Are dyno numbers correct? Article on dyno's and how to understand them - Sport Auto

[Boss330]

The last issue of Sport Auto had the announced article on dyno testing.

The background for the article was their Supertest of the 991 Turbo S and the dyno numbers they published. In that test Sport Auto published a dyno tested crank HP (metric) number that was significantly higher than the factory stated numbers.

Porsche 911 Turbo S - 560hp
Sport Auto 911 Turbo S - 607hp (MAHA dyno)

That's a significant 8,4% higher, or 47hp more, than the factory issued numbers.

Sport Auto had a lot of feedback and comments on this and decided they needed to go deeper into the matter. They needed to find the answer to 4 questions:

• How does rolling road dynos work?
• What is important to be aware of when dyno testing?
• Where are the possible faults to find?
• Did the Porsche really have to much power?

I am not going to translate the entire article, but just point out some of the highlights, or main findings and their conclusions:

The first question on how rolling roads work can probably be answered by going to the manufacturers websites or reading online articles, so I won't use more time on that question here.

The second question is more interesting, as it goes into some of the issues some of us also discussed here a year or two ago. It's also interesting because in reality it means that most of the published dyno numbers, for cars like the F8x M's and F1x M's, in reality are useless and wrong...

The dyno operator in the article explains that they use on road test data (like air intake temps at 200km/h etc) to make sure that they get the most realistic test conditions to ensure realistic testing environment and test numbers.

On the MAHA dynos it's important to understand that the Prad (or Pwheel) isn't the same as wheel horsepower. Prad will allways be lower on a MAHA dyno than the wheel power on the road. The article explains that the "car and dyno is one combined unit, take the car away and the losses and mass is gone from the dyno. That's why they need to measure "Schleppleistung" (towing power) and add that to the Prad number"


Where are the faults to be found?

The dyno's biggest enemy is slip, especially on high powered turbo cars this is a factor that needs to be controlled.

The most significant fault source on modern turbo engines however is using the correction factors as per EU 80/1269, SAE or STD. Just as we have discussed on these forums before, using correction factors on these engines constitutes a "double dipping"! The modern turbo engines have the capability to self correct for different ambient conditions and altitude. The latest edition of the standards also state that if the engine has this feature, no correction is to be applied to the dyno number. The engine adapts boost, timing etc to cope with high or low temps, high or low altitude etc. On a NA engine the power is directly affected by these factors, but on a modern turbo engine boost is increased at higher altitude to maintain the stated power output for instance.

On the 911 Turbo S, using the correction added a whopping 47hp... When in fact no correction should have been added at all...

So to test these theories, Sport Auto took a 911 Turbo S to three different dynos in Germany (all MAHA).

The dynos where on three different altitudes (757m, 271m, 85m) and outside temperature varied from 15,3 deg C to 20,5 deg C

NO correction was added of course. The results was as follows:

Dyno 1 - MAHA in Haldenwang, 757m, 15,3 deg C

• Crank HP - 567hp @ 6340rpm
• Crank TQ - 764,8NM @ 2770rpm

Dyno 2 - "Müller prüft" in Backnang, 271m, 20,5 deg C

• Crank HP - 563,4hp @ 6565rpm
• Crank TQ - 764,3NM @ 2815pm

Dyno 3 - DTE Systems in Recklinghausen, 85m, 19 deg C

• Crank HP - 564,1hp @ 6550rpm
• Crank TQ - 758,5NM @ 3465rpm

So, on three different MAHA dynos, at different altitude and temps, the measured (or more correctly - calculated) crank HP varied by ONLY 3,6HP... The consistency between those numbers are VERY impressive.

It also answers the fourth and final question. No the Porsche does not have "too much" HP, in fact it is just between 3,4-7hp higher than factory numbers.

The interesting thing is that if correction factors had been added to all of these three dyno results, the end result would have been very inconsistent since the difference in altitude and temps would have meant that the correction factors would have varied between the three locations.

The most important lesson here is to NOT use correction factors when dynoing a modern turbo engine. If correction is added the result is false and comparisons between different dynos becomes even more meaningless than it allready is (unless you compare between MAHA dynos apparently).

One other lesson here is to pay close attention to the dyno numbers posted for cars like the F1x M's and F8x M's. If it shows that the numbers are corrected (SAE or STD) the numbers are wrong!!!
(unless tested at EXACTLY the SAE or STD prescribed conditions, since then applyed correction is 0).

That's also one of the main reason we see so many claims about "much higher than stated hp" for cars like the F8x. And why the difference between the S65 dyno numbers (which shall have correction added) and the S55 dyno numbers seem to be much higher than the adverticed difference.

Adding SAE or STD correction to a S55 dyno can inflate the numbers significantly, as the 911 Turbo S dyno showed (40-47hp added).

[CanAutM3]

Thanks for sharing and translating the important bits.

This article confirms what some of us have argued for quite a while.

On the MAHA dynos it's important to understand that the Prad (or Pwheel) isn't the same as wheel horsepower. Prad will allways be lower on a MAHA dyno than the wheel power on the road. The article explains that the "car and dyno is one combined unit, take the car away and the losses and mass is gone from the dyno. That's why they need to measure "Schleppleistung" (towing power) and add that to the Prad number"

I also like this passage, as it confirms what I have suspected and debated numerous times in the past

It also shows how the Maha is pretty accurate and repeatable for a chassis dyno for peak power. On the other hand, the shape of the power curve varies sensibly from one dyno to the other. Maybe an effect of the varying elevations. I would be curious to find out what the S55 number would be on the Maha without the correction factor .

[Boss330]

Originally Posted by CanAutM3

I also like this passage, as it confirms what I have suspected and debated numerous times in the past

I actually thought of you when I read that in the article

[BigMacSmallFries]

Glad you posted this for everyone to read. Was frustrated with forum members not understanding that turbo engines compensate for any elevation, humidity, and temp. Hence hp is exactly the same all the time when it's sweltering in the summer or you're driving in the winter.

Now, with that said do we have any MAHA dynos of the F8x M3 uncorrected? Still think it's around the 450HP mark. The 480hp or more thrown around these forums from time to time is way off. Thanks!!

[CaryTheLabelGuy]

Originally Posted by BigMacSmallFries

Glad you posted this for everyone to read. Was frustrated with forum members not understanding that MODERN turbo engines compensate for any elevation, humidity, and temp. Hence hp is exactly the same all the time when it's sweltering in the summer or you're driving in the winter.

Now, with that said do we have any MAHA dynos of the F8x M3 uncorrected? Still think it's around the 450HP mark. The 480hp or more thrown around these forums from time to time is way off. Thanks!!

There, fixed it for you.

And when I say modern, I mean within the last few years. Before that, weather and elevation played a huge part in how much power a turbo engine made. It is only recently that the ECU (DME) has taken complete control over boost and added or subtracted boost to keep the engine at a constant output regardless of elevation or weather. Electronic Wastegates certainly helped create the ability to offer such fine control over boost. Don't get me wrong, ECU controlled boost has been around for many many years, but controlling a mechanical, spring loaded and variably adjustable Wastegate Actuator using a boost solenoid via vacuume/boost pressure doesn't offer enough resolution to be able to adjust boost pressures so finely.

[Boss330]

Originally Posted by BigMacSmallFries

Glad you posted this for everyone to read. Was frustrated with forum members not understanding that turbo engines compensate for any elevation, humidity, and temp. Hence hp is exactly the same all the time when it's sweltering in the summer or you're driving in the winter.

Now, with that said do we have any MAHA dynos of the F8x M3 uncorrected? Still think it's around the 450HP mark. The 480hp or more thrown around these forums from time to time is way off. Thanks!!

The uncorrected HP is 425hp...

The same car did 415whp on a Dyno Dynamics roller...

[Blksnowflake]

Sticky worthy.

[SECOND2NONE]

i understand the modern engines compensating on stock cars. But im just not a believer that this also goes for tuned cars

if a tuner sets max boost at 25 psi to make 450whp. that same car in diffrent conditions would require the ecu to raise boost by another 3 psi to to make the same consistent power. how would this work if max boost is already set to 25 psi

[BMW M4 PWR]

Originally Posted by SECOND2NONE

i understand the modern engines compensating on stock cars. But im just not a believer that this also goes for tuned cars

if a tuner sets max boost at 25 psi to make 450whp. that same car in diffrent conditions would require the ecu to raise boost by another 3 psi to to make the same consistent power. how would this work if max boost is already set to 25 psi

Very good example, it's gets even worst when that hypothetical 25psi you set it at, represents the max efficiency the said turbos can push.. or the max fuel system limits too. That's the whole premise why tuners use dyno's to compare achievements and set them to a set CF like SAE. This article, or all the commentary we heard earlier from earlier members claiming essentially corrected dyno results mean nothing lol don't take the mechanical limitations into account. They're looking at it one dimensionally. The ONLY reason some "modern turbo motors" can make near identical power by self correcting themselves is by having power well beyond stock configuration built into them so that it can essentially raise the boost or fuel pressure or whatever it needs to in order to compensate. This is a very easy concept to understand. Hence, tests like the magazine tests illustrated above only make sense for essentially stock turbo cars with a lot of room for power built into them. Take that same Porsche put a GIAC flash, raise boost to 25psi taxing its stock turbos and fuel system to the max... THEN test it at sea level and once more at a dyno 5000ft ASL and look at the results That's when the whole "correction factors aren't needed because motors self correct BS goes right out the window..." But then again this is elementary common knowledge information I'm stating here, AGAIN, that any serious tuners & engine builders already have known for years... but what do they ALL know lol..

[Boss330]

Originally Posted by BMW M4 PWR

Very good example, it's gets even worst when that hypothetical 25psi you set it at, represents the max efficiency the said turbos can push.. or the max fuel system limits too. That's the whole premise why tuners use dyno's to compare achievements and set them to a set CF like SAE. This article, or all the commentary we heard earlier from earlier members claiming essentially corrected dyno results mean nothing lol don't take the mechanical limitations into account. They're looking at it one dimensionally. The ONLY reason some "modern turbo motors" can make near identical power by self correcting themselves is by having power well beyond stock configuration built into them so that it can essentially raise the boost or fuel pressure or whatever it needs to in order to compensate. This is a very easy concept to understand. Hence, tests like the magazine tests illustrated above only make sense for essentially stock turbo cars with a lot of room for power built into them. Take that same Porsche put a GIAC flash, raise boost to 25psi taxing its stock turbos and fuel system to the max... THEN test it at sea level and once more at a dyno 5000ft ASL and look at the results That's when the whole "correction factors aren't needed because motors self correct BS goes right out the window..." But then again this is elementary common knowledge information I'm stating here AGAIN (for the my last time) that any serious tuners & engine builders already have known for years...

Of course a system operating at it's very limit in "ideal" conditions doesn't have room to compensate anymore. There simply isn't room to add more boost (or rather the turbos can't deliver more air volume/mass even though the DME tries to increase boost).

But on most tunes, apart from the "all out" ones, there still is room for compensation. Tunes like a AC Schnitzer, Manhart Stg 1, Dinan, JB4 stg 1 etc doesn't normally take neither turbos or fuel systems to the max. Adding compensation like SAE or STD here is still just as flawed as it is on a stock engine...


And why does the tuners use SAE/STD correction on their stock runs as well? I bet that quite a lot of rolling road operators simply doesn't understand how these engines work and that adding correction basically makes the dyno results false... Because by adding correction on both before and after numbers (on a tune that maxes out the turbos), the before and after results simply aren't comparable...

[Boss330]

Originally Posted by SECOND2NONE

i understand the modern engines compensating on stock cars. But im just not a believer that this also goes for tuned cars

if a tuner sets max boost at 25 psi to make 450whp. that same car in diffrent conditions would require the ecu to raise boost by another 3 psi to to make the same consistent power. how would this work if max boost is already set to 25 psi

Remember that on a piggyback system the tune basically tricks the DME into believing that boost is lower than it actually is by providing a false signal to the DME (say 1,3V instead of 1,5V at 15PSI boost). The DME then adds boost to reach the correct boost level for the conditions. The DME now has a 1,5V signal and believes it has 15PSI of boost, but in reality it has 17PSI.

The same goes with the max boost level. If the DME is programmed to cut boost at 25PSI then it does that when it gets a 25PSI boost signal from the boost sensor. But as explained above, the piggyback alters the signals from the sensors. Let's say that 25PSI boost equals a 2V signal from the boost sensor to the DME. Without a piggyback the DME gets a 2V signal at 25PSI of boost. With the piggyback it gets perhaps 1,8V and believes that boost is just at 23PSI. The DME then adds boost until it sees a 2V signal. But that 2V signal really means 27PSI...

So all the DME self correction capabilities are still working in full even with the tune. But as pointed out by BMW M4 PWR, if the turbos are at max capacity on a cool day at sea level, there simply isn't room for adding boost at high altitude or high temps anymore.

[BMW M4 PWR]

Originally Posted by Boss330

So all the DME self correction capabilities are still working in full even with the tune. But as pointed out by BMW M4 PWR, if the turbos are at max capacity on a cool day at sea level, there simply isn't room for adding boost at high altitude or high temps anymore.

You gotta remember, (and this is the core of my point on why correction factor are always very applicable, and used) even if you're well within the turbo and fuel systems capacity allowing the DME to compensate. The motor is essentially raising the boost, fuel pressure and power on itself to spin the rollers by the identical uncorrected/raw exerted force in less than ideal conditions (versus say, better than ideal conditions) taxing itself much more to achieve that same power at the rollers. The correction factor the dyno's use illustrate this fact.

It doesn't matter if a stock M3 shows 400rwhp uncorrected at sea level and then the same car does 400rwhp uncorrected at 5000ft ASL. That cars motor, turbos, systems, etc are working MUCH harder to achieve that same number at altitude* The correction factors not only illustrates this, but more importantly, shows you what that same motor COULD mechanically do if hypothetically its DME gave itself that same high altitude tune at sea level. Which is why a motor that could self compensate to the degree of spinning the rollers with the same force at 5000ft ASL (or less than ideal conditions) will show a higher dynojet number* ---> AND that higher number is justified*** Because it took way more motor, turbos, intercooling, etc to mechanically achieve it.

A dyno corrected 607HP 991 TT Porsche which is "factory stock" means that motor IS genuinely *mechanically* capable of 607HP in ideal conditions even though on the uncorrected MAHA it could mechanically regulate its rear wheel power to its preprogrammed factory target - sans conditions.


Someone sticky this^^^ "Why correction factors are still important*" lol..

[CanAutM3]

Originally Posted by BMW M4 PWR

You gotta remember, (and this is the core of my point on why correction factor are always very applicable, and used) even if you're well within the turbo and fuel systems capacity allowing the DME to compensate. The motor is essentially raising the boost, fuel pressure and power on itself to spin the rollers by the identical uncorrected/raw exerted force in less than ideal conditions (versus say, better than ideal conditions) taxing itself much more to achieve that same power at the rollers. The correction factor the dyno's use illustrate this fact.

It doesn't matter if a stock M3 shows 400rwhp uncorrected at sea level and then the same car does 400rwhp uncorrected at 5000ft ASL. That cars motor, turbos, systems, etc are working MUCH harder to achieve that same number at altitude* The correction factors not only illustrates this, but more importantly, shows you what that same motor COULD mechanically do if hypothetically its DME gave itself that same high altitude tune at sea level. Which is why a motor that could self compensate to the degree of spinning the rollers with the same force at 5000ft ASL (or less than ideal conditions) will show a higher dynojet number* ---> AND that higher number is justified*** Because it took way more motor, turbos, intercooling, etc to mechanically achieve it.

A dyno corrected 607HP 991 TT Porsche which is "factory stock" means that motor IS genuinely *mechanically* capable of 607HP in ideal conditions even though on the uncorrected MAHA it could mechanically regulate its rear wheel power to it preprogrammed factory target - sans conditions.


Someone sticky this^^^ "Why correction factors are still important*" lol..

I think you are missing an important point of the thread. It is about the widespread misconception that the S55 is massively underrated from factory where many folk believe they have a 480~500hp car.

You are also missing another important point, is that the whp numbers obtained on the majority of chassis dynos cannot be used to compare absolute numbers from different engines tested on different dynos. They are good tuning tools to test the relative gains of mods done on a given drivetrain tested on a given dyno. Not more.

IIRC, the SAE standard specifies that engines should be tested in conditions that are as close as possible to the standard. The SAE standards specify, than when it is appropriate to use correction factors, corrections of no more than 3% should be applied to be "certified numbers". There is a big reason for that, it is the only way to obtain accurate and repeatable absolute power numbers.

Correction factors are appropriate when their use is well understood to make the right comparison and analysis. However, way too many dyno users and tuners use them inappropriately. Hence the massive confusion.

Also note that, at altitude, the engine itself is not working any harder to produce the same (ECU compensated) power. Obviously as it is producing the same power. Only the turbos are working harder.

As for your last paragraph, I am not sure I follow the logic. How would you know the true thermodynamic margin (what you call "mechanically capable")? For instance, if a stock engine tested in less than ideal conditions puts out 400whp SAE, you then put a tune on it and now gives you 450whp SAE, it certianly did not gain 50whp in "mechanical capability". Further, all this it is pointless, because it is not what the engine is making anyhow.

[BMW M4 PWR]

We are making many of the same points in different ways.. "you say tomato I say tomaato" otherwise if not then I agree to respectfully disagree* (And so does an ENTIRE industry which uses dynojets with correction factors for the same reasons I specified) lol...

[CanAutM3]

Originally Posted by SECOND2NONE

i understand the modern engines compensating on stock cars. But im just not a believer that this also goes for tuned cars

if a tuner sets max boost at 25 psi to make 450whp. that same car in diffrent conditions would require the ecu to raise boost by another 3 psi to to make the same consistent power. how would this work if max boost is already set to 25 psi

That is exactly the thing. As we have discussed in the other thread, most tunes only tap in the engine's thermodynamic margin to extract more power. As conditions worsen, there will come a point where the tune will not be able to provide any more power than stock. Correction factors will not tell you that or where that threshold is.

[CanAutM3]

Originally Posted by BMW M4 PWR

We are making many of the same points in different ways.. "you say tomato I say tomaato" otherwise if not then I agree to respectfully disagree* (And so does an ENTIRE industry which uses dynojets with correction factors for the same reasons I specified) lol...

That is the thing, a big portion of the "industry" does not have the competence to appropriately interpret dyno numbers .

[BMW M4 PWR]

Originally Posted by CanAutM3

That is the thing, a big portion of the "industry" does not have the competence to appropriately interpret dyno numbers .

The tuners, engine builders, dyno developers, and 100's of automotive engineers that have been and ARE currently using them clearly DO NOT have the competence* But fortunately we have YOU, are own resident engineer from Canada that can... lol

[CanAutM3]

Originally Posted by BMW M4 PWR

The tuners, engine builders, dyno developers, and 100's of automotive engineers that have been and ARE currently using them clearly DO NOT have the competence* But fortunately we have YOU, are own resident engineer from Canada that can... lol

No need to get arrogant and insulting like in the other thread. We know how that one ended .

http://f80.bimmerpost.com/forums/sho....php?t=1201355

BTW, not sure what my being Canadian has to do with the topic at hand .

If the tuners, engine builders, dyno developers and 100 of automotive engineers you refer to are inappropriately using correction factors, then they are either incompetent or dishonest. Period.

Truth be told, the SAE and DIN standard have just been recently adjusted for modern turbo engines. It is more than likely most of the "industry" has not yet fully caught up.

[BMW M4 PWR]

lol

[Vectors2final]

Was coming here expecting a MAHA the Master Race dyno pitch.

However...



I'll catch up eventually.

[Boss330]

Originally Posted by BMW M4 PWR

The tuners, engine builders, dyno developers, and 100's of automotive engineers that have been and ARE currently using them clearly DO NOT have the competence* But fortunately we have YOU, are own resident engineer from Canada that can... lol

Don't quite get your attitude here, but one of the points of the Sport Auto article was that even the German MAHA operators falsely applied correction because they didn't understand that it was wrong to do so.

In this post you apparently thought that the dynos had a "FIXED" correction factor... http://f80.bimmerpost.com/forums/sho...3#post19089163



You might mock our Canadian member, but he has the experts at MAHA and both SAE and EU standards on his side.

http://eur-lex.europa.eu/LexUriServ/...12:0046:EN:PDF

2.3 Power correction factors
By way of derogation from paragraph 5.1 of Annex V to UNECE Regulation No 85, when a turbo-charged engine is fitted with a system which allows compensating the ambient conditions temperature and altitude, at the request of the manufacturer, the correction factors α a or α d shall be set to the value of 1.

And the SAE standard is even clearer on NOT using correction factors. Do you disagree with the SAE std as well???



EU standards has a +/- 7% correction window

[Boss330]

Originally Posted by BMW M4 PWR

We are making many of the same points in different ways.. "you say tomato I say tomaato" otherwise if not then I agree to respectfully disagree* (And so does an ENTIRE industry which uses dynojets with correction factors for the same reasons I specified) lol...

If so, all of you are not up to speed on how these engines work and how the mesurement standards have been changed to allow for these engines to be tested without the correction factors...