Why cant my engine rev to 7k rpm?

[camlob]

So I have a 993 with a 4l P&C, PMO ITBs 50mm, DC 43 -112, LWF, AASCO valvetrain, Pauter rods, solid lifters, TEC3r ecu, 1 3/4 headers etc.

My tuner didnt follow John D's cam timing specs which were: Set lift at overlap to 3.2 to 3.4mm. My tuner set it at TDC intake 111 degrees and exhaust 115 degrees.

Sorry I dont know if the tuner went with this cam timing. My problem is the car only made 295rwhp on a dynapack dyno. For comparisons sake, my brothers 997 GT3 RS made 350rwhp.

We just finished tuning today and it cant make hp beyond 6.7k rpm. At 7k power drops. I bought the pauter rods to rev up to 7.5k. So I am wondering why the cut-off in hp.

I am happy with the hp increase from 267rwhp from my 3.6 because of the increase in torque. But I am dismayed on why I cant rev higher. That is the reason why I bought pauter rods.

Is it maybe I should follow John D cam timing specs? The car is still in the shop so I can have them time it to Johns specs.

Should I give it a try? Or is the cam just too mild for a 4l? I visited a 993 in Singapore and it had a 3.8l with ITBs and had a RSR profile cam. The increase in displacement in my case could make my even mild cam milder.

John told me to try adjusting the intake to 115 degrees. But my tuner said that is useless since it wont make a impact. He wants me to get a 305 duration cam. John says to first try his cam timing specs.

I am pleased the way the car drives. Just want to make the most of the pauter rods to rev to 7.5k with power.

My tuner says, its no problem making peak hp at 6.7k, and extend the rev limit to 7.5k. He says even if the dyno shows a drop in hp, it will still be worth it. That made me scratch my head! I always thought that if you stop making power at a certain rpm, cut it since its no use in extending the rev limit.

What do you guys think? Tnx

[Mo_Gearhead]

QUOTE: "The increase in displacement in my case could make my even mild cam milder. "
__________________

Always remember that an internal combustion engine is just a glorified "air pump."

There are several factors involved that make moving that air most efficient; valve size, intake manifold size and design, head configuration, piston dome shape, exhaust/header design ....but even when those items are optimized, it is the camshaft design and proper installation that allows the engine to move the "air" most efficiently ... the 'camshaft' is the most critical piece in making the power you desire.

[camlob]

Good words of wisdom. There is a tug of war between me and the tuner. He says that adjusting the intake cam to 115 degrees is very minimal. He wants me to get a bigger cam.

I just cant reconcile the two methods of cam timing. One, using the degree wheel data pts and the other, using John D's straight linear method.

I am inclined to adjusting my cam timing to Johns cam specs next week. I just wanted to get some input over here.

Oh well.....

[johndaroza]

I would think the 305 duration cam would put you were you want to go! I had a cam with hi lift and 300 duration and it spun up to 8000 rpm! that thing burnd the tires at 130mph hitting 4grear hard! best of luck

[Walt Fricke]

Camlob

For sure, reset the cam timing to see what it does. Not that big a deal.

And your brother's 997 is not a fair comparison. A full on air cooled 911 race motor can make about 110 hp per liter. A water cooled, 4 valve 997 can make from 165 to 175 hp/liter. Its volumetric efficiency is just so much better. Of course, your's isn't a full on race motor with that cam, and you may well not want that anyway. Shifting up around 8,000 greatly reduces engine life between rebuilds.

You are wrong about your "rev limit" thinking. Assuming that the mechanical limits of the engine aren't exceeded (and with race valve springs and race rods and bolts I'd expect that it will hold together fine at 7,500), the real question is how does the shape of the torque curve determine optimal shift points for your gearing. Which depends on your gearing too.

Without doing the math (which isn't all that tricky), what you want to do is have the torque multiplied by the gearing in the lower gear match the torque multiplied by the gearing in the new, higher gear after you shift.

For any kind of torque curve likely from a 911, this is always going to be above the HP peak. Unless you have a race gear box with the gears evenly spaced, the optimum shift point can vary from gear to gear. First gear's optimum, because it is usually way low compared to second, is apt to be way up toward the mechanical limit.

I had a race motor which I shifted at 8,000 rpm. Then I finally got the car on a dyno, and did the math for shift points. Oh - peak HP was about 7,100, and the calculated shift points were about 7,600. I'd been wasting time running it up to 8. It felt like it wanted to keep going (unlike CIS motors,which you can feel just run out of steam if you disable the rev limiters), but that was misleading - there were the facts.

Anyway, that's probably what your tuner is trying to tell you about where to shift. You see "rules of thumb" like "shift at 500 rpm over HP peak" and so on, but with dyno information a guy really should just do the math. Then you will know.

[camlob]

Originally Posted by Walt Fricke

Camlob

For sure, reset the cam timing to see what it does. Not that big a deal.

And your brother's 997 is not a fair comparison. A full on air cooled 911 race motor can make about 110 hp per liter. A water cooled, 4 valve 997 can make from 165 to 175 hp/liter. Its volumetric efficiency is just so much better. Of course, your's isn't a full on race motor with that cam, and you may well not want that anyway. Shifting up around 8,000 greatly reduces engine life between rebuilds.

You are wrong about your "rev limit" thinking. Assuming that the mechanical limits of the engine aren't exceeded (and with race valve springs and race rods and bolts I'd expect that it will hold together fine at 7,500), the real question is how does the shape of the torque curve determine optimal shift points for your gearing. Which depends on your gearing too.

Without doing the math (which isn't all that tricky), what you want to do is have the torque multiplied by the gearing in the lower gear match the torque multiplied by the gearing in the new, higher gear after you shift.

For any kind of torque curve likely from a 911, this is always going to be above the HP peak. Unless you have a race gear box with the gears evenly spaced, the optimum shift point can vary from gear to gear. First gear's optimum, because it is usually way low compared to second, is apt to be way up toward the mechanical limit.

I had a race motor which I shifted at 8,000 rpm. Then I finally got the car on a dyno, and did the math for shift points. Oh - peak HP was about 7,100, and the calculated shift points were about 7,600. I'd been wasting time running it up to 8. It felt like it wanted to keep going (unlike CIS motors,which you can feel just run out of steam if you disable the rev limiters), but that was misleading - there were the facts.

Anyway, that's probably what your tuner is trying to tell you about where to shift. You see "rules of thumb" like "shift at 500 rpm over HP peak" and so on, but with dyno information a guy really should just do the math. Then you will know.

Thanks man. I just spoke to John D and he said that I shouldnt expect much from using the cam specs. Actually I should retard a bit the lift pts from 3.2 to around 3.0 but it wouldnt give much effect. He highly recommends the DC 62 which is about in 304 duration.

I will extend my rev limit and will try to do the math. I was thinking maybe 7,200 but will decide later. So he made sense! You guys are a helpful bunch, Thanks.

[camlob]

Walt Fricke - What pt in the torque curve data point should I get to multiply? The highest pt?

[Walt Fricke]

It is a little more complicated than that. What you do is write town the torque at, say, 500 rpm intervals from 3,000 to as high as you got on the dyno, say 7,000 or 7,500.

You then make a table where, for each gear you do the multiplication of each of these RPM intervals. And, again for each gear, you associate a speed with rpm in that gear. You can be arbitrary about using your actual differential ratio, and your tire diameter if you want, but it makes more sense to use actual ratios. If you are doing this by hand you could just use one of those gearing charts which shows (for standard gearing and tire diameter) speed versus RPM per gear.

Then you plot, for each gear, the resultant. The first point, say, would be 3,000 rpm in 1st gear, producing 300 units of "thrust" at 20 mph. And you keep doing that until you are done with 1st gear. Then connect the dots on the plot. You will get a curve which goes up as speed increases, and then goes down.

Then do this for second gear. Again, you get a curve. It is not as sharply pointed as 1st was. But it will intersect the 1st gear curve (arguably) at some point. That will be the speed where you would want to upshift. You can then backcalculate to see what RPM that would be in 1st. Or you just interpolate as to where it is between the points on the graph.

1st is a bad example, because you might run out of revs before you hit the crossing point. But for 2d, 3d, 4th, and 5th, you will get crossings. Each successive curve will start out lower, not rise as much, and will fall off a bit less - it will be flatter. But these are the key points of intersection.

It doesn't take genius to see that this kind of thing is what spread sheets are for. I have one I got from someone. I've messed it up some because I don't really know how to deal with graphs in Excel. And I don't know how to turn a computer page of its graph (or just of the table) into a JPG that I can post here. I'll see if I can use my camera, as a kind of blunt force method.

Anyone wants a copy of such a spread sheet, worts and all, send me a private message with your e-mail address and I'll attach an Excel sample.

Walt

[Walt Fricke]

Here are photos of my computer screen with the spreadsheet up, showing various parts of the page. The actual numbers for the RPM where shifting is optimum were entered manually, not from a formula, after inspecting the charts. I have another spread sheet where a buddy who really gets into Excel had the program determine where the curves crossed. Impressive computational power to a guy like me. But this is a simpler one.






You can easily see the triangular areas under the curves which you would give up by shifting too early or too late.

This is a spread sheet a guy named Dave Ferguson did 15 years or so ago, but in which I entered the data for my stock engine/stock transmission SC.

Before I had the spread sheet I sort of knew about doing this kind of thing. There was an article from Porsche in Panorama, reprinted in Upfixin, which explained doing this in the era before the widespread use of spread sheets, and I made up a graph like this on graph paper with a pencil and calculator. I used Porsche's published data for my torque curve. The one here uses Dynojet data from my car.

You can kind of do the same thing if you have G data recording in your car. The Gs just before an upshift should be the same as the Gs just after.

[camlob]

You got PM. Great work Walt.