MyE28.com DEV

T_C_D wrote:Ken,

The fact is that your car doesn't drive around with SAE corrected HP and therefore all your mention of it in a comparison manner for reliability, clutch capacity, etc isn't relevant. What was the uncorrected rwtq? 440/450ish?

Todd

Let's set my car aside for the moment, Todd. The vast majority of the people on this forum, and those that do FI in particular, live at low elevations. That being said, being able to make a well-reasoned determination of what kind of power outputs will work and how that influences component selection has a great deal of relevance.

Power losses at altitude are a fact of life, but that environmental constraint is a level playing field. Conversely, bring one of your builds up to Mile High and do the pulls. I can assure you that the printouts will indicate far less power than you got in the Midwest or on the Eastern Seaboard. Does this mean the components you used are incorrect, or that something's changed where reliability is involved?

Once again, the SAE corrections are applied to allow meaningful comparisons between tests conducted under differing conditions.

Back to my motor. Uncorrected power at the rollers with boost at 7.25 is somewhere around 465.

For your reading enjoyment, you can bring up the not2fast turbo calculator and input the relevant variables for my motor. Bring up the model and PM me for the numbers you will need. That model generates a fwhp value of 578 hp. This with an altitude input of 5328' ASL. Reset that altitude to zero (sea level) and the fwhp # is 663. No other changes.

My experience with the n2f model is it tends to be a bit optimistic, but the computational errors seem to be consistent over a large number of simulations. So not using a rubber yardstick here.

As to what happens with the outputs at lower elevations. While the car lives at a bit above 6000'ASL, it has been driven at sea level--the Cali Central Valley to be specific. I am here to tell you that at sea level it will rip off your bawlz and use them for castanets. This was with boost @ 7.25. Full boost was simply stupid.

Looking at it another way, consider absolute manifold pressure. At my altitude, atmospheric pressure is generally about 83 kpa or 12 psi. If I install a 1 bar/14.5 psi spring in the wastegate I get about 180 kpa manifold absolute pressure. Its making about 14.5 psi boost, of which 2.x psi are consumed just to get to what would be atmospheric pressure at sea level. Considering that, you could say the actual, corrected boost pressure is 11.6 psi. I like that better than adjusting the actual measured output. That would knock 7.5 psi down to about 5 actual psi at altitude, which would account for a pretty substantial difference when running the car at lower altitudes.

If thats correct, the affect diminishes as boost increases. Its a flat 2.x psi tax for me, which is a smaller and smaller percentage of manifold absolute pressure as boost goes up.

Wait a minute. If the wastegate reads manifold pressure (absolute) against (constant) spring pressure, it should regulate to the same manifold pressure, right? Same for closed loop boost control, unless it explicitly has atmospheric correction.

Manifold pressure x volumetric efficiency (as a surrogate for tq, ignoring timing and AFR) shouldn't be affected, then, right? The only difference I can think of should be in MAT due to the lowered compression ratio for the same mass flow.

One side of the diaphragm sees atmospheric pressure, the other sees boost pressure plus atmospheric pressure. If you run 1 bar of boost on top of .83 bar atmospheric pressure, you get 1.8 bar in the manifold. At sea level you would have an additional .2 bar of pressure holding the WG diaphragm closed in addition to the fixed amount of spring pressure. Increase atmospheric pressure to 1 bar and you get the .2 bar free.

I'd love to get some datalogs on the subject. Hell of a drive though.

Ken H. wrote: Once again, the SAE corrections are applied to allow meaningful comparisons between tests conducted under differing conditions.

Back to my motor. Uncorrected power at the rollers with boost at 7.25 is somewhere around 465.

Thanks. 465rwhp is the the number you should reference when talking about the reliability of your engine and all of it's drivetrain components. That is the amount of power that is applied to the ground at your home base. A corrected number is irrelevant.

Damn this is good shit!

My previous major (though unfinished) was EE. Hearty conversations/discussions inspire me to change to ME with an emphais in Thermodynamics.

\carry on & thanks

T_C_D wrote:

Ken H. wrote: Once again, the SAE corrections are applied to allow meaningful comparisons between tests conducted under differing conditions.

Back to my motor. Uncorrected power at the rollers with boost at 7.25 is somewhere around 465.

Thanks. 465rwhp is the the number you should reference when talking about the reliability of your engine and all of it's drivetrain components. That is the amount of power that is applied to the ground at your home base. A corrected number is irrelevant.

Then what happens when the sled is taken down to the Cali Central Valley--altitude 6 ft ASL?

Todd, Keep in mind that the thing has Stage 2 boost as well--this measured @ 14.95 psi--as close to 15 psi as we could reliably set.
It was built to be happy running at sea level atmospherics; no compensation factors needed.

Ken H. wrote: Then what happens when the sled is taken down to the Cali Central Valley--altitude 6 ft ASL?

Put it on a Dynojet!

Bahhhhhh...my Tiger will kick its ass no matter the elevation.

That is, once its back together again.

T_C_D wrote:

Ken H. wrote: Then what happens when the sled is taken down to the Cali Central Valley--altitude 6 ft ASL?

Put it on a Dynojet!

You're sayng that the eddy current dyno at Mile High Performance isn't accurate? Jay Kidwell, MHP's owner and John Force might take exception to that. JF uses MHP's dyno to dial things in when they come up to Bandimere for the MileHigh Nationals.

Ken H. wrote:Then what happens when the sled is taken down to the Cali Central Valley--altitude 6 ft ASL?

Todd, Keep in mind that the thing has Stage 2 boost as well--this measured @ 14.95 psi--as close to 15 psi as we could reliably set.
It was built to be happy running at sea level atmospherics; no compensation factors needed.

I think the point is that BMEP or peak rwtq is a much better metric for predicting long term reliability of the motor and other drivetrain components than boost pressure alone.

A stock s38 might make 400 rwtq at 10psi, but a stock m30 might need 18psi to get there. Given equal rwtq figures, either setup will have a similar peak BMEP, put similar stresses on the drivetrain components, and have similar longevity expectations despite the large difference in manifold pressure.

Jeremy

Ken H. wrote: Then what happens when the sled is taken down to the Cali Central Valley--altitude 6 ft ASL?

Todd, Keep in mind that the thing has Stage 2 boost as well--this measured @ 14.95 psi--as close to 15 psi as we could reliably set.
It was built to be happy running at sea level atmospherics; no compensation factors needed.

Anything could happen. That's the point. IF you have enough fuel pump/injector, and IF you have enough clutch, and IF you have enough turbo, and IF the motor/tune/fuel octane supports the additional combustion pressure for the engine to run without detonation, you will make more power.

Do a pull at your local dyno and increase the boost pressure by an amount equal to the difference in atmospheric pressure between where you live and sea level. That is what it would make at sea level using your "normal" boost setting.

Ken H. wrote:

T_C_D wrote:

Ken H. wrote: Then what happens when the sled is taken down to the Cali Central Valley--altitude 6 ft ASL?

Put it on a Dynojet!

You're sayng that the eddy current dyno at Mile High Performance isn't accurate? Jay Kidwell, MHP's owner and John Force might take exception to that. JF uses MHP's dyno to dial things in when they come up to Bandimere for the MileHigh Nationals.

WTF , I said dynojet because it's the most common and the most useful for comparisons.

Jeremy wrote:I think the point is that BMEP or peak rwtq is a much better metric for predicting long term reliability of the motor and other drivetrain components than boost pressure alone.

A stock s38 might make 400 rwtq at 10psi, but a stock m30 might need 18psi to get there. Given equal rwtq figures, either setup will have a similar peak BMEP, put similar stresses on the drivetrain components, and have similar longevity expectations despite the large difference in manifold pressure.

Jeremy

Good point about BMEP, Jeremy. FWIW, 400 psi BMEP seems to be the "line in the sand." as far as FI motors being feasible from a build perspective.
Consider the following formula:
HP = (Ap * BMEP * Sp) / 132,000
Where
Ap = piston area in square inches
BMEP = Brake Mean Effective Pressure in psi
Sp = average piston speed in feet per minute
132,000 = a constant.
Using a stock displacement M30,
Ap = 61.82 sq. in
Sp = 3943 ft/min
86 mm(3.38 in stroke) @ 7000 rpm.
2 x 3.38 x 7000 = 47320. 47320 / 12 =3943 fpm
BMEP taken @ 1.5 bar (22 psi) = 398.5 psi
(61.82 *398.5 * 3943) /132,000
97,145,081 / 132,000
= 735.94 defined HP ceiling @ 1.5 bar or 22 psi.

T_C_D wrote:Put it on a Dynojet!

You're sayng that the eddy current dyno at Mile High Performance isn't accurate? Jay Kidwell, MHP's owner and John Force might take exception to that. JF uses MHP's dyno to dial things in when they come up to Bandimere for the MileHigh Nationals.[/quote]

WTF , I said dynojet because it's the most common and the most useful for comparisons.[/quote]

Sorry, Todd. The people around here who are into serious drags don't put a lot of credence into DynoJet or Dynapack numbers--far too much room for error coming from inertial overrruns. Hence the reliance on the eddy current chassis machines.
But I understand your point, namely it might be interesting to get a definitive set of pulls done at sea level at a shop with a properly-calibrated machine.

Duke wrote:Bahhhhhh...my Tiger will kick its ass no matter the elevation.

That is, once its back together again.

Provided it lives long enough to get there in the first place.

Ken H. wrote:
Sorry, Todd. The people around here who are into serious drags don't put a lot of credence into DynoJet or Dynapack numbers--far too much room for error coming from inertial overrruns. Hence the reliance on the eddy current chassis machines.
But I understand your point, namely it might be interesting to get a definitive set of pulls done at sea level at a shop with a properly-calibrated machine.

I have dynojet files for every car I have owned and maybe 15 customer cars dating from 2003 all on dynojet. The comparisons I am able to make are very helpful.

When I want to know how quick/fast the car actually is, I take it to the dragstrip. The dragstrip MPH is the definitive indicator of true power.

T_C_D wrote:I have dynojet files for every car I have owned and maybe 15 customer cars dating from 2003 all on dynojet. The comparisons I am able to make are very helpful.

When I want to know how quick/fast the car actually is, I take it to the dragstrip. The dragstrip MPH is the definitive indicator of true power.

I suppose if you know the particular machine's quirks it helps a lot. In my case, I rely on Jay K's experience with the eddy current machines.
While I've put the Hammer thru the lights on just one evening, I don't necessarily put a whole lot of emphasis on quarter-miles being a definitive indicator of what I'm looking for in terms of over-the-road performance.
It doesn't necessarily provide max hp or other measures, but I look at how the car has done on long-distance cross-country usage. I have a very good idea of what the top end is and how it responds in the 3000-5500 rpm power band. (Nicely.)
I keep reminding people, this thing was built as an endurance machine, not a dragster. I want it to be able to get out of its own way, but I expect it to run for 12-14 hours at a stretch without breaking a sweat. This it will do.

Ken H. wrote:

T_C_D wrote:I have dynojet files for every car I have owned and maybe 15 customer cars dating from 2003 all on dynojet. The comparisons I am able to make are very helpful.

When I want to know how quick/fast the car actually is, I take it to the dragstrip. The dragstrip MPH is the definitive indicator of true power.

I suppose if you know the particular machine's quirks it helps a lot. In my case, I rely on Jay K's experience with the eddy current machines.
While I've put the Hammer thru the lights on just one evening, I don't necessarily put a whole lot of emphasis on quarter-miles being a definitive indicator of what I'm looking for in terms of over-the-road performance.
It doesn't necessarily provide max hp or other measures, but I look at how the car has done on long-distance cross-country usage. I have a very good idea of what the top end is and how it responds in the 3000-5500 rpm power band. (Nicely.)
I keep reminding people, this thing was built as an endurance machine, not a dragster. I want it to be able to get out of its own way, but I expect it to run for 12-14 hours at a stretch without breaking a sweat. This it will do.

Times are NOT a good measure of a cars HP. MPH is. Amazingly my cars are built as street cars but are also able to run at a road course and a drag strip and drive wherever I want. Any street car in good condition can drive 12-14hrs. I am not sure what you have done to yours that distinguishes it from every other car on this message board in that respect?

I drag race my street car not street drive my drag car. Get it?

T_C_D wrote:Any street car in good condition can drive 12-14hrs. I am not sure what you have done to yours that distinguishes it from every other car on this message board in that respect?

I drag race my street car not street drive my drag car. Get it?

Thats the truth. Would be far too much work if it were the other way around. I'll be daily driving mine again as soon as I get around to fixing the massive oil pan leak. Not a boost related failure there, BTW. I believe this short block came with a factory original gasket.

turbodan wrote:I'll be daily driving mine again as soon as I get around to fixing the massive oil pan leak. Not a boost related failure there, BTW. I believe this short block came with a factory original gasket.

I cannot wait to finish the e28. I am planning a good road trip in June/July. NY to TO to Milwaukee (via the UP) to NC and back to NY.

Ken H. wrote:

mooseheadm5 wrote: What correction methods were being used?
The SAE also disallows corrections greater than ±7%.

Using the SAE J1399Aug04 formula, the correction is:

What this all says is dyno readings done on turbo motors at altitude will show less of a correction to a stipulated sea level power level.
But SAE is silent on the topic.

I think you mean 1349 (1399 is a spec for an electronic tach.)

Seems like the SAE is pretty clear on the subject. 5.5.2 says no correction for absolute pressure controlled boosted engines, such if one were running boost control that was referenced to absolute pressure vs. one referenced to atmospheric. I don't know if you are running such a control on your or not, but if you are not, then you would be better off doing Dan's correction, which is to simply turn the boost up a tad. If the ambient air pressure is X psi less than it is at sea level, to find out what the car ought to make at Y psi at sea level, crank it up to X + Y psi and go for it. That will get you much closer than assuming it is down a certain percentage of power because the compressor inlet is at a lower pressure than sea level. From there, you could run your own calculations based on your compressor map to see what the effect of the change in pressure is on the efficiency of the compressor at your boost level. That would be far more accurate than any SAE correction intended for NA motors because it would be a specific correction based on how environmental changed affect your compressor. It wouldn't be perfect (because there are then issues with exhaust pressure changes as well) but it would probably be a heck of a lot closer than the SAE spec.

5.5.3 also specifically says no more than 3% for air conditions and 3% for fuel conditions, so 11% is thrown out (especially since most of that 11% is for air) and your motor is simply listed as being tested under non-standard conditions. Attempts to standardize those conditions using these formulae would not be allowed by the SAE for power rating.

In essence, if Ford tried to advertize their turbo cars' power ratings based on dyno charts done at Mile High and corrected with these SAE formulae, the SAE would throw a fit. If they tried the same thing for NA and the air correction was over 3%, same deal. If they advertized that the power in Denver was what they actually measured in Denver, the SAE would not care. If they advertized that their motor made power measured at 500 feet but corrected to sea level with a Cf of less than 3% for air and less than3% for fuel, the SAE again would not care.

Also this:

Paul, I follow your reasoning and that of the SAE (I guess), but let me rephrase my question.
I have been unable to find any formal methodology for determining correction factors pertaining to engines using forced induction. Or should I say, a methodology endorsed by a body such as the SAE.
The air correction formulae I have found on various websites seem to make sense. Whether these came from various tuner's experience or had guidance from professional engineers or academics, I can't say.
What I have found has come from websites with a strong focus on small-displacement imports--Mazda, Honda, Toyota--where the owners are looking for major power increases.
What I have found seems to be both internally consistent and reasonable in terms of the calculation results.

But I remain puzzled that this topic, FI corrections for altitude, seems to be terra incognita.
I would think that somewhere in the literature related to FI piston-engined aircraft there might be some discussion. Specifically the issue of de-rating of powerplants at specified altitudes. But I have no sources.
Considering the case that use of turbocharging is becoming increasingly frequent as part of the USG's pressing for more efficient auto power plants, I'm surprised that an engineering body such as the SAE hasn't published some kind of papers on the topic.

I think the problem, and the likely reason that the SAE doesn't provide a one size correction for FI, is that there are more variables for FI engines than there are for NA. It would seem that a large portion of any correction would be dependent upon the compressor used on the motor and the pressure ratios. Since the SAE formulas given don't even address compressor efficiencies at the different flow rates, it would seem they are not a good model to use. As noted in the spec, an engine that references its boost pressure to absolute pressure (which is quite likely in the case of piston engined aircraft) do not experience the same negative effects due to ambient pressure changes. What definitely does change with altitude is the pressure ratio. Those corrections would have to be made for a specific engine/compressor/turbine combination.

Ken H. wrote: An example on corrections:
1. Sea level air pressure =14.7 psi. Air pressure at the Mile High shop, 5328'. = 12.08 psi. difference =17.82%. This number 12.08 psi may change if a cold front comes through or we have a summer high pressure cell, but the delta will be small.
2. Relative Humidity. Assumed to be 77 deg F and 20% r.h. for the moment. This creates about 6 millibars of additional air pressure due to drier air being more dense. Air pressure @ 5328" = ~812 millibars. 6/812 = .0074, or about 3/4 of 1% density difference.
3. Air temperature. Assume 72 deg F. The SAE "algorithm number" is 59 deg. F. so we are 13 degrees warmer. 13 deg F = ~7.22 deg Celsius. A commonly used rule of thumb is 1% power loss for every 7 deg C. temperature rise. 7.22/7.00 = 1.0314. So a loss of about 1.0314%.

Ken H. wrote:... Add back the 2.37% in the above and we have 8.84 +2.37 = 11.21% correction.
What this all says is dyno readings done on turbo motors at altitude will show less of a correction to a stipulated sea level power level.

I found more info on FI corrections as well. Using a bunch of assumptions (such as a linear relationship between boost pressure and power, which we know is not strictly true) we can use this formula (which pretty much matches what you found here):

C_f=(P_ref+P_b)/[P_dry (1+P_b/P_atm ) ]
P_ref=Dry air absolute pressure
P_b=Boost pressure
P_dry=Dry air partial pressure
P_atm=Total absolute air pressure

Plugging in the numbers from the first quote gives a C_f of 14.5%. Plugging in your max boost at the same pressures gets you 11.2%. Of course, these are not as rigorously proven out as the SAE 1349 for NA, but it is a start. Your 465hp uncorrected at 7.5psi would then correct to 532, not 567. Your max boost number correction would be even further off from what the printout shows. If you are running an absolute pressure referenced boost controller rather than the more common atmospheric reference, there would be very little correction (though I am guessing you are probably not doing that.) Absolute pressure boost control is available for Megasquirt and other standalones, so for future readers make sure you know what you have when you do your math.

As you can see, the C_f is lower when boost pressures are higher, which confirms what Dan says. I'd say the above calcs would be a much approximation for FI than using the SAE 1349 corrections.

The numbers your car posts are impressive either way.

The approach of adding boost to reach a desired MAP level has some appeal insofar as it is straightforward.

To play with an example, I want to run my motor at 15 psig. I am in Denver with an ambient air pressure of 12.08 psi. This is 2.62 psi less than sea level's 14.7. If my stipluated max boost number is going to be 15 psi, I need to be delivering 17.32 psig into the plenum (15+ 2.62). Think absolute pressures here: 12.08+2.62+15.00 = 29.7 psia

But I am going to lose some pressurization of my incoming charge as it moves from the compressor to the IC and thence to the plenum. This is inevitable due to (a) charge cooling, (b) frictional losses in the piping, (c) agitation of the air mass as it moves over the cooling structures [turbulators] within the IC core.

I want to keep these losses through the charge cooling system to something less than 10% of my boost pressure at the compressor outlet. A well-designed system can do this with a bit of noodling. So my desired 17.32 psig at the butterflies is calling for 17.32 divided by 90%, or 19.24 psi in the air mass as it exits the compressor. 19.24 +12.08 atmo @ Denver = 31.32. Divide 31.32 by the Denver atmo of 12.08 and we are actually looking for a Pressure Ratio of a hair over 2.59. This ignores the loss of pressure in the divisor due to restrictions of the air filter--typically anywhere from .5 to .8 psi.

Where the question arises is what happens when you go from Denver down to sea level. Ambient pressure goes up, and with it, Pr. Using the examples numbers, we get (19.24+12.08 +2.62), or 33.94. For a divisor, we have (12.08 +2.62) or 14.7. Pr is thus 2.30. Not an issue, one would think, but if the boost levels are pushing the design envelope at sea level, the inevitable higher Pr at altitude may become a problem.

Paul, I excerpted this from one of my earlier posts in this thread:.
"A turbo applies its compression on the air in the compressor, having been taken in at ambient pressure. So pressure correction for FI is taken as
(ambient psi + boost psig) / (sea level psi + boost psig)
Ambient pressure = 12.08. Boost = 14.95 psig. Sea level ambient air pressure = 14.7 psig.
I Googled horsepower corrections altitude and found this website: http://www.mazda3forums.com One of the posters had a really good succinct description of the formulas we’ve been hacking at. “Ah, gestalt!”

(12.08+14.95) /(14.7+14.95). 27.03 / 29.65 = .9116. So the air density correction values should be ~8.84% lower than what an equivalent sea level pull would show (ignoring any temp and RH effects). Add back the 2.37% in the above and we have 8.84 +2.37 = 11.21% correction.
What this all says is dyno readings done on turbo motors at altitude will show less of a correction to a stipulated sea level power level."
It may just be happenstance, but I find it interesting that the SAE correction you show isn't terribly different from what came indirectly from one of the "less formal" websites.

That being said, if doing this rigorous analysis demonstrates that my horsepower levels are less than what I have thought them to be, all I can say is I stand corrected. A bit chagrined? Sure. But knowing this doesn't make That Thing perform any differently.

I quoted part of that, too

The correction I found wasn't an SAE correction, it is just working back from air density ratios. As mentioned, it makes assumptions of linearity, but doesn't account for compressor efficiency changes, so those effects could end up being a wash.