Originally Posted by sofakingdom

TPI stands for Tuned Port Injection.

THIMK about the concept of "tuned" for a moment. That term implies that, at some certain frequency or frequencies, something happens that's different from what happens at other frequencies.

Let that idea sink in.

Now: the speed of sound in air at STP (standard temperature and pressure: about 760mm or 30" of pressure, aka 1 atmosphere, and 0°C or 32°F, aka freezing of pure water) is right around 1100 feet per second.

That means, if you have a tube (something like, let's say, a musical instrument), that the "note" it produces, which is controlled by the frequency at which the length of the tubing it's made out of has a resonance, is related to its length. Pretty basic stuff really.

In any engine, the air rushing down the intake tract, will slam into the back of the intake valve, just as it closes, producing a pulse of significant POSITIVE pressure. All intakes on all engines do this of course, but in a general way, it's not often used as a direct method of augmenting cylinder fill (power output), in anything resembling a "street stock" sort of application.

TPI however took advantage of this fact of physics. It is designed such that the pulse of positive pressure from ONE cylinder travels BACK UP the intake tube (which is about 22" long), into the plenum, and then down into ALL THE OTHER intake tubes. Let's keep the numbers nice and round: let's say that the sum of the 22" up the one tube, plus the 22" down the other tube, plus the length rattling around in the plenum, is 48", on average. That's 4'. TPI uses this reflected pulse to AUGMENT the fill of the NEXT firling cylinder, but only at some RPM at which the pulse arrives at the next cylinder after it has travelled about 4'.

Butt, the pressure in the intake tract is somewhat less than atmospheric, otherwise the air wouldn't be motivated to go down it. The speed of sound is somewhat less when the pressure is lower. Once again. to keep the numbers REAL SIMPLE, let's simplify the speed of sound inside a running engine's intake manifold, to 1000' per second.

What this means then, is that at some engine RPM at which the POSITIVE PRESSURE pulse generated by the sudden closing of the intake valve while air is rushing down the tube and SLAMS into the back of that valve, and then that pulse bounces back from that valve and finds its way to some OTHER intake valve that happens to be OPEN at the moment, would result in GREATER cylinder fill (torque, aka output) than would happen at ANY OTHER engine RPM. By selecting that length - 4' - we have TUNED the engine's intake tract to REINFORCE cylinder fill at some particular RPM, or more accurately in some range centered on that RPM, that corresponds to the time it takes for sound to travel 4'.

RPM is a count of events per minute; "frequency" is usually considered in terms of events per second; so all we have to do to convert from one to the other, is to apply a conversion factor of 60. Trivial to say the least.

Make sense so far?

All that remains then, is to find this RPM. So: the time it takes for sound to travel 4', is 1000 ÷ 4. How convenient: that's exactly 4 milliseconds, or, 1/256 of a second. What RPM is that?

Well, a V8 engine fires 4 of its cylinders per revolution. That means that the RPM at which the cylinders are firing exactly 4 milliseconds apart is the RPM at which it make one full revolution in 16 milliseconds. What RPM is that? 1 second ÷ .016 = 62.5; again, let's simplify it so the numbers come out even (which happens to take out some of the error of the last simplification), and call it an even 60. 60 revolutions per second is ... 3600 revolutions per minute. 3600 RPM. Let THAT number sink in.

TPI is carefully, rigorously, mathematically designed, according to well and widely proven and demonstrated principles of physics, to produce its optimum torque at 3600 RPM.

Now: did you notice ANYTHING about a real-world engine that was CONSPICUOUSLY ABSENT from all of these calculations? I certainly hope so.

The thing that is COMPLETELY missing is CUBIC INCHES.

What does this tell you? Basically, that NO MATTER WHAT SIZE the engine you put TPI on is, its peak torque is going to occur at 3600 RPM. Doesn't matter how large the tubes are; doesn't matter what cam you put in it; doesn't matter what the head flow is; doesn't matter what the throttle body diameter is; 3600 RPM. It's DOMINATED by the properties of those tubes on the intake. The best you can do is to make sure that the engine is as restriction-free as you can get it, at 3600 RPM; and that the cam you select produces the largest possible pulse of "reversion" pressure at ... 3600 RPM. Which is why TPI is SO PICKY about cam selection. Just jamming a bigger cam under it will often make it produce LESS power, if said cam doesn't agree with TPI's ... 3600 RPM ... constraint.

OK, so once again, lets go back to numbers. (after all, everything in life is numbers, whether you humans are willing to admit it or not) Power, whether measured in HP, kW, or whatever, is the time rate of the production (or expenditure, depending on which way you're looking at it from) of energy. X number of units of energy per Y units of time, in all cases. So how do we get from torque (work, or energy), to power?

1 HP = 33,000 lbs of weight lifted 1 foot in 1 minute. Pretty simple really.

Back in the 1700s, when steam engines were replacing horses, the single biggest use of such a thing was pumping water out of coal mines. (coal being the fuel that powered the early Industrial Revolution worldwide, but particularly in England and Germany) Everybody that built engines, was charging people that used them, according to how many horses their engine could replace. Well, how do you put a number on that? They would look at a horse walking around a turnstile sort of thing that was connected to a pump shaft, and taking into account how long the horse would live depending on how hard you worked it, they would figure out how much water the horse could lift in some unit of time, without working it to death. Believe it or not, people all over the industrializing world all seemed to come up with pretty similar numbers. Turns out, d00d named James Watt (ever heard of a unit of power called the watt?) in England, determined that one horse, under conditions that the horse could sustain more or less indefinitely (over a normal horse lifetime anyway), could lift 33,000 lbs of water 1 foot out of a coal mine, every minute; and could do this more or less all day, every day, day in and day out, from attaining full growth to visiting the knacker. That's one horsepower then: 33,000 lbs, 1 foot, 1 minute. In Germany, whoever made the same calculation there, came up with a number that was closer to 31,000 lbs. (maybe Germans didn't work their horses as hard? maybe theirs weren't as big or as powerful? maybe they didn't feed them as much or as well? iunno) Keep in mind this was all going on before the metric system was invented in France during the French Revolution, as some kind of pure scientific measurement method devoid of any religious or historical taint. (read about the French Revolution: you'll be simultaneously inspired and horrified, it was a time when things got tried that couldn't possibly ever be tried again under any other circumstances, even besides the guillotine)

Torque is a twisting force. 1 ft-lb is the amount of twisting that occurs when one pound of force is applied (or expended) on a rotating object 1 foot in radius (2 feet diameter) that applies 1 pound of force at its circumference. Think, a tire being spun, pushing a car. A point on a rotating thing with a 1 foot radius moves 2 pi feet in one revolution.

So do we get from torque, to horsepower?

Well, if horsepower is the time rate of doing work, and something that has 1 ft-lb of torque does 2 pi of work every time it rotates, then all it takes is another conversion factor. 33,000 ÷ (2 pi), or 5252.11, is this new Magic Number. Often referred to on dyno pulls as 5250 RPM, where the torque and HP curves intersect, if plotted on the same axis. Horsepower (at any given RPM) = the torque at that RPM, times the RPM, divided by 5252.11. Simple: it's all numbers.

A typical real-world engine like ours, capable of running on fuel we can actually buy at the street corner, under conditions we actually can run the engine under, can produce about 1¼ ft-lbs of torque per cubic inch, at whatever RPM it produces its peak torque at. You can move that RPM up or down by selection of cam, intake, exhaust, etc.; but in the end, it's still 1¼ ft-lbs of torque per CI. Therefore a 305 can produce roughly 380 ft-lbs max, a 350 about 440 ft-lbs, and a 400 about 500 ft-lbs. Butt, don't forget, you're working with TPI: by definition torque peaks at 3600 RPM. So if you have a 305, your HP will be around 265. All you have to do to find the horsepower you're going to make (assuming you've optimized everything... it's all downhill from the number this calculation gives you, in the RW) is plug your CID into that simple equation.

400 CID × 1.25 = 500 ft-lbs. If that occurs at 3600 RPM, then your HP will be 500 × 3600 ÷ 5252.11; or, around 340 HP. For a 350, you're going to get real close to 300 HP.

BUTT: that whole "reinforcement" thing comes with a penalty. And that is, at any HIGHER RPM than 3600, the pulse provides less and less reinforcement, which means the help it gives you is less and less (the torque falls off REAL FAST as RPM increases); and, worst of all, at SOME RPM, the pulse actually begins to INTERFERE with cylinder fill. It REDUCES engine output BELOW what it would be WITHOUT the tuned effect at all. And THAT RPM is real close to 1.5 times the peak RPM; or, 4500. What you get is the classic "Mt Everest" TPI torque curve: MASSIVE peak at 3600, craters above 4000, is ALL DONE by 4500.

Real simple. It's all numbers.

Originally Posted by sofakingdom

TPI stands for Tuned Port Injection.

TPI is carefully, rigorously, mathematically designed, according to well and widely proven and demonstrated principles of physics, to produce its optimum torque at 3600 RPM.

Now: did you notice ANYTHING about a real-world engine that was CONSPICUOUSLY ABSENT from all of these calculations? I certainly hope so.

The thing that is COMPLETELY missing is CUBIC INCHES.

What does this tell you? Basically, that NO MATTER WHAT SIZE the engine you put TPI on is, its peak torque is going to occur at 3600 RPM. Doesn't matter how large the tubes are; doesn't matter what cam you put in it; doesn't matter what the head flow is; doesn't matter what the throttle body diameter is; 3600 RPM. It's DOMINATED by the properties of those tubes on the intake. The best you can do is to make sure that the engine is as restriction-free as you can get it, at 3600 RPM; and that the cam you select produces the largest possible pulse of "reversion" pressure at ... 3600 RPM. Which is why TPI is SO PICKY about cam selection. Just jamming a bigger cam under it will often make it produce LESS power, if said cam doesn't agree with TPI's ... 3600 RPM ... constraint.