Rpi scoop and afe intake..

[exdos]

fair enough it's a ambitious claim

Only reporting the data from my Z4MC's ECU...

I said it's ambitious, that's not refuting it..

I'm just making it clear that I'm not making any claims I'm just showing recorded data. Don't shoot the messenger :guns: :guns:

 

[tertius]

Tertius,

This is LONG, but you did ask. :wink:

<EDIT>
Which formula have you used to calculate the BMEP of the Formula 1 car and mine, and what numbers did you use? What BMEP figure did you calculate for mine?
<EDIT>

I used the formula here: http://www.autospeed.com/cms/article.html?&title=The-Real-Way-of-Comparing-Engine-Designs&A=111718

For the F1 engines I used 750 bhp (=559.3 kW) @ 18,500 rpm; capacity 2.4L

For your engine I took the peak power point from graph you posted:370 whp at 7,800 rpm

I assumed 15% losses to arrive at a flywheel figure (needed for a BMEP calculation) = 435.3 bhp

Convert to kW = 324.6 kW

Run those numbers through the formula = 15.6 BAR

Firstly, I’d like to say that I do understand the skepticism that may surround this subject and my figures. I’ve read LOTS of discussions on a host of different forums, and ram-effect and the use of dynamometers seem to be the two subjects that really polarize opinion! Postings like that of SweetRide are not unusual, but I much prefer postings like yours, which are in the spirit of serious and polite discussion.

Good ...

As I’ve said earlier, I originally got into datalogging in order to find out for myself what really goes on in the air intake and the effect on performance because there’s very little factual information so discussion without the information doesn’t really lead anywhere. From my research it seems that ram-effect for car intakes was pursued in the 1950s, particularly by Chrysler for its commercial products, but also by various independent engineers for race applications. There were several problems though back in those days: electronic fuel injection (EFI) wasn’t available and cars were not equipped with a host of electronic devices working in conjunction with an ECU which could be observed and datalogged so understanding the principle of ram-effect in real world driving conditions was hard to asses. In those days, fuelling was with carburetors, which required balancing fuel flow with the airflow passing into the engine and if you’ve ever done any carburetor tuning, you’ll know that you have to find a sweet spot as the “best compromise” adjustment, whereas with EFI, fuelling is optimized under all airflow conditions.

The problem with an air-intake which harnesses the ram-effect in conjunction with carburetors, is that the flow of air is variable and consequently, this means that the carburetor will perform differently at different speeds and in different gears/engine rpms. As such, ram–intakes caused unsolvable problems and so ram intakes fell out of favour. As such, there’s very little information about the use of ram-effect for road cars using EFI in the public domain. Any information, with actual data is very hard to come by, and despite searching incredibly hard, I’ve yet to find the sort of data that I have produced, although there may be others treading the same or similar path to me. Consequently, I consider that the study of ram-effect and its possible usefulness is a puzzle for which I need to piece together the clues as and when I find them. I've had to follow my hunches and intuition, based on my findings, throughout.

Whilst a static dynamometer can be a useful tool, it is in fact a “one-trick pony” in that it provides information about torque and power, only. The accuracy of them, the repeatability of their measurements and comparability from individual dynamometers and those of different manufacture, seems is still open to debate and will always cause argument. In comparison, the DashDyno and other datalogging equipment connected to a car’s ECU via the OBDII socket simply reads and records whatever information the ECU will yield. The same information of Intake air temperature (IAT) air-flow through the MAF, engine rpm, and a host of other PIDs (Parameter Identities) are used by the ECU in controlling fuelling using the engine mapping system provided by the manufacturer, or from aftermarket remapping. I therefore consider that the raw data recorded by a DashDyno is 100% accurate and that the DashDyno is an advanced hi-tech piece of equipment and certainly not a toy.


So, when it comes to using the DashDyno as a real-world dynamometer to record “at the wheels” figures for torque and hp, the device uses the data from the car’s ECU coupled with a car specific “profile” which is entered by the user (i.e. me) and this is used by the DashDynos software to compute its dyno results. For all my dyno runs in any car I always use 3rd gear and so I set up the profile for each car with the published weight (allowing for my weight and fuel), 3rd gear ratios, Cd, frontal areas, tyre diameter, geographic elevation, ambient air temperature, pressure and humidity. I always do my dyno runs on the same private road and when it’s not very windy and I use a couple of local weather stations with websites to get the environmental conditions at the time of my runs. When performing the runs, I cruise in 3rd gear at low revs until the green light comes on the DashDyno and then I give it WOT and hang on! When I stop the car the DashDyno produces the saved data as a plot of torque and hp. I’ve been using the same methodology throughout my datalogging of all different cars and with various different modifications under test so that I can ensure repeatability and accuracy to show me whether a modification has any significant effect on performance. I have absolutely no interest in attempting to produce bogus figures to fool anyone, because I’m looking for enlightenment and knowledge of what actually happens to performance with mods.

To answer your specific questions:
1. The figures are produced by the DashDyno after a run using the profile loaded for the car, as described above. There’s no wheel spin in 3rd gear. The dyno results that I obtained from the Dashdyno for my Z3MC are very comparable to those obtained on the dynamometers of Evolve and Powerstaion (as I’ve posted earlier in this thread). Likewise, the dyno results I’ve obtained with other cars in OEM state are very comparable to the manufacturers published figures and with modding of their intakes, I’ve seen significant improvements in torque/hp. I therefore consider the DashDyno to be accurate and to be reliable. Since the DashDyno produces “at the wheels” figures, only, there is absolutely no estimation of “at the flywheel” figures as used by static dynamometers, and so what you see is what you actually get, at the wheels. I therefore consider this to be a better way of measuring a car’s performance, because there is no favourable guesswork of adding “losses” into the results.

I think this is where I start to have concerns:

Firstly there should be no “guesswork” in adding losses when using a chassis dyno – the losses are measured under coastdown and the correction applied to the results. You are correct that comparability between dynos is may not be 100% but when using the same dyno you should have extremely good repeatability.

Secondly it is important that you do use flywheel figures from chassis dyno results – the “wheels” figure on a chassis dyno is easily manipulated, for example by adjusting tyre pressures, but that manipulation will be corrected for when applying the coastdown calculated correction.

Thirdly flywheel figures are much more useful as they allow comparison to manufacturer figures which are always given at the flywheel.

Finally, whilst I do not doubt that the data gathered by DashDyno is accurate, the performance figures it generates are not gathered at all, but are derived from other data points.
As far as I can tell the DashDyno works in effect by saying “IF the engine is turning at this RPM; AND the car is in this gear; AND the final drive is X; AND the wheel/tyre size is Y; AND the frontal area of the car is Z AND the drag coefficient is A and the weight of the car is B AND the air is still AND the road is perfectly flat THEN the engine must be producing THIS much power in order to do that.
It is all second order stuff, there are a LOT of dependencies there and I would be much happier if there were independent data points that could cross-reference it – e.g. in gear acceleration figures from something like a VBox, that uses GPS.

In any event, even if my DashDyno were not 100% accurate and is over-reading the dyno results, any increases seen from my modding are "relatively" greater than OEM, so gains are still being produced "relative" to OEM, so the gains are still significant and real, although the quantum may be in dispute. However, I don't see anyone challenging the OEM figures my DashDyno produces.

I don’t think I have seen OEM comparable flywheel figures – perhaps I have missed it?

2. Because I lifted off before getting to the redline. As I’ve stated in a previous posting, in my ramcharger mod run, I lifted off slightly to go over a slight crest in the road and then went back on WOT. If you lift off too much the DashDyno considers the run has been terminated and so only calculates for the duration of WOT in that run.

That is unfortunate as it means an important part of the graph is missing.

However, it raises another question – is there a downhill after the crest? If so then it might explain the spike in your graph – the car will accelerate downhill but the DashDyno won’t know it is because of the hill and will assume it is due to power.

3. Yes, the data recorded by the DashDyno can be exported into .csv format, which can be used in Excel. I’ve made an Excel template, which calculates a “rolling” Volumetric Efficiency (VE) from specific data that I log. I’ve found that by observing the graphical performance of VE has been most helpful in helping me to understand exactly how the air-intake “works” and how ram-effect can be improved with specific mods to increase dynamic pressure (i.e. ram effect).

And does your graph show power increasing proportional to speed as you suggest would be the case?

As I have said I am not an expert but from the little reading I have done I understand that what you need to utilise ram effect is increased static pressure (ie increased density) not dynamic pressure.
I found this post useful in understanding it: http://forums.corvetteforum.com/c5-tech/336068-dispelling-the-myth-of-ram-air-effect.html

4. So far, I’ve found that the S54 engines in both my Z3MC and Z4MC have been able to use all the air that I can give them, without any need for a remap. The S54 has a published VE of 105% but in 1st gear it only achieves up to 70%, in 2nd gear up to 90%, but in 3rd gear and above the VE can go above 100%: this difference occurs due to the OEM intake harnessing the ram effect. In the case of the Honda S2000, the published VE is 115%, yet if you read the S2000 forums, owners seem to want to ditch the OEM air-intake, which makes a VE of 115% possible in favour of some aftermarket kit. I find this laughable! Most people just don’t believe or understand how all this works.

Surely there is much more to volumetric efficiency than the air intake? Most discussions I have seen about improving VE have focused on the cylinder head itself – porting the valve inlets, changing cams for more aggressive lift and so on?

As previously said I’ve done a similar ramcharger mod for my Z3MC and I gave a link to a video of me chasing another Z3MC. I know the owner of that car very well and have driven with him a lot. He is hugely experienced and has instructed many others at The Ring, and he taught me the lines on my first trip there. Both our cars have the same aero mods and I’ve followed that car around The Ring before and we were very evenly matched on previous trips. On my visit shown in the video, I’d recently manufactured my ramcharger and I was astonished at the way my car could catch him up on the straight sections (where driving ability is not a factor – that’s why I showed that section). Here's another one: http://www.youtube.com/watch?v=ICLoOSJEkpY

I’m not sure how much you can really conclude from those two clips. I certainly would not agree that the first section does not require driving ability ... that is Flugplatz approaching Schwedenkreuz, perhaps the single most dangerous section of the track and almost the only place where a car driver could die. Moreover, although there is some gain in the first section, the majority gain is when the lead car is approaching the crest and then over it into the braking zone. Similar cars will always close up in that situation.

In the second clip the two cars are relatively close so the aero behaviour of the second will be affected by the closeness of the first. Plus as soon as you approach the Bilstein bridge driver ability (and balls) definitely come into it.

(Also you might need to have words with your friend about his line into the first right of the chicane leading into T13 ... he's giving loads of track and hence corner speed away)

I’m going to The Ring in April and will have the opportunity to run with several other guys I’ve driven with at The Ring before. I shall record my runs with my DashDyno up and running and I will then have lots of data to see how the car performs at all speeds.

When in April? I will be there for the DN days right after the Easter weekend.

However, I am dubious how much can be learned from on track data during TF or track days, when you are usually compromised by other cars, conditions and incidents.

FYI, my Z3MC in OEM condition, will maintain the pressure, at ambient pressure or above, within the air-filter box at WOT, in all gears and at all speeds above 90mph, but with a series of different mods, including a ramcharger mod, I’ve managed to get the same car to now maintain the pressure, at ambient pressure or above, within the air-filter box and perform at WOT in all gears at speeds above 60mph, and my ramcharger reduced this from 80mph with all previous mods. My experience of developing my air-intake mods for my Z3MC was part of the reason why I bought my Z4MC, because I wanted to see if I could do the same/better with a S54 engine with a slightly different OEM intake.

Throughout my experimentation with modifying air intakes, I've never made any predictions of what performance increases that I expect to see. I merely make a change that I think might work and then datalog in a "suck and see" approach, just like all the inventors of the past have done.

Are you still awake?

Yes.
Sorry about the delay in replying, I didn’t have time in front of a PC until this morning.

I hope you understand that I am not criticising but challenging in, I hope, a constructive way. I think it would be fantastic if the results you are seeing are verifiable, but the numbers are so striking that I think they need to be tested.

I would very much like to see some independent testing in repeatable conditions – as I have said side by side runs against a known similar car, would be great; VBox data from acceleration runs; even chassis dyno results would all be very useful data points.

 

[ChawenHalo]

I think you should duct-tape a stock intake around your head and take a few deep breaths... Then do likewise with a AFE Pro5r filter. You'll find out INSTANTLY which breathes better. This is about as scientific as your bubble-gum approach to all of this. You'll be laughed down the street in any real-world capacity with your claim of an 8.5% increase in TQ with your 'pressure-water-tested-home-made-intake - let alone the outlandish HP numbers. Regardless, if you think BMW left all that power on the table with the S54... Then I've got a bridge to sell you.

You're claiming that your AFE intake breathes better than the OEM intake set-up, yet you're also telling me that I think BMW left power on the table. It therefore seems that you think BMW left something for you too! So why didn't BMW save itself a lot of money and simply fit an AFE filter like you've done?

Tell us; how is it possible for a normally aspirated F1 engine, with a 2.4l displacement, to make around 720bhp. I've guessed; they use your AFE filter!

Of course BMW left "some" power on the table. That's called safe optimum operating parameters. You know when the girlfriend pooddles along at 1800 rpm in 4 th gear; when over rev the engine with a bad heal and toe on the track or when you buzz it trying to overtake that pesky Exige on the main straight. I don't believe for a minute all these tuning companies who make this stuff engage in anything more than a "tried and tested" approch to their equipment? Tuning poarts are easy money.

Just a practicle example: When well known and respected tuner AC SChnitzer built their "Profile" Z4M Coupe, the carbon airbox and engine mapping wasn't finished and perfected when tested.

Its not at all surprising that Exdos has fount some measurable increases with his approach. On this forum we all (I don't think I talking out of turn hear) respect and like the fact that at least he is trying a more scientific approach, as basic as it may be.

I love the fact that I get more noise out of my set up. And yes in THEORY back boxes = +5bhp + Gruppe M 8bhp = a +13 bphr WHAO! Could I feel it - NO :roll: do I believe it in the first place of course not! :lol:

The only way I credibly believe I can more reliable power is to change some engine internals such as more agressive cams, new valves and springs, a massive airbox a piggy back/ standalone EMS; Alpha numerique etc. So looking at £8-10k worth. BUT then I would have to accept a drop in driveability and tolerance

Thats why Ferraris are 200k. (well + the badge)

 

[exdos]

I think this is where I start to have concerns:

Firstly there should be no “guesswork” in adding losses when using a chassis dyno – the losses are measured under coastdown and the correction applied to the results. You are correct that comparability between dynos is may not be 100% but when using the same dyno you should have extremely good repeatability.

Secondly it is important that you do use flywheel figures from chassis dyno results – the “wheels” figure on a chassis dyno is easily manipulated, for example by adjusting tyre pressures, but that manipulation will be corrected for when applying the coastdown calculated correction.

Thirdly flywheel figures are much more useful as they allow comparison to manufacturer figures which are always given at the flywheel.

Finally, whilst I do not doubt that the data gathered by DashDyno is accurate, the performance figures it generates are not gathered at all, but are derived from other data points.
As far as I can tell the DashDyno works in effect by saying “IF the engine is turning at this RPM; AND the car is in this gear; AND the final drive is X; AND the wheel/tyre size is Y; AND the frontal area of the car is Z AND the drag coefficient is A and the weight of the car is B AND the air is still AND the road is perfectly flat THEN the engine must be producing THIS much power in order to do that.
It is all second order stuff, there are a LOT of dependencies there and I would be much happier if there were independent data points that could cross-reference it – e.g. in gear acceleration figures from something like a VBox, that uses GPS.

But do the figures produced by static dynos account for the environmental conditions, which have considerable impact on performance? In other words: are the result “standardised”? Most people seem to only like to compare back-to-back results performed on the same day/dyno if they are claiming accuracy of measuring. In comparison, the DashDyno figures are standardised. It doesn't matter whether I do a dyno run in the heat of summer or the depths of winter; provided the road is dry and there is no wind I get comparable results, and that's what I want. I've got baseline data from when the car was OEM, so all my comparison are with that data.

You are right that the DashDyno computes the figures from the collection of data as you’ve described, but doesn’t a static dyno also have to do the same thing relating to the turning of drums? Aren’t both systems actually doing the same thing, but from the opposing points of contact between the tyres and the opposing surface (road or drum) i.e. measuring the same equal and opposite forces? The DashDyno measures the rotation of the car’s wheels and the static dyno measures the rotation of the drum. Therefore, as I see it, the accuracy of any dyno system, be it static dyno or DashDyno, will only be as good as the formulae they use in their software, and why should the DashDyno’s formulae be any worse than those of any static dyno?

Obviously, you are used to seeing dyno figures presented “at the flywheel”, whereas the DashDyno records “at the wheels”. As a user of the DashDyno, I don’t see a problem with this. Since I’m always recording and comparing DashDyno figures with each other, I am always using the same system with the same methodology, thus all my comparisons are direct. I’m always comparing apples with apples. At the end of the day, I’ve done all my R&D for me, and I’m just letting others have a glimpse at what I’ve been up to. Others might use different methodology to me, but I'm trying to be as scientific and unbiased as possible with the facilities that I have available to me.

In any event, do you really think that it’s possible to properly log engine performance on a static dyno, when the vehicle incorporates the ram-effect in the intake system? I don't. When I had my Z3MC dynoed at Evolve, I measured the pressure in front of the fan, and with reference to my earlier graph of pressure versus forward velocity, this equated to a velocity of just over 70mph, yet the car was dynoed in 3rd gear and was taken to a speed of 106mph at 7600rpm. As such, the TRUE performance of my car in “real world” conditions was not measured, because it was impossible to replicate and record the engine performance on the static dyno. As it happens, my Z3MC recorded the highest figures that Imran had seen for a S54 Z3MC on the Evolve dyno (that's been written about elsewhere), and that was with a few intake mods and modified silencers, and before I did my "ramcharger" mod for that car.

However, it raises another question – is there a downhill after the crest? If so then it might explain the spike in your graph – the car will accelerate downhill but the DashDyno won’t know it is because of the hill and will assume it is due to power.

I think this is irrelevant since I use the same road for all my runs. However, the road has a very slight uphill gradient followed by a plateau. I’d rather have a disadvantageous situation to measure.

3. Yes, the data recorded by the DashDyno can be exported into .csv format, which can be used in Excel. I’ve made an Excel template, which calculates a “rolling” Volumetric Efficiency (VE) from specific data that I log. I’ve found that by observing the graphical performance of VE has been most helpful in helping me to understand exactly how the air-intake “works” and how ram-effect can be improved with specific mods to increase dynamic pressure (i.e. ram effect).

And does your graph show power increasing proportional to speed as you suggest would be the case?

I haven’t said that power increases proportional to speed. I think that the best that can be achieved with ram effect is maintaining torque throughout the engine rev range rather than seeing the usual drop off after peak torque has occurred somewhere in the mid rev range. The Vanos and other Variable Valve Timing/Lift systems greatly assist in preventing torque loss, but improving the air supply with ram effect also assists.

As I have said I am not an expert but from the little reading I have done I understand that what you need to utilise ram effect is increased static pressure (ie increased density) not dynamic pressure.
I found this post useful in understanding it: ‪http://forums.corvetteforum.com/c5-tech ... ffect.html

Yes, you do need to convert the dynamic pressure into an increase in static pressure and I’ve been recording static pressure throughout as part of my R&D and recording static pressure has been very useful in guiding me in the right direction.

There’s loads of articles like that out there and it’s based solely on the guy's theoretical interpretation. He should try doing some experimentation and he would change his opinion.
You referred to the Autospeed website: you should read the articles on “Eliminating negative boost” http://www.autospeed.com/cms/article.html?&title=Eliminating-Negative-Boost-Part-1&A=0629 there’s loads of stuff on there about this subject. You’ll also find some very useful stuff here too: http://www.sportrider.com/tech/146_9508_ram/viewall.html Both sources actually go to the trouble of experimentation and measuring in the real world situation, same as I do.

Surely there is much more to volumetric efficiency than the air intake? Most discussions I have seen about improving VE have focused on the cylinder head itself – porting the valve inlets, changing cams for more aggressive lift and so on?

Of course there is! Volumetric Efficiency is simply a measure of the mass of air inside the cylinders compared to the mass at ambient pressure. If it’s above 100% then you’ve cleverly managed to shove a bit more air into the cylinder than static air pressure alone would do. A turbo or supercharger does this, as do resonant effects inside the air intake system, less restrictive exhaust etc. Getting performance increases is a case of stacking as many things in favour as possible and calculating VE is a way of measuring the effectiveness of engine "breathing".

When in April? I will be there for the DN days right after the Easter weekend.

Yes, I’m booked too.

(Also you might need to have words with your friend about his line into the first right of the chicane leading into T13 ... he's giving loads of track and hence corner speed away)

You’ll be able to tell him yourself in April. :wink: I very much look forward to trying to keep up with you and learning the correct lines from you! :wink:

However, I am dubious how much can be learned from on track data during TF or track days, when you are usually compromised by other cars, conditions and incidents.

It will be a great opportunity to datalog the car in real world conditions at all speeds and well above our legal limit. I’ll learn a lot from my data. I don’t do track days to try to be a hero, I want to come home unscathed and be able to do it all again many times more.

I hope you understand that I am not criticising but challenging in, I hope, a constructive way. I think it would be fantastic if the results you are seeing are verifiable, but the numbers are so striking that I think they need to be tested.

No problem at all. I’m always happy to discuss and debate in a civil manner. 8)

 

[Bad_Pritt]

lol... you guys have waaaaay too much spare time

 

[exdos]

lol... you guys have waaaaay too much spare time

Yep, I'm retired and I can highly recommend it! 8)

 

[Bad_Pritt]

that explains 8)
retirement is a little soon for me :roll:

 

[exdos]

that explains 8)
retirement is a little soon for me :roll:

I often wonder how you guys, who are supposed to be at work, have the time to participate on forums during the day time.... :wink:

I've just done my job for today; fitted a set of AC Schnitzer pedals to make heel & toe easier for my small feet.

 

[Bad_Pritt]

it's called "multitasking" 8)

 

[Bing]

it's called "multitasking" 8)

Or 'bored and looking for a new one'

 

[tertius]

I think this is where I start to have concerns:

Firstly there should be no “guesswork” in adding losses when using a chassis dyno – the losses are measured under coastdown and the correction applied to the results. You are correct that comparability between dynos is may not be 100% but when using the same dyno you should have extremely good repeatability.

Secondly it is important that you do use flywheel figures from chassis dyno results – the “wheels” figure on a chassis dyno is easily manipulated, for example by adjusting tyre pressures, but that manipulation will be corrected for when applying the coastdown calculated correction.

Thirdly flywheel figures are much more useful as they allow comparison to manufacturer figures which are always given at the flywheel.

Finally, whilst I do not doubt that the data gathered by DashDyno is accurate, the performance figures it generates are not gathered at all, but are derived from other data points.
As far as I can tell the DashDyno works in effect by saying “IF the engine is turning at this RPM; AND the car is in this gear; AND the final drive is X; AND the wheel/tyre size is Y; AND the frontal area of the car is Z AND the drag coefficient is A and the weight of the car is B AND the air is still AND the road is perfectly flat THEN the engine must be producing THIS much power in order to do that.
It is all second order stuff, there are a LOT of dependencies there and I would be much happier if there were independent data points that could cross-reference it – e.g. in gear acceleration figures from something like a VBox, that uses GPS.

But do the figures produced by static dynos account for the environmental conditions, which have considerable impact on performance? In other words: are the result “standardised”? Most people seem to only like to compare back-to-back results performed on the same day/dyno if they are claiming accuracy of measuring. In comparison, the DashDyno figures are standardised. It doesn't matter whether I do a dyno run in the heat of summer or the depths of winter; provided the road is dry and there is no wind I get comparable results, and that's what I want. I've got baseline data from when the car was OEM, so all my comparison are with that data.

Yes they do – chassis dynos can correct for barometric pressure (typically of course, I can’t speak for all).

You are right that the DashDyno computes the figures from the collection of data as you’ve described, but doesn’t a static dyno also have to do the same thing relating to the turning of drums? Aren’t both systems actually doing the same thing, but from the opposing points of contact between the tyres and the opposing surface (road or drum) i.e. measuring the same equal and opposite forces? The DashDyno measures the rotation of the car’s wheels and the static dyno measures the rotation of the drum. Therefore, as I see it, the accuracy of any dyno system, be it static dyno or DashDyno, will only be as good as the formulae they use in their software, and why should the DashDyno’s formulae be any worse than those of any static dyno?

No I don’t think they are comparable in the way you suggest – a chassis dyno applies load to the rear wheels and measures the actual force applied to the rollers. Dashdyno reads engine performance data and using input variables for gearing, air resistance and so on derives a power figure from that.

Example 1: if you attached a parachute to your car and did your test run then DashDyno would show much worse results because it wouldn’t know about the increased resistance, but the engine would be producing the same power as before.

Example 2: if you put the rear on axle stands in a Ferris Bueller sort of way and ran it up through the gears I imagine that DashDyno would show fantastic results because it would have no load, but again the engine would still be exactly the same.

Obviously, you are used to seeing dyno figures presented “at the flywheel”, whereas the DashDyno records “at the wheels”. As a user of the DashDyno, I don’t see a problem with this. Since I’m always recording and comparing DashDyno figures with each other, I am always using the same system with the same methodology, thus all my comparisons are direct. I’m always comparing apples with apples. At the end of the day, I’ve done all my R&D for me, and I’m just letting others have a glimpse at what I’ve been up to. Others might use different methodology to me, but I'm trying to be as scientific and unbiased as possible with the facilities that I have available to me.

But at the flywheel is the accepted measure for comparison, not using it simply adds an additional layer of complexity that makes comparisons very difficult.

In any event, do you really think that it’s possible to properly log engine performance on a static dyno, when the vehicle incorporates the ram-effect in the intake system? I don't. When I had my Z3MC dynoed at Evolve, I measured the pressure in front of the fan, and with reference to my earlier graph of pressure versus forward velocity, this equated to a velocity of just over 70mph, yet the car was dynoed in 3rd gear and was taken to a speed of 106mph at 7600rpm. As such, the TRUE performance of my car in “real world” conditions was not measured, because it was impossible to replicate and record the engine performance on the static dyno. As it happens, my Z3MC recorded the highest figures that Imran had seen for a S54 Z3MC on the Evolve dyno (that's been written about elsewhere), and that was with a few intake mods and modified silencers, and before I did my "ramcharger" mod for that car.

Well, I think that the fundamental question is whether ram effect is actually real for a normal road car at (relatively) normal road speeds isn’t it? That is why I have proposed some unambiguous tests that would help provide some pretty clear answers.

However, it raises another question – is there a downhill after the crest? If so then it might explain the spike in your graph – the car will accelerate downhill but the DashDyno won’t know it is because of the hill and will assume it is due to power.

I think this is irrelevant since I use the same road for all my runs. However, the road has a very slight uphill gradient followed by a plateau. I’d rather have a disadvantageous situation to measure.

But it is very relevant because we don’t have that part of the curve for the earlier runs! Plus your final curve has a very noticeable spike at that point, which is quite exceptional for an NA car.

3. Yes, the data recorded by the DashDyno can be exported into .csv format, which can be used in Excel. I’ve made an Excel template, which calculates a “rolling” Volumetric Efficiency (VE) from specific data that I log. I’ve found that by observing the graphical performance of VE has been most helpful in helping me to understand exactly how the air-intake “works” and how ram-effect can be improved with specific mods to increase dynamic pressure (i.e. ram effect).

And does your graph show power increasing proportional to speed as you suggest would be the case?

I haven’t said that power increases proportional to speed. I think that the best that can be achieved with ram effect is maintaining torque throughout the engine rev range rather than seeing the usual drop off after peak torque has occurred somewhere in the mid rev range. The Vanos and other Variable Valve Timing/Lift systems greatly assist in preventing torque loss, but improving the air supply with ram effect also assists.

In an earlier post you said:

the increase in forward speed produces an increase in static pressure within the air-intake system, which increases exponentially. Therefore, as the rpms increase, the increase in ram pressure compensates for the expected loss of pressure within the engine, thus the torque level is maintained right up to the redline.

Which I understood to be you saying that the increased forward speed forces more air into the intake and in turn allows the torque to be maintained at higher revs – same torque at higher revs means more power, that’s the way power is calculated, ergo, the increased power was dependent upon and proportional to the increased forward speed.

As I have said I am not an expert but from the little reading I have done I understand that what you need to utilise ram effect is increased static pressure (ie increased density) not dynamic pressure.
I found this post useful in understanding it: ‪http://forums.corvetteforum.com/c5-tech ... ffect.html

Yes, you do need to convert the dynamic pressure into an increase in static pressure and I’ve been recording static pressure throughout as part of my R&D and recording static pressure has been very useful in guiding me in the right direction.

Your previous post, to which I was responding you referred to “mods to increase dynamic pressure (ie ram effect)”, hence my question.

When in April? I will be there for the DN days right after the Easter weekend.

Yes, I’m booked too.

Cool, be good to meet up.

(Also you might need to have words with your friend about his line into the first right of the chicane leading into T13 ... he's giving loads of track and hence corner speed away)

You’ll be able to tell him yourself in April. :wink: I very much look forward to trying to keep up with you and learning the correct lines from you! :wink:

I very much doubt if there will be much keeping up needed, fingers crossed it will be my first trip in a new car and I daresay I will mainly be learning about the car.

that explains 8)
retirement is a little soon for me :roll:

I often wonder how you guys, who are supposed to be at work, have the time to participate on forums during the day time.... :wink:

I work for myself ...

 

[exdos]

Yes they do – chassis dynos can correct for barometric pressure (typically of course, I can’t speak for all).

This discussion is a bit of déjà vu of one I had about 6 years ago on the M Coupe forum. Back then, to put what I’d been saying about my mods to the test, it was arranged for me to visit the dyno at Powerstation (well-respected outfit in the Subaru world) where my S54 M Coupe was to run at the same time (within minutes of each other) as another forum member’s S50 M Coupe, and my car produced some remarkable figures on that dyno. There are videos of it if you’re interested. The upshot of that was, Imran at Evolve (well respected in the BMW scene) invited me over for free dyno runs, which were carried out a week later, and as I reported earlier, my Z3MC recorded the highest figures that Imran had seen for a S54 Z3MC.

Shortly after those trips onto the dyno, I bought the DashDyno and the “at the wheel” figures that I obtained from the DD were approximately 15% lower than the "at the flywheel" figures produced from the Powerstation and Evolve dynos. I had written extensively about the modifications which I’d done to my car so all those who were interested knew precisely what I’d done to achieve those performance enhancements, and none of my mods had cost me anything other than my time and the costs of a few consumables in making the modifications.

As far as I’m concerned, I have no further need to prove anything to anyone. I’ve publicly and openly proven that performance gains are possible with some pretty simple modding of the OEM parts.

Example 1: if you attached a parachute to your car and did your test run then DashDyno would show much worse results because it wouldn’t know about the increased resistance, but the engine would be producing the same power as before.

Example 2: if you put the rear on axle stands in a Ferris Bueller sort of way and ran it up through the gears I imagine that DashDyno would show fantastic results because it would have no load, but again the engine would still be exactly the same.

I can see how I could achieve worse results in real world conditions but not better ones unless I use incorrect data in the car ”profile”. I know I didn’t do that because the dyno run file produced by the DashDyno from each run shows the profile of the car at the top and the first thing I did was check I’d used the correct profile, which I had.

But at the flywheel is the accepted measure for comparison, not using it simply adds an additional layer of complexity that makes comparisons very difficult.

The data from my DashDyno is for nobody else except me. At the end of the day, if the DashDyno’s “at the wheels” figures are approximately 15% less than the “at the flywheel” figures produced by static dynos, how hard is it to compute the difference? It’s only like converting Pounds to Euros when we’re at The Ring and we can both do that.

Well, I think that the fundamental question is whether ram effect is actually real for a normal road car at (relatively) normal road speeds isn’t it? That is why I have proposed some unambiguous tests that would help provide some pretty clear answers.

As far as I’m concerned “normal road speeds” in a car like a Z3MC or a Z4MC is any speed that I will do on my trip to The Ring, including lapping the NS. I know ram effect works and I gave up long ago attempting to convince anyone that it can be harnessed and is very useful.

But it is very relevant because we don’t have that part of the curve for the earlier runs! Plus your final curve has a very noticeable spike at that point, which is quite exceptional for an NA car.

The spike that occurs at around 7500rpm is because I’d slightly lifted off the throttle, which partially closed the throttle bodies, so that the air entering the intake had converted dynamic pressure into an increase in static pressure within the intake system, so that when I reverted to WOT, the pressure of air entering the cylinders was close, if not equal, to the theoretical pressure created by the ram-effect, with the consequential increase in Volumetric Efficiency at that point. You have just remarked on the ram-effect in action! Yes it is exceptional, and that is the WHOLE POINT of what I’m trying to achieve. I’m trying to produce the ideal ram-intake.

If I’d continued accelerating from 3rd gear to 4th, whilst I changed gear, there would be another build up of static pressure, which would then be available for cylinder filling in 4th gear, which would be using lower revs, so the loss of pressure in the air-intake system will be less than occurs in lower gears. Do you get it yet?

Cool, be good to meet up.

Definitely!

I very much doubt if there will be much keeping up needed, fingers crossed it will be my first trip in a new car and I daresay I will mainly be learning about the car.

It will also be my first venture on track in my Z4MC too and very much looking forward to it.

 

[tertius]

The data from my DashDyno is for nobody else except me. At the end of the day, if the DashDyno’s “at the wheels” figures are approximately 15% less than the “at the flywheel” figures produced by static dynos, how hard is it to compute the difference? It’s only like converting Pounds to Euros when we’re at The Ring and we can both do that.

Well, earlier in the thread you spoke about commercialising your idea and using the data to support the product. I'm making suggestions that will, in my opinion, make your data more powerful and relevant to potential customers. If in fact you aren't interested in that, then I agree, no need to take my feedback at all.

 

[exdos]

Well, earlier in the thread you spoke about commercialising your idea and using the data to support the product. I'm making suggestions that will, in my opinion, make your data more powerful and relevant to potential customers. If in fact you aren't interested in that, then I agree, no need to take my feedback at all.

I do appreciate all constructive comments, including criticism, such as yours. It is helpful.

The fact remains though, that ram-effect will not be seen on a static dynamometer unless a variable speed fan, which is powerful enough to replicate the change in dynamic pressure that occurs with the increase in forward velocity. I'm unaware of any such dyno, although a wind tunnel (mega expensive) would do the trick perfectly. At the moment, it is still a "work in progress" and I am able to interpret my data in the format in which the DashDyno presents it. If and when the time comes that I wish to commercialise my invention, then I agree I need to produce data other than my own to support it. To that end, I think I would seek to enlist the help of a University with an automotive engineering faculty to provide independent assessment, that way, I would hope that any figures would be considered as absolutely correct.

 

[tertius]

Well, earlier in the thread you spoke about commercialising your idea and using the data to support the product. I'm making suggestions that will, in my opinion, make your data more powerful and relevant to potential customers. If in fact you aren't interested in that, then I agree, no need to take my feedback at all.

I do appreciate all constructive comments, including criticism, such as yours. It is helpful.

The fact remains though, that ram-effect will not be seen on a static dynamometer unless a variable speed fan, which is powerful enough to replicate the change in dynamic pressure that occurs with the increase in forward velocity. I'm unaware of any such dyno, although a wind tunnel (mega expensive) would do the trick perfectly. At the moment, it is still a "work in progress" and I am able to interpret my data in the format in which the DashDyno presents it. If and when the time comes that I wish to commercialise my invention, then I agree I need to produce data other than my own to support it. To that end, I think I would seek to enlist the help of a University with an automotive engineering faculty to provide independent assessment, that way, I would hope that any figures would be considered as absolutely correct.

I don't think I suggested using a chassis dyno to support your figures, only to give us flywheel power reference value. However, I did make some other suggestions that would be fairly* cheap* and easy* to try.

* valid for a broad interpretion of "fairly" and "cheap" and "easy" ...

 

[exdos]

I don't think I suggested using a chassis dyno to support your figures, only to give us flywheel power reference value. However, I did make some other suggestions that would be fairly* cheap* and easy* to try.

* valid for a broad interpretion of "fairly" and "cheap" and "easy" ...

At The Ring, we can hook up your Z4MC to my DashDyno and we can compare. And I won't charge you... :wink:

 

[original guvnor]

Exdos I know you and I have had our minor spats in the past, but in this instance I applaud your perseverance in this matter.

The world needs innovators, people who aren't prepared to follow others like sheep and who put their head above the parapet and risk ridicule for it. All power to your elbow for doing this. Columbus and the flat earth spring to mind.

I'd happily do a straight line test against your car as mine has recently been static dyno'd at 321bhp at 7300 rpm so I'd guess it will be near enough stock at 7900. It is also completely unmodified from factory as well.

Keep going on this

 

[tertius]

At The Ring, we can hook up your Z4MC to my DashDyno and we can compare. And I won't charge you... :wink:

Ha!

Subject to a small dependency* we certainly can, and it would give a useful extra data point.



* I don't actually have a Z4M at the moment though hopefully I will by then ...

 

[exdos]

Exdos I know you and I have had our minor spats in the past, but in this instance I applaud your perseverance in this matter.

We've both got broad shoulders. Water under the bridge.

The world needs innovators, people who aren't prepared to follow others like sheep and who put their head above the parapet and risk ridicule for it. All power to your elbow for doing this. Columbus and the flat earth spring to mind.

I've been blessed/cursed with a maverick personality so it's just normal for me. I take Oscar Wilde's view; "Whenever people agree with me, I always feel I must be wrong". :wink:

I'd happily do a straight line test against your car as mine has recently been static dyno'd at 321bhp at 7300 rpm so I'd guess it will be near enough stock at 7900. It is also completely unmodified from factory as well.

Where, when? Why don't you book for the DN event at The Ring?

Keep going on this

No doubt about that.

 

[jimmybell]

needless to say, this thread blew my mind. 8)