Performance Tuner Magazine Article TECH : Dyno Discrepancies Transcribed because of image quality

Subtitle : Think of a Number

Performance is a numbers game. But just how reliable are those numbers when they’re produced on a rolling-road? We take one to five different chassis dynamometers on the same day to answer that question.

Story and photos : Peter Knivett

Rolling-road thrashes : don’t you just love them? Gathering a group of similar machines together at some willing rolling road always makes for pretty interesting reading – and particularly the owner’s reactions after what can all too often be a disappointing power run.

The textbook line is usually ‘I’m gutted, I thought it would make over 300bhp,’ when their pride and joy has actually made just 187 bhp. On the face of it, who’d argue with this disappointed punter’s line of thinking? After all, they’ve probably spent thousands of pounds on all the right tuning goodies, plus they ‘had a mate on an internet forum who’s told me it must be really quick, because his brother said so, and he worked at Prodrive once, and he knows what he’s talking about…’

Month in, month out, you see this sort of tale repeated across all sectors of the performance tuning market, and it never fails to amuse, because it’s a classic case of putting the component-expenditure cart before the testing horse. Despite melting his credit card by splashing out on tuning goodies our disappointed punter forgot to spend just 50-75 pounds having his car tested on a rolling-road, which could well have made a crucial difference to its ultimate performance. We know, because we’ve seen the benefits it can have on many occasions. Without rolling-road testing all power claims are simply guesses, some of them not particularly educated ones at that, because real-world testing always produces surprises, some less welcome than others.

Then there’s the other story that you frequently read from these rolling- road thrashes, usually with an owner saying, ‘I’m really pleased, it made 20bhp more than XXX’s rollers last month.’ He sounds pleased, but he shouldn’t be, particularly if he hasn’t tweaked the car since his last rolling-road run, because as we’ll show you, comparing the power figures from two different chassis dynos is a futile exercise that generally leads only to confusion and ultimately disappointment. Given that Performance Tuner is still a new title, we’ve the opportunity to advise you on how not only to avoid the mistakes others make, but also to arm you with some valuable information that will set you ahead of the opposition. Measuring power outputs and increases correctly is a classic case in point. How do you go about it, when should you do it, and what should you be looking for? Read on and we’ll be delighted to enlighten you.

WHAT IS BHP AND TORQUE?

First of all, what are we actually measuring on a rolling road? It’s surprising just how few people know how properly to define what brake horsepower or torque is, partly because there’s some maths required. But hold on, before you flick the page, it’s pretty simple stuff that’s worth knowing.

To understand what brake horsepower actually is, first you need to establish what torque is. Torque is the measurement of how much leverage, or turning force, an engine produces at the point where the work is really needed, namely the radius between the centre line of the motor’s crankshaft, rotates the flywheel, and ultimately provides the force that drives the car. Torque is measured in terms of pounds per foot, or the metric equivalent, Newton metres, and the amount of torque an engine makes is pretty much governed by two main factors; it’s internal design and how efficiently it breathes.

TORQUE – that’s the goal

Altering the internal layout of an engine isn’t easy, so the simplest way of improving a motor’s torque output is to increase how much fuel/air mixture it burns. At its simplest, that’s what just about every single tuning part sold on the market today aims to do – by increasing the air flow through the motor and adding the correct amount of fuel, you’ll raise the torque output.

Trouble is, torque is viewed by most car nuts as power’s boring cousin, which is wide of the mark, because it’s really more important than power. It’s torque – not power – that accelerates your car down the road. Don’t believe me? Well, chew on this, in 10 years of testing tuned car this writer can say that the best machines I’ve driven have always had almost as much torque as power – if not more. This is also the reason that turbodiesel general have such punchy in-gear acceleration times.

Still doubtful? The 2.0-litre turbo motors used in cars produce an often quoted 320bhp, but they also produce around 475 ft/lb of torque, because rallying is all about acceleration from low engine speeds between corners. Crucially, though, maximising torque is extremely important because power – brake horsepower – is a by product of torque WHAT IS BRAKE HORSEPOWER?

When people talk about an engine ‘power’ they’re referring to how much brake horsepower the motor produces. This bhp is a unit of measurement that was invented in the 19th century to grade the power output of steam engines. And before anyone had cars, it made sense to compare the power of the new engines to the amount of work that a horse could do in one minute, which was calculated to be 33,000 lb/ft of torque – a combination of force, distance and time. Now for the maths bit. Because torque is ultimately measured at the engine’s flywheel, we need to calculate the circumference of the flywheel to work out the distance the flywheel has turned using good old pi – but here’s the simplified, cut-out-and keep equation:

BRAKE HORSEPOWER = TORQUE x ENGINE SPEED IN RPM

Clearly, engine speed has a large effect on bhp output, which means that increasing revolutions per minute is a surefire way of increasing power. Is that why a Honda Civic Type R produces 195bhp from 2.0L and revs to 8500 rpm? Could be ... but if that’s left you completely baffled, here’s a simple description in abstract terms – thick of bhp as the measure of how much work an engine can ultimate do, and torque as a measure of how quickly it can do it. Keep that in mind and you won’t go far wrong.

HOW ARE BHP AND TORQUE MEASURED ON A ROLLING ROAD?

A rolling-road (or chassis dynamometer) is a piece of equipment that enables an engine’s brake horsepower and torque to be measured under load, without having to go to the trouble of removing the power unit from the car. Chassis dynos use rollers to act as the interface between the tyres and the measuring apparatus, although these days so-called ‘hub’ dynos are also becoming popular where the measuring equipment is bolted directly to the driven axle of the vehicle.

The engine’s power output is measured by calculating the force that it can generate against the rollers (or hub pickups), which are loaded to create a resistance by one of two methods. The most common is the eddy-current dyno, which uses opposing electromagnets to provide a variable roller loading, but an alternative is an inertia dyno, which simply uses the mass (in other words, the weight) of high-inertia drums to provide a consistent, measureable resistance.

WHEEL, HUB OR FLYWHEEL FIGURES?

Arguments have long raged about whether it’s best to measure the power figures from either the flywheel or the road wheel, and now we have hub dynos muddying the waters too. The problem arises because precious horses are shed between the engine’s flywheel and the road wheels, thanks to the power-sapping nature of a car’s transmission, it’s also compounded by the convention that the majority of manufactures (perhaps not surprisingly) publish their power figures as measured from the flywheel, and so thanks to the transmission losses wheel outputs tend to sound a touch weedy in comparison.

Sadly, short of stripping the engine ou7t of the chassis and running it on a bench-test dyno there isn’t a definitive answer to easily calculating transmission losses. The latest eddy current dynos use a ‘coast down’ process that seems fairly accurately to calculate losses, but this process isn’t uniformly reliable – some makes of dyno I’ve used have produced some very odd flywheel-loss numbers. Given that, I’d always rely on power figures measured at the wheels, or hubs, because they’re not open to debate or doubt indeed, this is standard practice among US tuners. As a rough guide a rear wheel drive car will lose 10-15%, a front wheel drive car will lose 15-20 per cent, and a four wheel drive car can lose up to 25% in transmission losses, but the variables involved (gear design, transmission oil qualities and so on) are vast, so these sums aren’t an exact science, hence why the use of wheel figures is preferable every time.

CORRECTION FACTORS

The internal combustion engine is sensitive to variations in air pressure and ambient temperature, so these will alter the power outputs from one day, or even hour, to the next. For example, cooler ambient temperatures being greater air densities which benefit power, while higher barometric pressures mean there are more molecules of air around for the engine to draw into tis intake. Clearly, if we took a power reading during these conditions, then ran the car again on a hot day with lower barometric pressure it would tend to produce less power – in reality it doesn’t, it’s just that the conditions have changed.

As you can see, if these external factors are left uncorrected, then inconsistencies creep into the test, so these must be corrected out. Unfortunately, there’s more than one method of correction, even though each system incorporates air-pressure readings and ambient temperatures to help correct the power outputs. In the UK, the most popular correction factors used are SAE, DIN, EEC, STD and JIS - of these, SAE and DIN are the only two worth bothering about. If you want consistent results, ensure that your car is testing using either SAE or DIN at all times. Also, be sure that the dyno operator inputs the ambient temperatures and barometric pressures before literally each and every run, in order to ensure accuracy, because it’s quite common for the weather conditions to change quite markedly during a test, particularly if the car has a prolonged dyno session.

GETTING IT RIGHT – BEFORE YOU GET TO THE ROLLERS

If you’re using a dyno, there are several areas that should be checked before your machine ventures onto the rollers. Obviously it’s pointless chasing every last bhp on the dyno if your car is overdue a service – neglected cars won’t produce the right numbers. Even if it’s been recently serviced ensure that the vital fluids – oil and coolant – are at the right levels. Be certain that the petrol you’ve filled up with represents what you’ll normally be running, because it’s pointless booking in a session on the rollers to have the car remapped to 98 octane, then filling it up with cheap 95 Octane juice from the Supermarket. Lastly, check your tyre pressures, because under-inflated tires affect the power output. To prove this, we subtracted 100psi from the tires on our test Ka Sport and lost over 2bhp on rollers. Not much admittedly, but then it only had 94 bhp to start with.

GETTING IT RIGHT – ON THE ROLLERS

For a first timer a rolling-road session can be quite nervewracking, so here’s what to expect. To begin with, a decent dyno operator should check the car’s fluids and tyre pressures as described above. Then they’ll load your car onto the dyno, which entails driving it onto the rollers, which are sometimes lowered into a ‘run’ position. Then the car is strapped down at both the front and the rear. Some operators, such as Guy Chamberlain at CPL Racing also use a ‘torque bar’ which helps keep the driven wheels aligned squarely on the rollers, an important factor with high-powered cars.

Next a probe is inserted into the exhaust system to read the AFR – that’s the air/fuel ratio, one of the critical factors in helping an engine produce power, if the engine’s excessively lean or too rich the AFR meter will flag this up, and if so the operator may have to abort the run until adjustments are made.

From here an ignition pickup is often used to take an engine speed trace to help the rolling-road’s computer calibrate roller speed versus rpm, thus ensuring the accuracy of the results. Alternatively, the newer rolling-roads can calibrate real-time roller speed against the car’s speedometer with good levels of accuracy.

Some rolling-roads also feature an inlet-air temperature sensor, which needs to be positions within the engine’s inlet trunking, upstream of the throttle body. It’s not vital, but again it’s another stream of data that can help build a coherent picture of how a given motor is performing under load.

Finally, but crucially, the rolling-road cooling fan is placed at the nose of the car. Different rollers use a wide variety of fan sized, from small to huge, and a fan must be used during power testing, since this will prevent a temperature build up under the bonnet. Expect plenty of noise at the moment it is switched on.

That done, once the oil and coolant are up to operating temperatures the operator will start to accelerate the car up through the gears, typically into third or fourth, where they will commence the power run. Once there they will load up the rollers and floor the throttle from 1500rpm or so up to rev limiter. Occasionally an additional coast down run may be required to calculate the transmission losses back to a flywheel figure, or this may take place on the wind-down from the power run itself. Generally I’d advise taking three runs to be certain of consistency, and once done, you’ll be presented with a handful of printouts to pore over. Be aware that some dynos produce their print outs in metric kilowatt units, which will require conversion back to imperial bhp.

CONCLUSION

With a mixture of correction factors, and wheel and flywheel figures, this exercise highlighted the pointless nature of comparing your car on two different rolling roads. Because Janspeed’s inertia rolling road produces wheel figures only we’d use those as the direct comparison between outfits, but even so a 13.4 bhp difference between the highest and lowest wheel bhp readings is pretty shocking. Of them all, Surrey Rolling Road’s Dyno Dynamics rig got closest to the mark with a near spot-on set of figures, with CPL Racing’s identical dyno following close behind. Even accounting for the different correction – factor methods the other three rolling roads over-read by around 12 per cent, and this is on a car with less than 100 bhp. Imagine the over read figure if we had a genuine 300bhp car.

So the underlying message is very simple – to know your car’s power output you must test it on a rolling-road. That applies even if your machine is totally standard, because your first run will give you a baseline figure that’ll act as a reference point from which you can measure the effectiveness of future upgrades, plus the operator should detect if there’s anything wrong with the engine before you’ve started tweaking it. Even standard cars can sometimes have undetected power sapping problems – I once saw an MG ZR lose a good 20bhp purely because of an incorrectly assembled factory throttle body, so it does happen.

Then, after every motor upgrade, no matter how small, return to the same rolling road, use the same correction factor, run the same petrol and tyre pressures, and make sure that the operator conducts the power test in the same gear and then you’ll see what difference each tweak has made to the engine’s power output and fuelling under load. Outright power and torque figures are irrelevant (other than for pub or forum talk) because what matters is the level of increase over and above the standard numbers you originally recorded at the start of the process.

Bearing that in mind, our results don’t mean that any of the dynos tested here should be avoided, because they are all useful tools, particularly for before- and-after-style component testing. Just bear in mind that some fo the outright power figures should be taken with a pinch of salt. Ideally, though, try and use a consistent, repeatable chassis dynamometer, stock to wheel bhp output figures only and you won’t go too far wrong – plus you’ll know for certain whether your tuning efforts have been successful. Or not, as the case may be!

RESULTS :

TESTING TESTING

To illustrate the dangers of comparing one rolling-road directly with another, Performance Tuner took a totally standard Ford Ka Sport to five different rolling roads on the same day. Each test was carried out with the same fuel (Tesco 95 unleaded), while tire pressures were carefully checked to ensure consistency. Ford claims that the Sport Ka produces 94 bhp and 100lb/ft at the flywheel, which should equate to around 74-78bhp and roughly 80lb/ft of torque at the wheels. Correction factors were left in the hands of the individual dyno operators, so 457 miles and 14 hours later, what had we found out? See below!

Janspeed Engineering Ltd Rolling road model : Dynojet Inertia Wheel power : 86.17 bhp Wheel torque : 95.36 lb/ft Flywheel power : Not available Flywheel torque : not available Correction method : SAE

Brooklands Horse Power (BHP) Rolling-road model : Sun Ram 3000 Wheel power : 85.9 bhp Wheel torque : not available Flywheel power : 104.86 bhp Flywheel torque : 104.17 lb/ft Correction method : DIN 700200

Surrey Rolling Road Rolling Road model : Dyno Dynamics 450 Wheel Power : 76.1 bhp Wheel torque : 82.5 lb/ft Flywheel power : 95.0 bhp Flywheel torque : 103.2 lb/ft Correction method : SAE

Gerald Dale Motorsport Rolling road model : Sun Road-A-Matic XJ Wheel power : 89.5 bhp Wheel torque : not available Flywheel power : 106.2 bhp Flywheel torque : 109.4 lb/ft Correction method : DIN 700200

CPL Racing Rolling Road model : Dyno Dynamics 450 Wheel Power : 78.6 bhp Wheel torque : 85.11 lb/ft Flywheel power : 98.7 bhp Flywheel torque : 108.0 lb/ft Correction method : SAE