ago.

Is there any other industry in the country compelled by Government to continue to use technology not years, but decades out of date?

VOSA’s Alan Wilson, Head of Testing Standards Policy and Strategy, gave us his response:

What actually happens when as drivers we spot an emergency and hit the brakes – HARD? For the first third to half a second nothing happens while our brains process the data to reach a decision. Then a further time elapses while our right foot moves off the throttle to the brake pedal. Even more time passes as the pedal is depressed until it reaches the point where the brake fluid is pressurised. Then, as hydraulic brake pressure increases the first evidence of braking is when the brake lights go on. Even a Grand Prix F1 driver would need almost half a second, maybe more to get this far – but there’s still time to go before any braking happens…

Key: A. Coast down drag: 0.05G deceleration brake lights on, brakes not yet grabbing. B. Vehicle mechanical reaction time  The time from brake lights on to when the brakes start working at 0.2G deceleration. C. Vehicle mechanical response time  The time from brake lights on to when the tyres reach full braking force. D. Static Friction  Spin velocity of the tyre is reduced by 20% and maximum friction for braking occurs. E. Dynamic Friction ­ The tyres are now locked (standard brakes, no ABS) and sliding friction occurs (skidding). F. Stopped  no speed.

This graph is reproduced from a 1999 paper entitled “How much braking performance is enough?” By Michael Smith, Director of Vipac Engineers and Scientists Limited ­ Melbourne Australia.

When time ‘stands still’

Have a look at the graph of deceleration against time. From time zero taken as when the brake lights go on, the throttle is at idle, the car slowing down. Yet before the brakes take effect at point B on the graph, a fifth of a second has passed. It isn’t until point C, after just under a third of a second, that the tyres start to deliver a significant braking force. That’s the best part of a second since spotting the danger, before anything significant happens!

Not until point D, with tyre rotation reduced by 20% does maximum braking friction occur. This is called ‘static friction’, when the greatest ‘slowing down’ or ‘g’ force is experienced. This continues until wheels ‘lock up’ at E, and (without ABS), the vehicle skids. This is called ‘dynamic friction’ which lasts from E to F, when the vehicle has stopped.

Different vehicles in different condition

The point to note, however, is that this graph will be much the same overall shape for all cars, but will differ in detail for different vehicles. Even different vehicles of the same make. It depends on the mechanical condition of the brakes, the type of tyres, the condition of the tyres, the load on the vehicle and, a vital factor, the vehicle’s braking dynamics – very much a matter of suspension design, and braking distribution front to rear. For example, does the car ‘nose-dive’ under braking (suspension design)? Are the shock absorbers in good condition? Is there a device to change the braking distribution front-to-rear when the rear wheels are loaded? And so on.

Does it matter?

“Yes! it does matter.” For example, a long pedal travel, which still has sufficient reserve travel could take a tenth of a second longer to apply. In that time at 60mph braking distance is increased by nearly three metres – in some cases the difference between life and death! And other research has shown that shock absorber condition is another key factor affecting braking.

And none of this takes account of the fact that we could quite legitimately and unknowingly carry out a brake performance check with incorrectly inflated tyres – hence affecting brake performance.

A better test?

The truth is that the current method of checking brake performance for the MOT is decades old and whilst sufficient for a comparative test car for car, provides only a crude and inaccurate measure of how the vehicle’s brakes will perform in dynamic ‘on-the-road’ conditions.

Arguably then VOSA should be looking into alternatives to ensure that the MOT brake performance test better reflects actual driving conditions. In practice, however, this would need an accurate electronic decelerometer, which prints out brake performance with pass/fail depending on the shape and time scales on a deceleration graph timed from when the driver first touches a sensor attached to the brake pedal. Although brake testing 24 million vehicles on the road each year may have adverse political implications!

Yet simply fitting a sensor on the brake pedal linked to a roller brake test machine to measure the brake ‘lag’ would be a step in the right direction…

There is little doubt, that just as a more comprehensive inspection of the brakes would improve the MOT Test, so too would a fresh look at how we check brake performance.