Chassis dynos and Porsche hp
Discussion
We have discussed on here a few times about how chassis dynos are used for "tuning".
A base power run is done on a chassis dyno, adjustments on the map are carried out and another run is done to reveal the magic hp gain.
There are two videos below which are pretty self explanatory. We can probably guess whose new hp will be hanging around once off the chassis dyno and on the road.
I measure the American tuner with his "10%" 42hp power increase to be a 8 second run.
The Euro tuner's run took about 1 min 10 secs, It is plain to see how much load the engine is taking -now imagine holding it on an engine dyno under brake at 6000rpm for 30 seconds while adjusting the map - there is hp and there is hp !!
http://www.youtube.com/watch?v=Bsjk2eImC7M
http://youtube.com/watch?v=FAKST_z2JX8
A base power run is done on a chassis dyno, adjustments on the map are carried out and another run is done to reveal the magic hp gain.
There are two videos below which are pretty self explanatory. We can probably guess whose new hp will be hanging around once off the chassis dyno and on the road.
I measure the American tuner with his "10%" 42hp power increase to be a 8 second run.
The Euro tuner's run took about 1 min 10 secs, It is plain to see how much load the engine is taking -now imagine holding it on an engine dyno under brake at 6000rpm for 30 seconds while adjusting the map - there is hp and there is hp !!
http://www.youtube.com/watch?v=Bsjk2eImC7M
http://youtube.com/watch?v=FAKST_z2JX8
kennyboy said:
The euro tuner still uses a chassis rather than engine dyno though TB - better but still not ideal?
Yes but if you notice he is uploading a file, which presumably has been developed during steady state tuning elsewhere ? He doesn't actually do any "tuning" using this dyno in this mode. This loaded run is just to verify the power is there after the re-flash...TBH I am pretty sure the likes of WMs DD dyno would also give correct numbers used in this way when in its shootout mode - with a turbo car things would be different and the rear bumper would have to be removed.
Edited by TB993tt on Thursday 28th June 08:48
This is actually a huge topic.
However a test is done - if there is a huge power gain between tests then the results on the road are going to be better - so nit picking over methods would be less relevant.
However - if you are interested in small yet significant differences between power sources - there are differences between testing an actual car under acceleration on the road and testing in ANY OTHER WAY - that can mislead people and not always transfer to road going performance.
If you are only actually interested in bhp figures taken during a particular test and are not really interested in how much faster one car/engine is than another - then these discussions may be what you rely on but if you are more intrested in how much faster one will be on the road than another - you need to actually test it on the road.
Compared to a less tuned engine - a highly tuned naturally aspirated engine relies upon large inlets and exhausts (to move the increased air flow), extended cam timings (to increase specifc time area) and higher compression pressures (as a result of increased trapped air volumes).
However - large holes and time areas also suffer from unsteady gas flow initially until time passes and they settle down - so the rate of acceleration of an engine and the load it is under during that acceleration - all change the benefits measured compared to testing under different conditions.
This also means that one set of tuning may be better in lower gears and worse in higer gears than another as the gear ratios also change the loads and time for the engine to accelerate between it's rev band between the gears - and of course - for the same reason so to the gear ratios affect the result. Not fully understanding this was behind the dissapointment owners initially had who bought the Porsche 944S, VW and Peugeot Gti's when they first went to 16 valves and unsteady gas flow in the lower ranges produced less acceleration than the power graphs suggested would be possible.
Now an on car dynamometer measures how quickly the car is accelerating in any gear and works out from that the bhp required from the simplistic formula that acceleration is proportional to torque divided by resistance (in all its forms). The system works out the engine revs from the spark timing of the engine and as it has and already been programmed with the gear ratios and tyre diameter etc can work out the answers.
This not only gives you acceleration times between points but also a true comparison of power in the conditions the car is actually required to drive in - i.e. it is the most accurate measurement of the meaningful or useful horsepower generated as it is directly related to actual performance. In other words the bhp you calculate from the results reflects exactly how much faster or slower one car or engine is compared to another and makes the comparison of the results - discussing bhp differences between cars, engines or tuning mods - completely meaningful.
It also standardises tests in different places or even Countries as long as the resistance parameters used for the same models are the same.
Anything else - while comparing differences between cars or tuning - can still only be an indication of how the changes should affect the performance but never how it actually does.
A forced induction car is less likely to suffer such differences (and be a better guide to preformance changes) because it tends to increase torque accross the range between the gear change revs (not just peak bhp) and therefore provides a more consistent increase in bhp - although temperatures of exhaust, turbo pressures and intercooler temperatures between testing static and on the road can still make a difference.
While these are less significant in a four stroke - they are more crucial in a two stroke where power bands are smaller - and if anyone is interested there is more information about all this and other testing problems - on our web site buyers guide section 3. www.hartech.org Hopefully this will convey what a big subject it actually is! - such a big subject - in fact - that still comparatively few people fully understand it.
Baz
However a test is done - if there is a huge power gain between tests then the results on the road are going to be better - so nit picking over methods would be less relevant.
However - if you are interested in small yet significant differences between power sources - there are differences between testing an actual car under acceleration on the road and testing in ANY OTHER WAY - that can mislead people and not always transfer to road going performance.
If you are only actually interested in bhp figures taken during a particular test and are not really interested in how much faster one car/engine is than another - then these discussions may be what you rely on but if you are more intrested in how much faster one will be on the road than another - you need to actually test it on the road.
Compared to a less tuned engine - a highly tuned naturally aspirated engine relies upon large inlets and exhausts (to move the increased air flow), extended cam timings (to increase specifc time area) and higher compression pressures (as a result of increased trapped air volumes).
However - large holes and time areas also suffer from unsteady gas flow initially until time passes and they settle down - so the rate of acceleration of an engine and the load it is under during that acceleration - all change the benefits measured compared to testing under different conditions.
This also means that one set of tuning may be better in lower gears and worse in higer gears than another as the gear ratios also change the loads and time for the engine to accelerate between it's rev band between the gears - and of course - for the same reason so to the gear ratios affect the result. Not fully understanding this was behind the dissapointment owners initially had who bought the Porsche 944S, VW and Peugeot Gti's when they first went to 16 valves and unsteady gas flow in the lower ranges produced less acceleration than the power graphs suggested would be possible.
Now an on car dynamometer measures how quickly the car is accelerating in any gear and works out from that the bhp required from the simplistic formula that acceleration is proportional to torque divided by resistance (in all its forms). The system works out the engine revs from the spark timing of the engine and as it has and already been programmed with the gear ratios and tyre diameter etc can work out the answers.
This not only gives you acceleration times between points but also a true comparison of power in the conditions the car is actually required to drive in - i.e. it is the most accurate measurement of the meaningful or useful horsepower generated as it is directly related to actual performance. In other words the bhp you calculate from the results reflects exactly how much faster or slower one car or engine is compared to another and makes the comparison of the results - discussing bhp differences between cars, engines or tuning mods - completely meaningful.
It also standardises tests in different places or even Countries as long as the resistance parameters used for the same models are the same.
Anything else - while comparing differences between cars or tuning - can still only be an indication of how the changes should affect the performance but never how it actually does.
A forced induction car is less likely to suffer such differences (and be a better guide to preformance changes) because it tends to increase torque accross the range between the gear change revs (not just peak bhp) and therefore provides a more consistent increase in bhp - although temperatures of exhaust, turbo pressures and intercooler temperatures between testing static and on the road can still make a difference.
While these are less significant in a four stroke - they are more crucial in a two stroke where power bands are smaller - and if anyone is interested there is more information about all this and other testing problems - on our web site buyers guide section 3. www.hartech.org Hopefully this will convey what a big subject it actually is! - such a big subject - in fact - that still comparatively few people fully understand it.
Baz
It is a big subject, and an interesting one.
I always read with interest the input from TB, Baz Hartech and alike (plus guys on rennlist) and enjoy learning more about the obvious 'discrepancies' out there from these insightful posts.
Hopefully these discussions could in some way, one day, have a positive affect on the tuning industry - maybe some regulatory body in the future could borrow from these discussions.
There is so much money spent these days on 'perceived power upgrades' but the common flaw is that the results of these are measured in specific condition, one off, peak power figures on that specific dyno. As mentioned - the only real measurement is what affect the upgrade has on the car's performance.
Only then will the market understand why Porsche charge £8k for 30bhp when the remap outfit down the road charge £800 for 25bhp.
OK the £8k probably includes a margin so a company as large as Porsche can make a worth while profit but looking at the x51 kit for example you see where the money goes.
Another way of looking at it is - if the advert says 10% more power then surely I can expect my performance to increase by 10%, so many factors come in to play but it is still relative.
I always read with interest the input from TB, Baz Hartech and alike (plus guys on rennlist) and enjoy learning more about the obvious 'discrepancies' out there from these insightful posts.
Hopefully these discussions could in some way, one day, have a positive affect on the tuning industry - maybe some regulatory body in the future could borrow from these discussions.
There is so much money spent these days on 'perceived power upgrades' but the common flaw is that the results of these are measured in specific condition, one off, peak power figures on that specific dyno. As mentioned - the only real measurement is what affect the upgrade has on the car's performance.
Only then will the market understand why Porsche charge £8k for 30bhp when the remap outfit down the road charge £800 for 25bhp.
OK the £8k probably includes a margin so a company as large as Porsche can make a worth while profit but looking at the x51 kit for example you see where the money goes.
Another way of looking at it is - if the advert says 10% more power then surely I can expect my performance to increase by 10%, so many factors come in to play but it is still relative.
In reply and addition to what Baz has said some ppl may find this technical background usefull
http://www.dynamometer-info.co.uk/
http://sensors-transducers.globalspec.com/LearnMor...
These guys make a form of interesting dynometer that some ppl like and others don't so much:
http://www.rototest.com/rototest-dynamometer.php
Of course one of the most frequently talked about dynometer manufacturers is these guys:
http://www.dynojet.com/automotive_dyno/about.aspx
What may not be obvious is that AFAIK high end engine dyno's can be programmed to generate loads equivalent to those seen in the final vehicle application. This ultimately means that one can see the effect of how well the car will accelerate through the gears and determine for example the benefit from things like lightened flywheels which have 0 benefit to power measured at a static / fixed rpm.
http://www.dynamometer-info.co.uk/
http://sensors-transducers.globalspec.com/LearnMor...
These guys make a form of interesting dynometer that some ppl like and others don't so much:
http://www.rototest.com/rototest-dynamometer.php
Of course one of the most frequently talked about dynometer manufacturers is these guys:
http://www.dynojet.com/automotive_dyno/about.aspx
What may not be obvious is that AFAIK high end engine dyno's can be programmed to generate loads equivalent to those seen in the final vehicle application. This ultimately means that one can see the effect of how well the car will accelerate through the gears and determine for example the benefit from things like lightened flywheels which have 0 benefit to power measured at a static / fixed rpm.
When someone claims "10% more power" from a remap they may only be refering to a particular spot on the power curve. For example, at 4000 rpm the standard engine may produce 200 bhp and the "tuned" one may produce 220bhp. Hence 10% improvement. At all other points on the curve there may be no improvement or even some loss but the claim can be substantiated.
It really depends on what you are trying to do. If you want "bragging rights" at the pub then a "600 brake" remap may be good enough. The "tuner" may well be able to show 600 bhp on his rolling road but the reality, under real operating conditions, is more than likely far, far less. The 600 bhp is probably only, at best, a flash reading and cannot be supported by the engine / intercoolers / installation for more than a few seconds.
On the other hand, how many times do you really hold the throttle wide open for more than 30 seconds without a gearshift or a lift? I'll wager not many - certainly not in the UK! Under these conditions a remap may be good enough as the engine may not reach its thermal limit in such a short time. Now, take it on the track and it will be a different story. I wonder how many times a remapped car that runs "perfectly" on the street blows up or shows signs of distress when used on a track day? More than some people might like to admit I guess!
There is nothing clever about a dyno. It is a brake, pure and simple. Most nowadays are eddy current (like running an electric motor as a generator) but in the old days they were hydralic devices. Most will allow the transient response to be measured and I think this is the function that is used for the so called "shoot outs". The trouble with this is that they do not allow the engine to load up at any given speed hence lag, rotational inertia and other parameters come into play. Its all been said before but I reckon its horses for courses. Fundamental development and reliable, repeatable results can be had from a proper dyno, rolling roads are a compromise as they introduce many more variables. A good friend of mine who knows far more than me on the subject told me that he would never try to test an engine when it is coupled to a dyno via a piece of elastic! When you think about it, thats exactly what a rolling road has to do!
My advice, for what its worth, is have your engine remapped on a rolling road by all means. However, use a reputable company and do not go for all out power. Believe me, those last few horsepower ALWAYS cost a fortune!
Cheers,
Richard.
It really depends on what you are trying to do. If you want "bragging rights" at the pub then a "600 brake" remap may be good enough. The "tuner" may well be able to show 600 bhp on his rolling road but the reality, under real operating conditions, is more than likely far, far less. The 600 bhp is probably only, at best, a flash reading and cannot be supported by the engine / intercoolers / installation for more than a few seconds.
On the other hand, how many times do you really hold the throttle wide open for more than 30 seconds without a gearshift or a lift? I'll wager not many - certainly not in the UK! Under these conditions a remap may be good enough as the engine may not reach its thermal limit in such a short time. Now, take it on the track and it will be a different story. I wonder how many times a remapped car that runs "perfectly" on the street blows up or shows signs of distress when used on a track day? More than some people might like to admit I guess!
There is nothing clever about a dyno. It is a brake, pure and simple. Most nowadays are eddy current (like running an electric motor as a generator) but in the old days they were hydralic devices. Most will allow the transient response to be measured and I think this is the function that is used for the so called "shoot outs". The trouble with this is that they do not allow the engine to load up at any given speed hence lag, rotational inertia and other parameters come into play. Its all been said before but I reckon its horses for courses. Fundamental development and reliable, repeatable results can be had from a proper dyno, rolling roads are a compromise as they introduce many more variables. A good friend of mine who knows far more than me on the subject told me that he would never try to test an engine when it is coupled to a dyno via a piece of elastic! When you think about it, thats exactly what a rolling road has to do!
My advice, for what its worth, is have your engine remapped on a rolling road by all means. However, use a reputable company and do not go for all out power. Believe me, those last few horsepower ALWAYS cost a fortune!
Cheers,
Richard.
Some really good comments so far - especially relating to the question do you want bragging rights or a faster car?
However - I must make one important point and that is that unless you first read off the test results from a car in motion - and then programme a dynomometer to reproduce the exact dynamic load curves that were experienced (to give the same results) then you will never be sure that what you read on a dyno will reproduce exactly on the road.
If you did do what I suggest above then I think any dyno that does that (even an engine dyno) would be a useful tool to use - but to achieve the result you would first need that road going dyno and if you have that anyway - why go any further as they are incredibly accurate?
I am pleased to see that so many of you understand that it is actually very easy to get better breathing in a small rev band and at slightly higher revs - to give a higher bhp result - to boast about or sell a product from - and that it is the average power (or more acurately the average torque) within the gear change revs that determines ultimate performance and that to sacrifice a nice flat torque curve for a peaky one can result in a higher bhp peak figure but lower overall acceleration.
Hopefully many of you also appreciate that unless you change the gear ratios - raising the peak revs of an engine while using the same gearbox - actually requires you to have a broader power band as the revs you drop between gearchanges is a proportion of the peak revs reached and so any increase in peak revs results in the need for a broader power band - which is usually a very difficult thing to achieve.
Higher revs (and therefore usually peak bhp) can be more useful if you can then gear down the gearbox overall and make the ratios closer together - to achieve a similar top speed and a narrower rev drop/power band - because the acceleration is proportional roughly to the rear wheel torque - which in turn is proportional to the engine torque (and therefore bhp)and gearing it down increases the rear wheel torque again. This way an increase in revs can be useful but if (as most people are) you are stuck with the same gearbox - then raising the revs to get more bhp can be counterproductive as the car becomes basically overgeared for the application.
Fitting (for example) a 944 S2 gearbox to a 944 turbo can increase acceleration (although unfortunately they increased the rev drop between 4th and 5th) and so can be even better if a turbo 5th gear is also installed.
However road dyno's are very cheap to buy, easy to use and actually tell you eactly what you want to know - how fast is this car now and what kind on change in bhp did I achieve to get that extra performance I have just measured.
All other measures use correction factors that can make any result read well if you want it to.
Baz
However - I must make one important point and that is that unless you first read off the test results from a car in motion - and then programme a dynomometer to reproduce the exact dynamic load curves that were experienced (to give the same results) then you will never be sure that what you read on a dyno will reproduce exactly on the road.
If you did do what I suggest above then I think any dyno that does that (even an engine dyno) would be a useful tool to use - but to achieve the result you would first need that road going dyno and if you have that anyway - why go any further as they are incredibly accurate?
I am pleased to see that so many of you understand that it is actually very easy to get better breathing in a small rev band and at slightly higher revs - to give a higher bhp result - to boast about or sell a product from - and that it is the average power (or more acurately the average torque) within the gear change revs that determines ultimate performance and that to sacrifice a nice flat torque curve for a peaky one can result in a higher bhp peak figure but lower overall acceleration.
Hopefully many of you also appreciate that unless you change the gear ratios - raising the peak revs of an engine while using the same gearbox - actually requires you to have a broader power band as the revs you drop between gearchanges is a proportion of the peak revs reached and so any increase in peak revs results in the need for a broader power band - which is usually a very difficult thing to achieve.
Higher revs (and therefore usually peak bhp) can be more useful if you can then gear down the gearbox overall and make the ratios closer together - to achieve a similar top speed and a narrower rev drop/power band - because the acceleration is proportional roughly to the rear wheel torque - which in turn is proportional to the engine torque (and therefore bhp)and gearing it down increases the rear wheel torque again. This way an increase in revs can be useful but if (as most people are) you are stuck with the same gearbox - then raising the revs to get more bhp can be counterproductive as the car becomes basically overgeared for the application.
Fitting (for example) a 944 S2 gearbox to a 944 turbo can increase acceleration (although unfortunately they increased the rev drop between 4th and 5th) and so can be even better if a turbo 5th gear is also installed.
However road dyno's are very cheap to buy, easy to use and actually tell you eactly what you want to know - how fast is this car now and what kind on change in bhp did I achieve to get that extra performance I have just measured.
All other measures use correction factors that can make any result read well if you want it to.
Baz
I don't think anyone is suggesting that a dyno run and specialist tuning will not make a car better - it is when the bhp figures that result are bandied about between people that they do not always correlate to the improvements in performance that they suggest - for the reasons already stated.
Baz
Baz
In outright performance terms, it's the increase in area below the BHP (torque) curve that's important.
The early TurboTechnics Escort RS' were a perfect example of how the challenge for the highest BHP figure (at a given point) can make a car almost undriveable in normal conditions.
I drove one which almost set the tyres on fire between 3.5 and 4k --- but which was a pain everywhere else.
The early TurboTechnics Escort RS' were a perfect example of how the challenge for the highest BHP figure (at a given point) can make a car almost undriveable in normal conditions.
I drove one which almost set the tyres on fire between 3.5 and 4k --- but which was a pain everywhere else.
I dissagree - it is not the area under the bhp graph that makes a car fast. Of course more bhp is usually better than less - but that would also provide more rear wheel torque (if the gearing was for the same top speed). I could plot a bhp graph that had more area but accelerated more slowly than one that had more area under the torque graph - which is more relevant.
Firstly you must plot a graph showing the gear change points to establish the revs that drop each time you change gear through the range. From this you will see that as each gear change takes place as you go up through the gears, the revs that you pick up the next gear in - are higher - so the rev-band or power band for each gear is different. This is why you need to stipulate between which revs you are looking at the area under any graph - because it changes in different gears. A Peaky engine that produced a larger area just in the last 400 revs (but less area at lower revs) may be faster in top gear but would loose out more in first and second where the rev drop may be say 1000 or 1200 revs - because that weaker area of the graph is being used there.
If you then plot bhp in each gear as the revs rise and look at the areas under the graph netween the gear change points - this is a measure of the potential to do work but a similar graph showing the area between gear changes under a torque graph actually shows the potential to accelerate. Indeed if it is plotted for rear wheel torque (which reduces as the gears go up through the box) it reproduces quite accurately the rate of change of acceleration through the gears (such a graph is shown in the tuning article on our web site www.hartech.org)
Bhp is the same measure regardless of the revs but torque is not (so if you half the output speed through gearing then apart from minute transmission losses the output bhp is the same but the torque is double) - so if you alter gearing the bhp is the same but obviously the acceleration is not the same because it is proportional to output rear wheel torque not bhp. A flatter torque curve will therefore provide good torque in all the gears whereas as more peaky shape will benefit performance in some gears and handicap it in others. For serious track racing therfore both gear ratios and torque curves need to be considered.
One factor that is extremely important (and agrees that a peaky engine is not as beneficial as one with a flatter or more progressive torque curve) is to take account of strain energy losses.
The torque supplied twists the parts of the transmission and that absorbs energy. The more the rate of change of the rate of torque delivery - the more energy is lost in the system - and the same applies to tyres and their ability to transmit that torque - so a smoother torque delivery (or a bhp curve that more closely reflects a rise in proportion to revs) - the more of the torque that you are applying - is transmitted.
It is vital to understand that whatever shape the power curve - more of it is in use in the lower gears and the amount in use reduces as the gear changes rise. It is worth repeating that the shape of a power curve that may be best for say 1st or 2nd gear will not be the same as a curve for 5th or 6th gear which could have a smaller but higher peak. Similarly the rate of change of revs slows down as the gear changes increase (because the rate of acceleration reduces) so there is longer for the engine breathing dynamics to settle down and the engine could benefit from larger port areas than in lower gears when there is less time for it to stabilise. Furthermore, the combustion and exhaust temperatures inside the engine increase as the loads increase in higher gears - so the thermal (and resonant) dynamics also change. Fortunately the extra resistance at higher speeds means that any design parameters for optimising that rev band will benefit high speed applications - but most tests of acceleration (whether they are round a race track or sprint) are measures of acceleration at speeds much lower than the ultimate potential top speed of the car - and therefore serious contenders must tune for those areas first.
Because of all this - no - one set up will be perfect for each gear and an ideal engine would need to change many of its parameters as the gears change up to be optimised.
This is why - whatever you do on a dyno - there is no substitute for on the road comparative performance (say sprinting for example). A road dyno can record all this information as well and simply converts it into a reading for bhp that reflects how it performed.
So if 2 owners in different parts of the world were comparing an engine only torque or bhp graph (or static rolling road dyno results) for the same model - it may not be that the differences seen on the graphs between the two cars reflect how differently each car performed on the road - but if they both had the same road dyno and compared results - it would be far more meaningful.
Baz
Firstly you must plot a graph showing the gear change points to establish the revs that drop each time you change gear through the range. From this you will see that as each gear change takes place as you go up through the gears, the revs that you pick up the next gear in - are higher - so the rev-band or power band for each gear is different. This is why you need to stipulate between which revs you are looking at the area under any graph - because it changes in different gears. A Peaky engine that produced a larger area just in the last 400 revs (but less area at lower revs) may be faster in top gear but would loose out more in first and second where the rev drop may be say 1000 or 1200 revs - because that weaker area of the graph is being used there.
If you then plot bhp in each gear as the revs rise and look at the areas under the graph netween the gear change points - this is a measure of the potential to do work but a similar graph showing the area between gear changes under a torque graph actually shows the potential to accelerate. Indeed if it is plotted for rear wheel torque (which reduces as the gears go up through the box) it reproduces quite accurately the rate of change of acceleration through the gears (such a graph is shown in the tuning article on our web site www.hartech.org)
Bhp is the same measure regardless of the revs but torque is not (so if you half the output speed through gearing then apart from minute transmission losses the output bhp is the same but the torque is double) - so if you alter gearing the bhp is the same but obviously the acceleration is not the same because it is proportional to output rear wheel torque not bhp. A flatter torque curve will therefore provide good torque in all the gears whereas as more peaky shape will benefit performance in some gears and handicap it in others. For serious track racing therfore both gear ratios and torque curves need to be considered.
One factor that is extremely important (and agrees that a peaky engine is not as beneficial as one with a flatter or more progressive torque curve) is to take account of strain energy losses.
The torque supplied twists the parts of the transmission and that absorbs energy. The more the rate of change of the rate of torque delivery - the more energy is lost in the system - and the same applies to tyres and their ability to transmit that torque - so a smoother torque delivery (or a bhp curve that more closely reflects a rise in proportion to revs) - the more of the torque that you are applying - is transmitted.
It is vital to understand that whatever shape the power curve - more of it is in use in the lower gears and the amount in use reduces as the gear changes rise. It is worth repeating that the shape of a power curve that may be best for say 1st or 2nd gear will not be the same as a curve for 5th or 6th gear which could have a smaller but higher peak. Similarly the rate of change of revs slows down as the gear changes increase (because the rate of acceleration reduces) so there is longer for the engine breathing dynamics to settle down and the engine could benefit from larger port areas than in lower gears when there is less time for it to stabilise. Furthermore, the combustion and exhaust temperatures inside the engine increase as the loads increase in higher gears - so the thermal (and resonant) dynamics also change. Fortunately the extra resistance at higher speeds means that any design parameters for optimising that rev band will benefit high speed applications - but most tests of acceleration (whether they are round a race track or sprint) are measures of acceleration at speeds much lower than the ultimate potential top speed of the car - and therefore serious contenders must tune for those areas first.
Because of all this - no - one set up will be perfect for each gear and an ideal engine would need to change many of its parameters as the gears change up to be optimised.
This is why - whatever you do on a dyno - there is no substitute for on the road comparative performance (say sprinting for example). A road dyno can record all this information as well and simply converts it into a reading for bhp that reflects how it performed.
So if 2 owners in different parts of the world were comparing an engine only torque or bhp graph (or static rolling road dyno results) for the same model - it may not be that the differences seen on the graphs between the two cars reflect how differently each car performed on the road - but if they both had the same road dyno and compared results - it would be far more meaningful.
Baz
What a fantastic reply.
So, if an engine management system could take account also of the actual gear selected, and could optimise the power curve for the next change-up point, then one could have the best of all worlds --- within the limtations of the adjustment available.
Does this play to the stregths of an SMG type gearchange ?
So one would mechanically tune for the maximum overall power, and then use the electronics to tweek according to the actual gear selected and point in the rev range ... ?
So, if an engine management system could take account also of the actual gear selected, and could optimise the power curve for the next change-up point, then one could have the best of all worlds --- within the limtations of the adjustment available.
Does this play to the stregths of an SMG type gearchange ?
So one would mechanically tune for the maximum overall power, and then use the electronics to tweek according to the actual gear selected and point in the rev range ... ?
ChrisW - yes - that is exactly what you would achieve and naturally aspirated engines with electronic management of cam timing, valve lift, and obviously ignition, fuel etc with the addition of a sensor for the throttle opening and the gear you are in (or the ability to calculate it from the revs and road speed) would be able to optimise the performance of the engine throughout the rev range in different gears (allowing for and creating the changing power band requirements in each gear).
Such controls feedbacks exists today in many road cars (and I am sure all F1 cars) and as they are used more and more - the amount of tuning you can do will become more limited.
Fortunantely for the "chip" men - manufacturers have to tune down a little to allow for different fuels, mileages, standards of service etc - and also emmision outputs - so it is always possible to improve performance a little especially if the application changes slightly.
The actual power created depends very largely on the compression pressure inside the cylinder head at the point of correct ignition timing. All these variables either improve the amount of inlet charge that is trapped (so getting higher actual compression) or adjust the ignition etc to suit that same compression pressure for the revs, load and road speed etc. Anything that increases that compressed pressure makes the car faster (up to the point of detonation). Even in the most advanced engines there is probably scope to increase static compression a little and of course any kind of blower will also increase the inlet charge. For race purposes - smooth power at very low revs is unneccessary as well giving scope to tune a road car for the different application of racing and - as you will have already worked out - changing the gear ratios brings in a whole new range for the power band to work in and provides even more scope - but the more complex engines become and the more variables they build into the parameters - the less tuning will benefit output and perhaps more attention should then go towards reducing the other parameter in the acceleration = torque/resistance of weight and inertia loads etc.
Baz
Such controls feedbacks exists today in many road cars (and I am sure all F1 cars) and as they are used more and more - the amount of tuning you can do will become more limited.
Fortunantely for the "chip" men - manufacturers have to tune down a little to allow for different fuels, mileages, standards of service etc - and also emmision outputs - so it is always possible to improve performance a little especially if the application changes slightly.
The actual power created depends very largely on the compression pressure inside the cylinder head at the point of correct ignition timing. All these variables either improve the amount of inlet charge that is trapped (so getting higher actual compression) or adjust the ignition etc to suit that same compression pressure for the revs, load and road speed etc. Anything that increases that compressed pressure makes the car faster (up to the point of detonation). Even in the most advanced engines there is probably scope to increase static compression a little and of course any kind of blower will also increase the inlet charge. For race purposes - smooth power at very low revs is unneccessary as well giving scope to tune a road car for the different application of racing and - as you will have already worked out - changing the gear ratios brings in a whole new range for the power band to work in and provides even more scope - but the more complex engines become and the more variables they build into the parameters - the less tuning will benefit output and perhaps more attention should then go towards reducing the other parameter in the acceleration = torque/resistance of weight and inertia loads etc.
Baz
Actually most modern ECU's do take into account the gear you are in. On a turbo 911 there are different corrections for intake air temperature for different gears. This allows the ECU to take advantage of the increased efficiency of the IC's at higher speeds. As a matter of fact there are loads of variations being gear dependant. The throttle sensitivity is different in reverse for example and can be different in 1st - 3rd and 4th - 6th. This is not used in any Porsche code but the provision is there. They also have different rev limits in different gears to keep the pecking order of the 911's in sync.
I completely agree about tuning on the circuit or road. There is simply no way to reproduce air turbulence for the intake and rads, temperature, and load except for doing it live and on the road.
McLaren take a new engine and set the telemetry data from the ECU and load it into their dyno. They then make an exact duplicate for load, rpm, temp, etc, and run the new engine as if it had just completed the last race. Then they load the data for the next race and run it again to pre test for potential failures. And they thought the engine rules were saving money… Hah!
I completely agree about tuning on the circuit or road. There is simply no way to reproduce air turbulence for the intake and rads, temperature, and load except for doing it live and on the road.
McLaren take a new engine and set the telemetry data from the ECU and load it into their dyno. They then make an exact duplicate for load, rpm, temp, etc, and run the new engine as if it had just completed the last race. Then they load the data for the next race and run it again to pre test for potential failures. And they thought the engine rules were saving money… Hah!
Well I am truly amazed - this is the first time I have posted something about engine dynamics and power/performance measurement - when there seems to be general agreement - it usually ends up with huge arguments (which it turn makes me reluctant to comment too often). It seems that ones belief in what makes engines and cars go fast is almost as controversial as religion and politics for many people - but there are increasingly few of us who have actually had the opportunity to put their understanding into practice and created engines that have beaten all opposition - for which your own understanding obviously must be quite good!.
Being aware that modern cars pick up all the information they would need to operate such a programme - it is no suprise to me that F1 teams do exactly what I have described as the only way to properly tune an engine/ECU for racing competitively using a fixed dyno. There are many benefits in a fixed dyno - not the least of which is the ability to measure and/or alter parameters easily - and - as long as you know what you are doing and your comparisons are between the same car or engine - it is not difficult to make gains that work out quicker on the track.
However - using figures gained in different establishments, for different cars and different applications and then arguing about what they mean in terms of comparative performance is something that actually a road dyno is better able to support.
Fixed dyno engine testing is expensive and yet a road based dyno can be bought for under £150 and easily used by anyone with a little knowledge of computers and cars to compare the actual results of tuning. I have absolutely no connection with anyone making them - it just seems to me ironic that such a cheap and simple device that all those interested in tuning or racing could use for themselves - is often regarded as a cheap, nasty and inaccurate device when actually it is probably the most useful and accurate that anyone without expensive equipment could have - and the most relevant to judge and compare the results of tuning devices and products that they may have bought, fitted or want to compare with others who have.
Unfortunately I will not be able to connect to the Internet for a couple of weeks - so - if the arguments do start - my lack of response is for that and no other reason.
Baz
Being aware that modern cars pick up all the information they would need to operate such a programme - it is no suprise to me that F1 teams do exactly what I have described as the only way to properly tune an engine/ECU for racing competitively using a fixed dyno. There are many benefits in a fixed dyno - not the least of which is the ability to measure and/or alter parameters easily - and - as long as you know what you are doing and your comparisons are between the same car or engine - it is not difficult to make gains that work out quicker on the track.
However - using figures gained in different establishments, for different cars and different applications and then arguing about what they mean in terms of comparative performance is something that actually a road dyno is better able to support.
Fixed dyno engine testing is expensive and yet a road based dyno can be bought for under £150 and easily used by anyone with a little knowledge of computers and cars to compare the actual results of tuning. I have absolutely no connection with anyone making them - it just seems to me ironic that such a cheap and simple device that all those interested in tuning or racing could use for themselves - is often regarded as a cheap, nasty and inaccurate device when actually it is probably the most useful and accurate that anyone without expensive equipment could have - and the most relevant to judge and compare the results of tuning devices and products that they may have bought, fitted or want to compare with others who have.
Unfortunately I will not be able to connect to the Internet for a couple of weeks - so - if the arguments do start - my lack of response is for that and no other reason.
Baz
Gassing Station | Porsche General | Top of Page | What's New | My Stuff