When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
This is partially to have somewhere to host this to help settle stupid "Miata Club" debates, but can also help make decisions with engine upgrades and swaps. I started a gearing spreadsheet when I decided to swap in the LS1, to help decide on a rear end ratio. Just mimicking the online calculators that take your gear ratios and tire diameter and spit out max speed in each gear wasn't enough. So I incorporated the torque curve, to show how torque at the wheel is affected by engine modifications and gear ratio changes. I'm currently working on turning this data into a quarter mile simulator, then when I finish my swap I'll work on correlating the inputs to make it useful. It'd be cool to change the torque curve, say by estimating how the torque curve will change when going from a stock cam to a medium lift cam, and see how it'll affect trap speed and ET. While there are a lot of estimates here, I'm sure it'll still be useful to at least get an idea. Is shifting the torque curve 500 RPM to the right and increasing peak torque by 10 ft-lbs going to drop 0.1 or 0.5 second off the quarter mile? You really can just take the average power of the usable range (RPM between shifting and redline) and the higher one wins, but this is more fun. Eventually I'll clean up the spreadsheet and post it here, but for now we'll just look at screenshots.
I've heard too many old school claims of "horsepower is how fast you hit the wall, but torque is how far it'll get pushed" and "horsepower sets trap speed, but torque is what gets the car going". Also, "horsepower is just calculated from torque, torque is the driving factor". I added my friends swap into my spreadsheet (Rollie, his build is on here, turbo F20C) to help show what really matters in a dyno graph. Acceleration is a function of weight and the forces acting on the vehicle, drag and rolling resistance are negative, tire propulsion is positive. If you're comparing the same chassis/tire size, then wheel torque is directly comparable between each vehicle. This gets mistaken as engine torque, because everyone knows that torque is a physical measurement. Torque gets converted to force, and force moves the vehicle, torque must be the important factor, and power is just a calculation to show how much torque is sustained at high RPM's! The missing aspect is speed (obviously, it's the only other part of the equation), which also means that gearing isn't taken into account. To start off, below are the torque curves I'm going off of. They're both just based off of searching for a bunch of similar dyno graphs online. Basically, both are 400 hp motors, the LS1 at 400 lb-ft and the F20C turbo at 275 lb-ft.
The LS1 is mated to an F-body T56 with a 4.10 rear end ratio. The F20C is mated to an S2000 transmission with a 4.30 rear end ratio. The T56 is a little more spread out, and is longer gearing overall, it's basically a 4-speed with 2 overdrives, 1st gear is 2.66 and 4th gear is 1.00. The S2000 trans is very short starting with a 3.6 1st gear and even 5th gear is still greater than 1 (1.126). Gear for gear, that puts the F20C at a much greater reduction, so higher torque multiplication and lower speed. Below shows the graphs of torque at the wheel versus ground speed.
Now this wheel torque versus wheel speed is split up to what the engine would actually see during an acceleration event.1st gear goes from 2500 RPM to redline, it's assumed anything under 2500 RPM isn't a valid area, because the engine would rev over this and either the clutch or tires slipping would keep the engine above this RPM. Then the shift to 2nd at redline, and so on. This translates to the forces the tires are seeing, not accounting for acceleration due to tire forces when the vehicle is traction limited. This tractive effort is plotted with both vehicles below. The area under this curve is what will correlate into actual acceleration.
Both vehicles follow basically the same trend of torque versus ground speed. This means they would be very close in a race where the only variable was engine related (same tires, driver, weight, etc). This is because power is the determining factor of what the torque ends up being at the wheel. Engine torque is not a valid comparison, because it gets multiplied by gear ratios, and if different vehicles have a different gear reduction for a given speed (maybe because one motor can be screaming over 8000 RPM, while the other has to shift at 6000 RPM), then the engine torque is no longer comparable. Because of this reduction, engine torque is different from driveshaft torque, and that is different from wheel torque. Power, however, stays the same throughout the system. At the engine, you take power, divide by RPM, and multiply by 5252 to get torque. This is also the case at the wheel. If both vehicles are going the same speed and both make 400hp, they will have the same torque (Power/RPM*5252, if power and RPM are the same). One engine could have 200 lb-ft and the other could have 400 lb-ft at that wheel speed, but if they both have 400 hp, the wheel torque will be the same and thus acceleration will be the same. If you take the same torque curve but shift it to higher RPM, this means that at the higher RPM you can stay in the lower gear for longer. So even though the torque isn't greater at the engine, the reduction gear is greater for a longer time.
Torque isn't a completely useless variable to measure, but that's only because we've gotten used to evaluating low end acceleration with torque. We're accustomed to comparing motors, and saying one has more low end torque because people often drive all motors at similar RPM's on the street. If the LS1 is driven from 1500 to 3000 RPM when leaving a stoplight and so is the F20C, sure it's valid to say the F20C is gutless because it doesn't have as much torque. But it certainly would be more accurate to say that the LS1 has more power between 1500 and 3000 than the F20C. Even though engine RPM will be the same, for a given ground speed the F20C will have a greater gear reduction to bump that torque up. Comparing the difference between peak torque and peak power gives a good idea of how the torque curve changes with RPM, which is useful info. Torque is also useful because it is the real measured value, so having a linear torque curve means that your acceleration will be fairly linear throughout the flat torque band. This is only useful on the street, because even if you still make the same amount of torque at 4000 RPM as you do at 6000 RPM, if power is higher at 4000 RPM, then you should downshift to maximize acceleration.
Saying that horsepower matters and torque doesn't matter doesn't allude to street manors, reliability, and such like that. In terms of a pure drag race, power matters, torque doesn't. But an engine that has the torque band shifted to higher RPM in order to increase HP also needs to take into account reduced reliability, extra fuel, extra cooling needs, whether the driver is willing to rev that high on the street, etc.
What I've used this for most is gearing selection. If the gearbox is short enough to keep the engine in it's power band, then when choosing a rear end ratio, you basically just need to keep the first and last of your gear ratios in check. In the T56 the change in acceleration from using rear end ratios to make 1st through 4th the usable gears, or 2nd through 5th usable is basically negligible. If you drew a trend line on that last graph of torque versus wheel speed, changing the rear end ratio just shifts the squiggly line up or down that trend line. If you have a favorite track, then there will be a point in shifting the squiggly line where torque is maximized between the low and high speeds on the track, and you can also tune to eliminate as many shifts as possible. Gearing can be selected effectively for autocross because top speeds are fairly consistent, so eliminating that 2nd to 3rd shift is important. Track is less important because speeds vary so much from track to track, unless you're willing to change gears for tracks or operate in a few tracks that are all similar. For my case, I could either use a long gear like a 3.42 with a Getrag or a short gear with a TII diff, a 4.10. The 3.42 means 1st gear could be modulated, then 2nd gear becomes pretty usable, and 3rd and 4th would be solid track gears. You'd get into 5th gear on long straights (shift at 120) but probably wouldn't be sustained long. For the 4.10, 1st is basically unusable for putting power down, but 2nd will likely be launchable, and be able to put power down easier than 1st gear with a 3.42. Then 3rd and 4th are still the most used gears on track, but 5th becomes more important since the shift point is now 100. A downshift to 3rd may only be necessary on shorter tracks. 6th gear is still only 2200 RPM at 70 mph. I wanted a short 1st gear because I have a very aggressive clutch, and I'm trying to spin the tires the least on the street. I'm trying to mimic the feel of a truck in 4 Low where you just let the clutch out, and the truck moves, because the speed at which the engine has reached idle RPM with the clutch out is so low. Then a quick shift to 2nd, but all drag or track launches will just start in 2nd. Then I had to make sure street manors are in check at the high gear, and 2200 RPM is plenty low, especially since I don't want to buck at low RPM if I cam it. This is just an example of how I've used these calculations to make real decisions, I chose the 4.10 and I think it was the right choice to match the rest of the car.
Another conclusion to draw is how higher revving motors pair well with closer and shorter ratios. Both vehicles have shift points that are fairly similarly spread apart (in relation to vehicle speed), but the F20C has much much shorter gears. The extra 3000 RPM redline makes up for this.
I remember a program called Desktop Drag Strip and its companion program Desktop Dyno being a lot of fun to explore back in the day. Changing gear ratios, tire heights, shift points, transmissions, etc. made for hours of entertainment. It wasn't expensive.
I think i used similar math to prove the age old (incorrect) argument about how sloooowwww an S2000 is out of vtak on these forums. It's soooo slooooowwww my Miata is wayyy faster.
You can spend an awful lot of brain power thinking about it in terms of engine torque, then conversion to wheel torque, then adding speed, etc.
Or, you can think about it in terms of energy. If you maximize horsepower production at all time, you will maximize energy production. Transfer that energy into kinetic energy (speed) and you will go faster.
That! Powertrain matching (engine plus drivetrain) is bigger than a single torque or HP number. Area under the curve too, peak numbers alone say nothing. BTW here's an interesting presentation on powertrain matching as some (more) food for thought:
That! Powertrain matching (engine plus drivetrain) is bigger than a single torque or HP number. Area under the curve too, peak numbers alone say nothing. BTW here's an interesting presentation on powertrain matching as some (more) food for thought:
Great timing on this post. I was just setting up a similar calcuation to compare final drives on a power limited car to see which was the better option.
For the ones that are so inclined, you can try OptimumLap. It is a free, point mass simulator but works pretty well in determining what's important and what not. Not only in straight line, too.
So why do manufacturers bother making engines with different power ratings when you can achieve the same thing with gearing.
Wouldn't designing, manufacturing, calibrating and validating a range of gearboxes be vastly cheaper than doing the equivalent for a new range of engines?
So why do manufacturers bother making engines with different power ratings when you can achieve the same thing with gearing.
Wouldn't designing, manufacturing, calibrating and validating a range of gearboxes be vastly cheaper than doing the equivalent for a new range of engines?
I agree with Andrew, rate of energy produced is all that matters.
To simplify, it is work. A basic physics class can teach such. Average force over distance. If we can optimize force provided at all times, given the same distance (this is a race over a fixed distane right?) the vehicle that provides more force over a distance wins. What is work over time? (Like the vehicle that out accelerates the other one?) Oh, i know... Power.
11-07-2017, 06:04 PM
Power versus Torque and Gearing
7
