Why do we measure with peak horspower? In reality it tells us what could be potentially the highest spike on the curve, which may only last for a 100 rpm. So if you could run you engine at that specific RPM you could use that HP all the time, but you can't. So wouldn't average or integrated HP be a more appropriate measure?

Think about it this way would you buy an amplifier with 100 Watts Peak or 100 Watts RMS? RMS is a calculated scientific measurement (basically the average power output of the amplifier). As where Peak is a couple guys watching a VU meter bounce around and guess at the max. Also the peak could be frequency dependent, as a dyno plot is. So you could get 100W peak at 1Khz, but only 20W at 50Hz.

By taking the average you can really predict how the motor will behave over a perioid of RPM. Or if you take the integrated HP (area under the curve) you will have the total HP expended over the entire run. This of course would help smaller engines with less peak HP and hi redlines.

 

Yep, ditto.

Screw peak, give me the meaty (and lumpy) mids!

 

It occured to me the other day while checking some of these dyno plots. You could have a spike that is completely unsuable. Besides there are so many variables to the dyno. You could run 165 at 8 am when it's cool, and run 155 when the temp hits 100F. 10 ponies right there! (just a guess don't quote me on that number) The average or integrated HP would see very little change because of atmoshperic conditions.


[Modified by itr1275, 3:24 PM 8/20/2001]

 

Jerold,
I agree with you...I had Lawrence help me tune the car for mid-range power at the expense of about 2 peak whp...now only if I can get the plots scanned in.
Austin

 

If they can put out the plots to excel we could calculate it easily. Add up every point, and divide by the number of data points. Average HP used. You could also pick sections of RPM to work with.

 

As Jack always says, area under the curve =

 

So you could choose the points by where the rpm will be when you shift. So for each gear you could have a horsepower rating. Or how about calculating thrust instead? What I'd really like to see is a program that will allow you to plug in your PEP files and your gear ratios that will give you a real world estimate of how your car will perform. It's possible, and there programs that do both separately, but none (that I know of) that will do both together.

Yeah, I guess it's a little more complicated than just an Excel file, but wouldn't that be cool?


[Modified by B18C5, 7:19 PM 8/20/2001]

 

Yep. Most important the part of the curve that your engine is in when under WOT.
6k-8,400

 

This brings us way back to the Torque and Horsepower debate that has been raging on for decades.

Horsepower is just the product of torque multiplied by (RPM divided by 5252) thus your crossing point at 5252 RPMs.

I would have to agree that the torque band must be flat and useable, hence the "area" underneath the curve will yield a stronger useable power.

Alot of things affects the HP rating, and in reality the, the car sitting on top of a rolling mass with air blown from recirculating air in the garage into the engine is not the type of thing we would encounter. If we could, it would be more acurrate to use a wind tunnel and a car strapped to a dynometer inside the tunnel with simulated airflow with hoods closed. But that just isn't too practical nor affordable.

On another note, it seems that the Horsepower war is more of a war about who has the highest number. It's like an example I told my father today. I told him he could have any car he wanted to race against my R. I told him he could have the most horsepower he wanted. So he said he wanted a 50,000 horsepower engine. Then I bet him that I would still beat him in the 1/4 because he will sit there and spin his wheels. Power is nothing without control. Horsepower is just a number.

Just a worthless thought.

 

A good description of this - and what makes a car go - can be found here.

Peak horsepower DOES matter - because it's a reflection of a flat TORQUE CURVE, and torque (at the wheels) is what REALLY matters. Here's why.

What makes a car accelerate is torque AT THE WHEELS. This equals engine torque (at the crank), less any drivetrain losses, times gearing. So torque should be all that matters, right? Well, not exactly.

If a car has a relatively flat torque curve - torque doesn't drop off a whole lot before redline, thanks to say a VTEC cam - and it can extend that torque high in the revband - then it will be able to use that torque at higher revs. It will have a gearing advantage over a car whose torque curve drops off sooner and which would have to upshift sooner. Because horsepower equals torque times revs, horsepower is a reflection of that gearing advantage. And yes, we're talking PEAK horsepower here; flat torque curve means horsepower rises (is not flat).

Check out the link above for a more detailed description.

 

What I see is guys all the time with a gain of "2 WHP" (Peak) after an air cleaner install. My point is, what conditions changed during the 10 mins it took to change the filter? 2 HP is nothing, on a test like this (~1% change), what is the margin of error or absolute error on this test? This is where I think a new method of measurement could be effective.

What I was pointing out is the current method of measurement is a bit screwy. Check out http://www.dynospotracing.com/. Some of the plots have a 5-10 hp Spike near the end of the curve (ie http://www.dynospotracing.com/s2000s.htm). So what good is that? You can only use it from 8500-9000 RPM? Pretty small width if you ask me. Does that 10 HP really exist? Is it usable? The usable peak HP is probably around 279 HP not 289 HP on this plot.

By taking a larger sample (area under the curve or avg) it will remove all the small peaks and valleys.

If someone has data from some dyno runs in excell, TXT, CSV, etc... I would be greatful to get it. I would like to play with it. Also I think running all gears from 0-Top Speed would be useful info too. That way you know you can accelarate faster in which gear given the RPM. You could essentially run a vitural 1/4 mile run.

Just some thought flying through my head..... Most go right through.

 

Jerold,
If you look at the photos in my signature you can view the intake comparison, which has these numbers available for you...
Austin

 

Do you have a soft copy? It would save some typing.

 

Guess I'll throw in my 2 cents here. I guess the reason for measuring peak horsepower can be a little more useful in helping with shift points dring racing. If you look at your power curve as it starts to fall off at a certain RPM point, there really is no sense in taking it past that point when racing. If you can time your shift in conjuction with your power curve, it should theoretically drop you right back in the upswing (for lack of better terms)of your power curve. Better that then to shift to far beyond and have it drop you back at the top of the curve and immediatly have the curve fall off again right?

Can you see what I'm saying? By the way, this is just a thought on my part. By no means am I an expert or anything (so no bashing OK )

James

 

That's close but not quite right. You want to shift at the point in your powerband where you'll be making as much or more power after you shift as you would by continuing to rev up the engine. Our cars are a good example, peak is at 7800rpm, but you want to shift at redline for maximum acceleration. Shifting at 7800 will drop you down into an area of the powerband that is making less power than the engine makes from 7800-8400rpm. By shortshifting, not only are you losing power to the wheels but you are also shortening your time spent in the lower gears where torquemultiplication is highest.

The only way that I know to determine the right shift points is to dyno the car and go off of the chart.

 

That's close but not quite right. You want to shift at the point in your powerband where you'll be making as much or more torque at the wheels after the upshift as you would by continuing to rev up the engine in the lower gear - or, if there is no such point, you want to shift at redline.

You can calculate torque at the wheels by multiplying torque by the gearing and assuming that drivetrain losses are a constant. For example, if you are upshifting from a gear with ratio 1.500 to a gear with ratio 1.200, you would want to upshift at the point where the engine torque in the higher gear is 25 percent higher (1.500/1.200=1.25) than the engine torque in the lower gear.

As it turns out, that's correct. The actual value is either at redline, or close to it, so that should be your target shift point.



[Modified by nsxtcjr, 3:47 PM 8/21/2001]

 

So say for my car, torque peak is at 6000-6500 rpm. HP peak is at about 8200. Are you saying that I should shift at 6000-6500 RPM???

 

I think what they are really trying to say, and I could be wrong is want to be as close as possible the peak HP point. To do that you will have to go past that, shift and then get back to it. ie, your average RPM is at your peak HP.

My point was saying that a small spike won't do squat because you can't live in such a small bandwidth of RPMs.

 

i'm sure everyone here realizes that people will always quote peak horsepower numbers. i'm a big "area under the curve" proponent but you're never going to convert people over to use an area figure when talking about their car (not that you're trying to convert anyone...). it makes me thing of trying to convert the u.s. to the metric system. i can't imagine people saying, "dude! i just dyno'd at 13,778!!!
anyway, my 2¢ though i may have gone kinda off-topic...

 

I disagree. At high rpm, power is what matters. Taking our car as an example again, after 7800 rpm torque drops off to less than than what it was before the VTEC crossover. In fact it drops to less than what it is making just after the VTEC crossover too. By shifting at redline in a Type R you are making less torque at the wheels before you shift than you are after you shift.

The car will still keep accelerating harder for a couple hundred more rpm after the torque peak though for the simple reason that torque is dropping slower than rpms are increasing. This leaves you with a higher number in the (torque*rpm) portion of the hp equation which will make a higher hp value after you divide the total by 5252. You can feel this when you are driving the car as well.

Any talk about torque without taking into account how fast it is being applied (which is horsepower by definition) is pointless IMO.

If anyone thinks that hp doesn't mean anything because it's calculated and not measured, think about this: If hp doesn't mean anything, why is the Type R so much faster than an LS? They both have the same 130 lb/ft of torque. The gearing on the two cars is not different enough to account for a nearly 2 second difference in 1/4 times either. So why?

The answer is simple: because it revs higher and makes more horsepower.

 

what newt said.

i cringe everytime i see people arguing to shift based on torque...ummm no, that does not compute. it's power that creates acceleration, not brute force (torque in this case.) if it were merely force, big beefy weightlifters would be dunking like jordan from sheer big thigh syndrome. or something like that...

 

Do you realize that your argument on shifting for maximizing torque at the wheels is exactly the same thing as maximizing the engine's power output? That is the definition of power.

You'll find in your calculations that the maximum torque at the wheels is in the gear that puts the engine at an RPM that provides maximum power.



[Modified by Gansan, 3:23 PM 8/21/2001]

 

Torque vs HP.

So where do we shift, max torque or max HP? Take a look at his then tell me where is max torque and max HP?
http://albums.photopoint.com/j/ViewP...p=52628574&f=0

I will take not only max HP but past that. I want to straddle the highest point in the power curve.


[Modified by itr1275, 3:54 PM 8/21/2001]

 

Here is something else to throw into the pile Effective Torque(seen as TE on spreedsheet below)...see the plot below, which is under my intake comparison. When you shift you want the change in rpm to place you in an area where your Effective Torque is higher than where you were before you shifted...correct?
Austin




[Modified by Austin, 4:04 PM 8/21/2001]

 

ITR1275, I am not talking about peak torque or horsepower. I am talking about the gear that puts the engine at the highest power at a given speed. You are right in trying to straddle the highest part of the curve.

Austin: As for "Effective Torque", I am not familiar with that term. If it means the torque at the wheels, then the data in the plot is not sufficient to determine the shift point. It is more likely the raw data that the dyno measures before calculating the engine's power and torque.

You need to know the gear ratios, which will let you know what the effective torque will be after the shift. You cannot just look at the effective torque in the dyno run, which represents only one gear. You need to compare the current gear with the next one up.