POWER OUTPUT, TUNING & PERFORMANCE TESTING.

WITH PARTICULAR REFERENCE TO THE PORSCHE 968

By Barry Hart
Joint Founder of Hartech, Bolton

INTRODUCTION

After many years involvement with the design, manufacture and development of racing motorcycle and car engines and gearboxes, I found myself testing many variables affecting engine power and performance. In the process I have proven certain facts; disproved many other theories and learned a great deal (often from trial and error). This article defines the important factors that influence power production and measurement that helped me achieve some success on the racetrack. I wrote this article in response to the many questions I get from 968 owners and to try and clear up some of the arguments on the web site that can handicap others from understanding what is really going on by allowing too many discussions to center around incorrect assumptions, issues and theories - wasting time and effort - while it takes too long to re-emphasise the right issues every time a response is made to a specific question. Web space and time have forced this to be a very short and quick fix article on a very complex subject (indeed my first attempt was many times the length and will soon be available on our own web site www.hartech.org.uk for those wanting to know more). I am indebted to Paul Follet for picking out the salient points and helping to re-write it in its present condensed form - thanks.

BASICS.

Piston engines are basically air pumps. They receive air in (often thought of as "sucking" in), mix it with fuel, compress it, ignite it and the resulting pressure pushes down on the piston and con-rod through the crank throw and angle to produce torque twisting the crankshaft (very like a human arm pushing down on a wrench fitted to the crankshaft) - finally pushing the exhaust waste products out. This is often called "suck, squeeze, bang, blow) The greater the push down on the piston - the greater the resulting torque and the faster this occurs the more power can theoretically be produced (as power is a function of torque and revs). Conversely - the higher the revs the shorter time is available to get that air and exhaust gas in and out - so eventually a "breathing" limit is reached. Cooler, denser air contains more oxygen so for a given pressure drop, more "air" is effectively taken in and trapped if it is cooler (hence intercoolers and external inlets) and it can be compressed higher before detonation (producing more power).

"Sucking air in" is actually a misleading concept because air only moves into the engine because the pressure inside is lower than outside (the pressure drop caused by the piston's downward movement). It could therefore be more appropriate to imagine the air being "blown" in by the higher pressure outside than being "sucked" in by the piston in some way. This is important because it changes the understanding of what is going on and somehow makes you appreciate that the forces that are helping the air to go into the engine are not actually very big as the greatest negative pressure possible inside the engine is zero and the maximum outside is atmospheric pressure at around 15psi - the maximum normally aspirated pressure difference. Imagine a tyre with 15 p.s.i. in it, opening an inlet valve in the side for 4 thousandths of a second and picture how little air would escape and a more tangible understanding of the problem emerges.

Generally the greatest volume of air you can theoretically get into a cylinder is therefore its own capacity. Some engines today can actually exceed this slightly in one small rev band due to clever camshaft timing and inlet tracts that enable the inertia of a column of air - already moving towards the cylinder - to carry on flowing in after the pressure has equalized. Furthermore - initially when the inlet valve opens - the piston has hardly started lowering the pressure inside - so flow then would be minimized - however by allowing some overlap between the exhaust valve closing and the inlet opening - the last flow of the outgoing exhaust gas can help lower cylinder pressure on overlap to encourage the flow of new air into the cylinder.

You can imagine Turbochargers and Superchargers creating greater than atmospheric pressure outside (and some linear flow) and intercoolers lowering the inlet temperature - both to help more air get in. To trap more air with naturally aspirated engines at high revs it would be an advantage to keep the inlet valves open longer & or make the lift greater & or shorten the induction length. If these settings were used at lower revs, air would spill back out and reduce the performance and drivability.

Most traditional engines had fixed timings and lifts (independent of revs) so were designed for a general power band. For competition engines it was then not too difficult to alter some things to create more power in a smaller selective power band. How much you could increase power in a smaller rev band was limited by the amount of gears useable and the amount the revs dropped between gear changes. Therefore engines with lots of gears and variable breathing devices can be so well tuned from the factory as to leave little else possible to improve them even for competition use. Hence the suitability for track use of the standard 968 engine with Variocam and a 6-speed gearbox.

Many indulge in promoting their own pet theory about what makes an engine more powerful - but - there really is only one issue - the cylinder pressure reached before ignition. All other major factors achieve better pressures (or try to). Assuming mixture and ignition timing are right then as long as the pressure and temperature in the cylinder are not so high as to cause detonation, the amount of power that can be produced depends almost totally upon the compressed pressure reached before ignition. This is increased by trapping more air (with ideal flow and dynamics), pumping in more air (using compressors), raising the compression ratio (or making sure that it doesn't leak out again under compression), having good valve seating and or rings and piston fit and lowering the charge temperature. The greater the trapped pressures (within these limits) the greater the resulting burn pressure, the force on the piston and therefore the torque and power produced. By further explanation - when you lift off the throttle and say drive on half throttle - the reduction in power is actually because you are limiting the amount of air that can pass into the engine and therefore by trapping less air it is compressed to a much lower pressure before ignition giving less power output.

SO WHAT COULD YOU DO TO TUNE A 968 ?

1. THE ECU CHIP
For road use, legislation restricts fuel ratios and ignition timings to operate below their most powerful settings. So a chip that richens the mixture (and adjusts the ignition timing accordingly) should produce a little more burn pressure (b.m.e.p.) while increasing un-burnt emissions. The chip could also be optimised for use only with a top grade of fuel such as super unleaded (thus removing some of the compromises within the standard arrangement).

Many chips lift the maximum revs permitted and this may feel faster (and allow a higher bhp) but may not actually be quicker. There are 2 things against - (1) torque drops as revs rise towards the breathing limit and it is the torque that accelerates the car (by the formula acceleration = torque/resistance). Since power is a function of torque x revs - the bhp may be higher but the torque may be less and may not result in a faster car. (2) - The rev drop while changing gear is a proportion of the ratio multiplied by the revs so if you do not change the ratios the rev drop between changes - increases as the revs rise (see later).

2. CAMSHAFT TIMING

Because of the Porsche Variocam system there is already a variable camshaft-timing device in operation. Therefore the potential in this area is not as significant as usual. Although it should be possible to make a camshaft that breathes better for track use it would then be worse for road use. Standard timing does vary as the cam belt and chain stretch and needs regular re-adjustment, this is exacerbated by wear on the camshaft belt drive sprockets that also promotes rapid belt wear and stretch. The automatic tensioner often reaches its maximum limit while the cars are still being used - thus losing tension and changing basic timing (although we manufacture something to minimize this problem).

3. INERTIA

Other basics ways to make the car faster involve reducing the mass (static weight of the parts) of the engine or the car to improve acceleration. Lighter flywheels, transmission parts (or even the driver) will always accelerate more quickly.

4. GAS FLOWING

Years ago a lot of time was spent polishing inlet ports and knife edging bridges etc - however the flow of inlet gas is sub-sonic so it actually has little effect. More likely is that the inevitable increase in area resulting from removal of material - suited better gas flow at higher revs (probably at the expense of a reduction at the now unused bottom end). Indeed there is some value in a slightly rough surface (not a glazed finish) enabling a film of slower gas to help the laminar air-flow go past it efficiently. A better design of inlet system could benefit any car but will be expensive and need a chip change to optimize results.

5. EXHAUST SYSTEMS.

Similarly racing exhausts have a different role to that for the road in which quietness is more of a feature. Generally as a gas expands the pressure drops, so a carefully designed pipe can help lower cylinder pressure and help evacuate the cylinder and promote the incoming charge efficiency. Also some clever pipe lengths and joints can enable the exhaust from one cylinder to assist the evacuation of another - but this is subject to considerable testing and alteration and may be too time consuming for anyone to bother in view of the comparatively few potential customers involved.

6. GEAR RATIOS

Once again a basic principle is important. People imagine that the ratios in a gearbox fixes the revs that drop as you change from one gear to another - but it doesn't - the revs only drop by the proportion of the revs that the engine was at when the change was made. If a gearbox drops 1000 revs when changing up @ 6000 revs in 4th say, it will drop only 500 revs in 4th if the change is made @ 3000 revs. You see if you consider an engine revving to 6250 revs/minute and dropping to say 5000 revs/minute when you change gear from say 1st to 2nd (a drop of 1250 revs/minute) - the rev drop is 20%. If you improve the revs to 6,800 revs/minute and change gear - the actual drop is still 20% - but 20% of 6,800 r.p.m. is now 1,360 revs/minute. This means that if you raise the maximum revs of the engine - then to make best use of the gearbox - you would need to not only raise the peak revs but also broaden the power band (usually almost impossible). It also means that changes in the revs of different engines or the same engine being developed could only be properly exploited if the gearing was also altered accordingly.

However assuming that the engine revs are not being altered there is still a good reason to change gear ratios for track use. Lets trace the gear ratios of a 968 and see what I mean (refer to your driver's manual graph "transmission diagram"). You will see that first gear can achieve about 40 m.p.h and 5th gear about 135 m.p.h. You will rarely on the UK circuits that I know - be able to use first gear - you will never be going that slowly and you will rarely be finding a long enough straight to exceed 135 (often much less) - so you will effectively - at most be using a 4 speed gearbox (often only 3 speeds).

Now if you draw a vertical line from peak revs in each gear - where it crosses the next sloping line is the engine revs you will drop to when changing up a gear. In 2nd to third this is about 1500 revs and from 4th to fifth about 1,200 revs - so allowing for building revs through a corner (as you couldn't "drive" the car into a corner while flat out in a lower gear) - you only need a powerful engine between say 4,500 revs and top revs (say 6,400) to negotiate a circuit quickly.

Next - look at the torque and power graphs at the same revs and you will see that you get maximum torque almost level between those revs already and a steadily improving bhp line (as the speed increases). Now a measure of the work done is the area under the torque graph - hence - unless you can increase it all (and because any timing changes that raise the graph at one end will almost certainly lower it at the other - achieving the same or less area)- it is already theoretically perfect unless the whole graph can be raised throughout its length. In practice - it may well be that a little more slope upwards @ 4500 revs and downwards towards then end of the torque graph would drive round a circuit more quickly but for most practical purposes it is already an ideal vehicle for track use (and hence it's huge popularity).

Although it may be disappointing that you can use so few of the 6 speeds on the track, many 5 speed Porsche's can only use 3 speeds for track use - which forces the rev range over which they have to be driven to be much greater and since most do not have variable engine dynamic systems - it is a very difficult compromise to tune them over such a wide power band and they consequently perform a lot slower.

Because the 968 already has 6 speeds (and assuming no one is going to the expense of changing the ratios) then already as a stock item it seems almost tuned to perfection theoretically as at least 4 gears are probably in use. They would always circulate faster with 6 closer ratios (all being used) in the box (and enable more extreme tuning) but such a car would be difficult in road traffic as first gear would be much higher and uneconomical as 6th gear would be lower for most racing circuits.

You may wonder why the rev drop is not the same in all gear changes - getting less as you change up the box. This is because the resistance to motion increases with the speed so you can get away with more slightly outside the optimum power band at lower speeds but need to gradually approach the rev band of the engine that performs best as the speed and therefore resistance rise. Furthermore if you used that final power band for all gear changes then you would need about 8 speeds (as drawing a suitable graph will quickly demonstrate). If race tracks had got longer as performance has increased over the years - then top speeds would be higher and rev drops would need to be even less to cope with the extra resistance at higher speeds - but fortunately the tracks are the same and therefore effectively have got relatively shorter - so most road gear boxes can still cope.

7. BURN PRESSURE

The most important quest for power is to trap and compress more air in the cylinders. Regardless of everything else - whatever is done to improve that - is wasted if it leaks out again before ignition. Most 968's quickly lose ideal valve seating as a result of the long spindly valve stems and valve head flutter and this can lose around 10 to 15 bhp - so simply overhauling this area can achieve significant improvements and can easily be justified since most head gaskets last 10 to 12 years and will need doing soon anyway.

With a naturally aspirated engine - everything you do to make it breath in and trap more air at high revs (large inlet ports, large valves, late camshaft timing etc) works against you to efficiently trap the maximum air at lower revs, because the air going into the engine has both mass and resistance. At any engine speed the right sized inlet ports help the inertia of the incoming gas to allow it to continue rushing into the cylinder even when the negative pressure in the cylinder is minimal - but different sizes are needed at different mass flow rates (or engine speeds/throttle openings etc).

Big port holes and valves are needed to reduce resistance at high air flow rates (or revs) to produce maximum power - but at low revs (when the pressure drop is not so high) - big ports make the gas stream too slow to get the ram effect (and are open for too long allowing the charge to escape back out again). So the power band choice has to be made at the design stage unless the valve sizes, timing, valve lift or inlet port sizes can all be varied with engine revs and load. For road use - you need both torque at low revs (to make it drivable) and power at high revs (to enjoy it's performance). Porsche's variable inlet timing systems (although mechanically crude) help achieve this by optimizing inlet timing in both ranges.

If achieving a high compressed pressure is the ultimate aim then higher compression pistons will always be a quick fix but then anything that makes a 968 produce more torque must by the same rule produce more burn pressure and consequently head gaskets will inevitably become vulnerable - possibly needing upgrading to cope.

CHANGING DEMANDS FROM OWNERS

Although the 968 with its 6 speeds and variable camshaft timing already improves the range over which good torque figures can be produced - many owners are beginning to seek different performance characteristics for track use. America is usually 10 years or so ahead of the UK and there are a lot of Americans who can afford to own a Porsche not only as a third car but also only for track day use. This seems sensible in view of increasing road safety legislation - speed cameras etc. Furthermore once you have experienced the thrill of driving quickly on the track - you find it is so impossible to reproduce anything like that sensation - even illegally on the road - and soon you find that you confine your need for speed to track use only.

This is starting to become affordable and more sociably acceptable in the UK. So more owners will be seeking advice on how to improve their track performance. The internet is helping to foster these developments as more technical information and diverse opinions are being contributed to web based forums.

Now the suspension, gear ratios and engine tuning that will be best on the track will not be the same as for the road - so owners will have to decide which area is more important to them and how far they want to go. Because the U.S.A. has been more advanced in this area - I find they seem to have more suppliers of this type of information and products etc than we do at present. However the 968 must be one of the best cars for track use that can also be driven to the track and costs comparatively little to provide an exciting and satisfying drive from standard.

PERFORMANCE TESTING

Once owners seek improvements they become potential victims of misleading publicity, technical claims, unscrupulous suppliers of goods and services or just the inability to assess and convey the results of various changes and products. The 968 web forums provide invaluable experiences and advice but even within that a lot is actually incorrect or misleading.

The best way to minimize these problems is for contributors to provide reliable information that helps others and the method they choose of performance testing is crucial to the usefulness of this exercise.

All my extensive experience has shown me that it can be very misleading to test an engine in any way other than by measuring the vehicle's acceleration as a whole in true road conditions - i.e. while driving it.

I do not recommend the use of a static rolling road or drum type dynamometers, let me explain why. Engine breathing is not stable as the revs rise and the rate of acceleration changes depending on how quickly an engine stabilises its gas flow. So unless the acceleration curve or resistance reproduces exactly what happens on a road - the results will always be misleading (and by a greater margin than anyone who hasn't been able to test these variations would ever believe possible).

Similarly the temperature of all the engine parts (including the exhaust system) does change the breathing characteristics and can only be reproduced accurately in a real road situation. Additionally, the fuel pressure alters the fuel flow and is influenced by the dynamic motion of the car. Rolling roads do not reproduce the same airflow and the roller diameters and loads needed to transmit the energy are not the same as on the road, often suffering creep and tyre temperature problems. Furthermore the resistance characteristics are unlikely to exactly match road acceleration.

The fastest way to travel over a distance between braking is to increase acceleration and the amount of work done is theoretically a function of the area under a torque graph between the revs in the gears and this will reveal the best tuning modifications. A sudden burst of torque feels great but inevitably reduces the overall area under the graph and this can only be measured dispassionately by a device of some sort.

A driver's ability to assess the value of tuning changes is flawed because the ability of the human brain to accurately compute information is limited by its response to sensation. In the same way that flying at 450 mph on a plane feels like standing still, a constant rate of acceleration does not feel impressive to the senses. Conversely a sudden surge (like with a turbo or hot camshaft), a loud exhaust or higher revs can feel more impressive but actually reflect a slower run. Ironically with the physical limit of acceleration being the ability of the tyres to grip the road - and with the best possible "grip limit" being an almost constant rear wheel torque throughout the rev range - actually the slowest feeling acceleration might also be the fastest through the gears if it is generated by constant torque.

Although the torque graph theoretically demonstrates the fastest potential characteristics. I have found that the inertia of the engine, the size and frequency of the power pulses into the engine "system", the mass of the various moving parts and the strain energy lost within the whole transmission system - can produce completely unexpected results. It means that you cannot deduce exactly how an engine will perform in a vehicle in motion unless it is in that vehicle measured in motion. You can get clues and yes - if you squeeze more torque out of an engine it should produce more acceleration but the rate of acceleration will only be a reliable guide to eventual performance if the rate of acceleration of the test is the same and still other dynamic factors can confuse things. So, an engine improvement tested at the flywheel will not necessarily transmit the same torque improvement to the wheels under actual acceleration.

Similarly adjustments made to chip parameters while an engine is running at constant load and revs on an engine or roller dyno - will not necessarily produce the fastest acceleration when submitted to a changing rate of acceleration and load in road use.

This is probably the most important factor to understand. Generally a smoother torque delivery loses least strain energy. Equally while the torque is dropping off at higher revs, the strain energy is also reducing while the number of power pulse delivered to the vehicle is increasing and this can have a benefit in terms of acceleration that theory does not explain. It means that usually the best overall characteristics can be a relatively flat torque curve that drops off towards peak revs - while the bhp is still rising or level. Those of you who have seen the standard curves of a 968 engine will notice the similarity to what I describe.

It is actually impossible for mortals to predict exact perfect characteristics for any vehicle and only comparative testing of the actual car while measuring the results will confirm the very best set up. Early 16 valve engines on early Gti's and indeed the Porsche 944S (from which the 968 is derived) demonstrate this anomaly well. They all claimed higher bhp than their 8 valve predecessors - but were disappointing in practice - often being thought of as unresponsive or "gutless" to drive - many preferring the older 8 valve model. This was due to the bigger inlet ports suiting high power outputs at high revs but suffering from unstable gas flow and lack of ram effect at low revs under acceleration.

So we need to "measure" how a car actually accelerates (not what power is produced) to judge what difference we have made. The resulting "maximum" torque or bhp measured or calculated is actually irrelevant and does not necessarily correlate to the fastest acceleration engine characteristics between revs. Indeed at this point it is worthwhile considering if we need to relate accelerating performance to bhp or torque data at all. Perhaps we only need to compare acceleration data, however while the manufacturers, magazines (and everyone else) refers to those old bhp and torque parameters all the time - some comparisons would at least enable meaningful discussions to take place with others.

THE "ROAD DYNO"

So I propose the use of a method that measures acceleration (the thing we are most interested in) and then computes from that - torque and power - as the car is actually being driven along a real road. What we need is a "road dyno". Now if it was very expensive, complicated, time consuming or technically too challenging to set up a "road dyno" then we would be stuck with alternative testing methods which are a compromise. However the "road dyno" is in fact the simplest, cheapest and quickest method of accurately testing your car's performance.

Bearing in mind that it takes a few minutes to set up and about 30 minutes to analyse the data, in my view it is the biggest breakthrough in vehicle testing I have ever come across.
For around £100 it is possible to use your own "road dyno" to measure the performance of your own car with no hard wiring. With simple and easy instructions and with the help of a straightforward computer programme, the answers are revealed and graphs can be produced.

It provides the results we need simply and in the most relevant way possible with excellent comparative powers and recording systems.

Many comments on various web sites state that it doesn't matter if the results of a particular testing method are right or wrong because they all measure changes. This is absolutely wrong because each testing method produces different results that may or may not accurately explain the changes in road going performance and it becomes meaningless to try and compare results with others using different techniques or apparatus.

One thing that successful racing experience develops is pragmatism and first and foremost you have to decide if all you seek is bigger figures to impress others with or a car that goes faster. I believe that the only way to be sure of the latter is to test the whole car in motion on a road.

Previously - systems did not exist to do that and yet now - an inexpensive system exists to easily and quickly test your own car and get nearer to the result you probably seek. The alternatives on the market are expensive and private individuals could not justify or find room for the equipment.

So how does it work and why is it so accurate?

We are trying to measure the rate of acceleration. The engine drives the wheels through the gearbox and final drive ratio. If we measure the sparks of one cylinder - a computer can deduce the engine revs and change in revs (acceleration). By knowing the diameter of the wheel (and tyre profiles) and the gearing (gearbox and final drive) the computer can easily work out the rotational speed of the wheels and therefore the acceleration and speed of the vehicle. As long as the wheels do not spin - the result of taking 260 readings every 7.5 seconds will be a very accurate measure of the acceleration of the vehicle - which we are primarily interested in (also systems using accelerometers and even sat nav exist that I will be testing soon - but these are more expensive and cannot be any more accurate).

Next the interpretation of torque and power. This is also relatively easy. Torque = Inertia x acceleration so if the acceleration is known (because it was calculated) the only unknown is the Inertia (or resistance). This resistance is quite complicated because it takes into account all the things that make the vehicle not want to accelerate - like weight, frontal area, rotating masses, tyre drag etc. Much of this can be measured anyway and what cannot is remarkably consistent across different cars and clever research has come up with variables in the computer system that get very close to reproducing the correct answers.

Finally there is a "drive train loss" figure that you must input yourself that covers losses from the flywheel and acts as a general correction factor over and above the calculated ones for all the other variables. Now imagine you take a new stock 968 and test it with a road dyno and find out what figures to put in for that drive train loss - so that the "road dyno" results are exactly the same as the factory predictions. From that point onwards the road dyno will provide more accurate results for changes made to the vehicle than any other testing system I know. Furthermore - if all interested 968 owners who are planning improvements - had this dyno - and all used the same parameters, adjusting just for weight (including fuel load and driver weight) and air conditions, then the results of discussions about the various tuning mods and alterations Worldwide would be as valid as they could possibly get because even if there was a minute error in the factors used for say frontal area - this drive train loss correction will compensate for it.

We use a simple spreadsheet to work out vehicle weight for different models, driver weights and fuel load - then the following are inputted.

Test Weight: 1497 kg (adjusted for fuel load and driver weight etc)
Engine Type: 4 stroke
Transmission Type: M
TC Stall RPM: N/A
Axle Ratio: 3.78
Test Gear: 2
Tire Width: 255 mm
Tire Aspect: 40 %
Wheel Diameter: 17 inches
Drag Coefficient: 0.32 (which we follow advice on)
Frontal Area: 22.00 Ft^2 (which is general)
Drive-train Loss: 0.24 (this we can adjust to suit ourselves)
Air Temp: 10 C
Relative. Humidity: 60 %
Barometer: 999 mb
Elevation: 64 Ft
Correction Fact: 0.971 SAE

This is a one line example of the results provided.

We always test on the same stretch of road (which is slightly uphill to minimise wheel spin on very powerful cars). We also adjust the drive train loss for that uphill slope, fuel load, model type, driver weight and weather conditions with a correction factor.

All test techniques have to use some correction factor for internal losses - if they are going to try and record a measure of flywheel performance while inevitably connected to a transmission system under accelerating conditions - and we are no different. We have found that a correction factor of 0.24 (24%) (including an allowance for the internal resistance and the uphill slope) provides very close results to those quoted by Porsche and as such provides something reasonably meaningful to discuss with others. On a flat road it seems to be about 19 to 20 %. But whether it is right or wrong - it is all relative and we use the same parameters on all tests to enable us to compare results.

THE SALES PLUG!

For around about £100 (plus p&p Vat) we can provide a "road dyno" that will accurately reproduce acceleration and power outputs. This will reveal the condition of your car and the improvements due to any changes that you may make. By supplying to - hopefully - many owners - we can standardize the test parameters and information sources, building a meaningful databases of information that enable the benefits and pitfalls of various products and ideas to be analysed from a Worldwide source to every ones benefit.

Remember - it is accurate because it compares the gearing and tyre size (etc) with the engine revs to calculate the speed at which the wheel accelerates taking 37 readings/second. As long as the wheel does not spin it will be 100% accurate (and it is quite easy to avoid wheel spin in most models by using a higher gear for the test). The computer system estimates the torque and bhp necessary to achieve that rate of acceleration with parameters that can be adjusted for different models. As long as the type of car being tested is the same then even if some parameters are marginally wrong - the comparisons will be accurate, indeed I suggest that if everyone using this system with the same parameters as us, then the results would be meaningful. You only need a home computer and the system can be up and running very quickly - it takes no hard wiring and can be fitted to a car in minutes.

The default settings take a little time to input but we can provide help with this part of the procedure as part of our sales package - ensuring we all use the same basic settings - after which you can always rely upon it to reveal the true performance of your car.

For further information please E-Mail Barry Hart on auto@hartech.u-net.com marking the message ROAD DYNO ENQUIRY.

Many "experts" and specialists, egg heads and engineers may rightly question some or all of the assumptions I have made and the priorities I have recommended. A more comprehensive story about this topic and how these various discoveries were made will soon be available on our own web site - www.hartech.org.uk. Because it explains the source of information for this article as more of a story or chronology - how various tests were conducted and how some discoveries were made - I will not even consider entering into any arguments about the content of this article until and unless the longer version has been studied and understood. This doesn't mean that it does not contain any mistakes, errors or omissions and as my intention is only to help everyone appreciate the main issues - it does not negate nor criticise anyone else's findings or conclusions. One thing that learning a lot does do - is to make one aware of how little you actually do know and understand and how much more you would still like to find out. I am no different - but what I have learned has been tested and proven by me and also resulted in considerable track success against formidable opposition - so it can't be all wrong - however I would always welcome more knowledge from reliable proven sources but not from purely academic or intellectual arguments that have not been physically proven in a competitive - results based way.

.