Optimal Shift Points
 

Alright, I figured this would go best in here. Other sub-forums didn't really seem to fit....

This is an article I got off another car forum, but it applies all the same here. I think it's always good to know when you should be shifting. This will take away any guesses you have as to what rpm you should be shifting at. Here is the article;

Optimal shift points
By Dave Lum - 10/22/98

Looking to get maximum effect out of your car when accelerating? Try optimizing the RPM's you shift your car at. If you're shifting at redline, you're probably not getting the best acceleration out of your car - in fact, the optimum shift RPM will probably be different for each gear. You can calculate the optimum shift points, but it takes two pieces of information : Gear ratios of all your gears, and a chart of your torque curve.

The trick here is to keep your engine in the meaty part of the torque curve, as your acceleration curve will match your torque curve - meaning if your torque peak is at 3000rpm, that's the RPM your car accelerates the hardest. How do you keep the engine in the best part of the curve? Shift when you will be at an RPM where the car will make more torque AFTER you shift than in your current gear. This may sound confusing, but I will give some examples to help demonstrate this. Gear ratio is important because you need to know what your RPM's will drop to once you shift. EXAMPLE : If you're at 6500rpm in first, you need the gear ratio to know what RPM that will translates to once you shift to 2nd.

Below is a torque curve for a fictitious engine :
Engine
Torque Gear ratio Now with your gear ratio in hand, figure out how many RPM's will drop going from 1st to 2nd. In this case, it'll drop by 1/3 (2:1 divided by 3:1). So shifting from 1st to 2nd at 6000rpm, it will drop to 4000rpm. This means you're going from 330ft/lbs (see chart - 6000rpm in 1st) to 220ft/lbs (6000rpm * .6666) = 4000rpm in 2nd). Not what you want - you want these numbers to be the same. How about 6400? 285 vs 240 - much better, but let's try 6800 : 210 vs about 250. Oops, too far. In this case, optimum shift point is between 6400 and 6800rpm for the 1-2 shift.

It works the same for each consecutive gear, as the shift RPM may be different for each gear. Below are graphs representing how it looks on "paper".
RPMS___Torque__3:1____2:1____1.3:1

3600____100____300____200____130
4000____110____330____220____143
4400____120____360____240____156
4800____125____375____250____163
5200____125____375____250____163
5600____120____360____240____156
6000____110____330____220____143
6400____95_____285____190____124
6800____70_____210____140____91
Gear -->_______1st____2nd____3rd


Notice anything? The two torque outputs are almost the same (6600 in first and 4400 in second), which is what you're going for to achieve maximum acceleration. What does it look like on the horsepower side? Take a look at the chart below :

http://www.datsuns.com/Tech/shiftcurve1-2-tq.jpg


http://www.datsuns.com/Tech/hpcurve.jpg


Well what do you know! You bracket the HP peak. HP after the shift is the same as HP before the shift. Now we know shifting at the torque peak isn't the answer, and neither is shifting at the HP peak.

The math required : next higher gear ratio / current gear ratio (i,e. second gear divided by first) = X

X * shiftrpm = nextrpm, where shiftrpm is the point your shifting AT, and nextrpm will be the RPM of the next gear at the same speed.

firstgeartorque = torque at shiftrpm times current gear ratio (example, at 6400rpm we have 95ft/lbs, multiply this by our gear ratio in 1st (3:1, or just 3) and you get about 245ft/lbs.

secondgeartorque = torque at nextrpm times next gear ratio (in out example, we have 120ft/lbs @ 4400rpm, times 2:1 (second gear ratio), and you get 240ft/lbs.

Change shiftrpm until firstgeartorque is equal (or close to) secondgeartorque.





I did all the math for a close to stock torque curve, and most of the mods won't change the shape of the graph that much anyways. Of course those with FI will have to change their numbers around.

Our gear ratios are
1st 3.76
2nd 2.269
3rd 1.645
4th 1.187
5th 1.0

For my caclculations, I didn't use sixth, because it's faster than most go, and it's not useful for most motorsport. Besides, you would be shifting fairly low (probably less than 8K) from fifth to get there. Anyways, on to the math!

Shifting from 1st at 9,000 lands you at 5,670 in 2nd. Torque at 9,000 is about 105 ft lbs, and torque at 5,670 is about 130 ft lbs. Multiply 105 by 3.76 and you get a total of 394.8 ft lbs in first gear. Multiply 130 by 2.269 gives you 294.97 ft lbs in second gear.

Since you are making less torque in second gear than first, you want to shift into 2nd after 9,000. I didn't find a close to stock graph that went up to 9,500, so I don't know the exact rpm.

Shifting from 2nd at 9,000 lands you at 6,525 in 3rd. Torque at 9,000 is 105 ft lbs, and torque at 6,525 is about 127 ft lbs. Multiply 105 by 2.269 and you get 238.2. Multiply 127 by 1.645 and you get 208.92.

Shifting from 3rd at 9,000 puts you at about 6,500 in 4th. Torque at 9,000 is 105, torque at 6,500 is about 127. Multiply 105 by 1.645 gives you 172.7 ft lbs. Multiply 127 by 1.187 gives you 150.75.

Shifting from 4th at 9,000 rpms puts you at 5,472 in 5th. Torque at 9,000 is 105, torque at 5,500 is about 130. Multiplying 1.187 by 105 gives you 124.6 ft lbs. 5th gear is 1:1, so at 5472 torque is at 130.

So for a stock car, you would want to shift above 9,000 for all gears except 5th and 6th. If you have any mods that are going to move your shift points up, then you'll have to shift even later. I don't think there are that many that make much of a difference though.


I hope this has been helpful to you guys!

So what your telling me is to redline it in every gear to accelerate fast? Wow, what a long article just to find out it doesnt make a dif. for us with a 9k redline. Although interesting, everyone is already driving "correctly" apperantly :)

He's explaining WHY :)

we have one of the few cars where redlining it is correct ^.^

Thanks for the write up, it definitely gives a better understanding of torque. It's easy to notice the difference of acceleration when shifting a second or two after the beeper rather than at the beeper.

If you read the article, you would have seen that I told you to run it over the redline. That I doubt most people do. I bet 90% run to redline.

didnt mean it to be an ass or anything, just was suprised that in the end it didnt really make much of a difference concidering our 9k redine. Like I said, interesting article, like raptor said it givees a good understanding of torque.

I know.. LoL I was just giving you shit. It wouldn't make a whole lot of difference if you shifted at 9,000 or 9,300... It's just always good to have something to help you figure out where you need to shift. Especially those with turbos and such, becuase their power graphs are so different.

Ok so you mean Mazda makes a car that makes best acceleration while over-redlining it, somehow I don't think Mazda or any car manufacturers recommends going over redline.

due to its extremely high torque curve, this is indeed correct. Most cars torque curves are easily 2-3K below redline. the 8's is closer to 500 RPM's.

Thus he's correct...the 8 is one of the rare cars where that redline should be utilized for top performance.

Actually, in every car that I've seen this done for, in the first two gears (even with a close ratio gear set) going over redline is best for acceleration. I'm not saying it's good for the engine. I know the Zetec can be over-reved by 1,000 rpms without any problems.

It's not so much the torque curve as it is the gearing. If first has really high gearing, and there is a big gap between first and second, then you are going to have to over rev the piss out of first to get a decent leg up on second. So even if your torque is falling off the face of the earth, it can still be beneficial to over rev.

I have played around on ocassion by shifting without taking my foot on the gas petal while holding it to the floor. I seem to get a slight hesitation when I drop the gas petal and shift rather than just keeping it to the floor. I was worried (the first time I tried it) as I thought I'd blow up the motor but it really makes a difference. You do have to shift fast though. !!! I do a lot of reading and not too much writing; but I love the site !!

You probably get hesitation because you are hitting the rev limiter. If you do that, it will cut timing, or gas, or do something so that you can't rev further.

Best thing for you to do would be to shift normally. IE not hold down your foot on the gas. If you are good enough, you can shift very fast, and be right on the gas again in less than half a second. It's kinda silly to keep your foot on the gas.

i'd imagine that never letting off the gas would cause needless wear on the clutch, as matching rev's wouldn't be possible when upshifting considering how the revs would always be RISING when your going from a gear in higher RPM's to one in lower.

:suspect:

I doubt it would make that much of a difference if you were shifting very quickly (IE when racing), since the flywheel doesn't have enough time to slow down. Normal driving, yes. I sure hope he doesn't stay on the gas during normal driving... LoL

Any chance for a sticky on this? I think people should read it...

So at 9500 rpms in 4th you're exceeding the speedlimit by how much? jk :)

4th gear you probably wouldn't need to really over-rev. As you go higher in gears, the multiplication factor decreases. The difference between third and fourth is half of a multiple. So you aren't getting much more torque than third. That's why as you go up in gears, the revs you should shift at go down (9,500 for first, 9,300 for second, 9,000 for third etc).

Very good article. If you think about it, it's very intuitive. I've often thought about this relationship between hp/torque curves and shift points for optimal acceleration but never spent the time researching or studying to figure it out. Now I don't have to!

Figured I'd bring this back from the dead... I always like to see where I need to shift, and I think it's important for folks to see it!

I appreciate the post and the effort to shed light on the issue.

The quote has the right idea, but I think it can be more accurate.

"Shift as close to the point where the upshifted output shaft torque IS RISING (cresting) and is less than or equal to that of the origin gear."

The way cars come from the factory, it would be really tough to find a car that makes more torque at the output shaft when upshifted. When one does this though, they would have wasted a small slice of the powerband.

You've implied in on your analysis, but I'll mention explicitly. Shift points are only relevant for real world torque curves that rise and drop (like a hill).

Essentially, one is using gears to prolong the 'decay' of the torque curve until the resulting torque cannot overcome the load as max speed is reached.

In the torque chart the OP posted above, try stringing the torque curves for the gears end on end in sequence to visualize this.

Shift points are generally irrelevant for torque curves that are linear (flat, rising, falling) because you simply shift at rev limit for max accel.

CORRECTION

from: In the torque chart the OP posted above, try stringing the torque curves for the gears end on end in sequence to visualize this.

to: In the torque chart the OP posted above, try stringing the torque curves for the gears end on end AT THE CORRECT SHIFT POINT in sequence to visualize this.

Ooh, didn't notice at first that the thread was so old.

Well, Doesn't help me, considering that my car grinds the hell out of 1/2/3 in the high RPM's.

Basically you want to put the most amount of torque to the wheels after gearing is factored. I cut and pasted that article for the most part, so that kind of slipped by me. But yes, you would ideally want to shift where torque is equal in both gears. Since first would be falling off, and second would be rising, you would have optimal acceleration. IE first making 330 ft lbs at 9,500, while second making 325 at 6,000. Those are guesstimations, but you get the point.

Stock ZX2s need to rev to about 7,200 in first for optimal results. The (series I) Mazda 6 V6 needs to shift around 7,500 or so. It feels like you won't be accelerating as hard because the engines power is dying off, but it does the trick.