Basically I'm just trying to figure out if I'm doing this right.

I went to the dyno last week and I'm using the torque numbers to find out where in the rpm range I should shift.

It's hard to explain, but here goes.

I figured out how much torque the engine is actually producing at the rear axles. For example the engine peaks at 200rwtq at 4400rpm. In first gear at 4400rpm (22.5mph) there is potentially 2839ft-lbs of torque applied to the rear axles (tq x diff x gear, 200ft-lbs x 3.38 x 4.2).

I have created a table with a torque numbers from 2000-7000rpm with a column for each year. I also created a table with my rpm range that shows how fast the rear wheels are moving for a given rpm with a column for each gear, so I know roughtly what rpm the engine will drop to on each gear shift (which can vary on the shifting delay and aerodynamic decceleration).

What I am doing is finding when the axle torque of the current gear and rpm drops below the torque of the next gear with the appropriate rpm reduction. That is when I think a shift would be made.

Anyway, first and second gear I deemed quickest to pull right to redline (7000rpm), for 3rd to 4th gear shift at 6700rpm and 4th-5th gear shift at 6400rpm.

- In first gear at 7000rpm, the rear axles are under 1865ft-lbs of torque, shifting there drops the rpm to 4200rpm in 2nd, which is at 1633ft-lbs.
- In second gear at 7000rpm, 1094ft-lbs at the rear axles, shifting into 3rd drops the rpm's to 4600rpm at 1061ft-lbs.
- In third at 7000rpm there is less torque produced at 7000 then there would be after the shift, so at 6700rpm in 3rd there is 813ft-lbs at the rear axles, shifting to 4th drops the rpms to 4900rpm at 809ft-lbs.
- In fourth at 6400rpm, 662ft-lbs at the rear axles, shifting to 5th drops the rpms to 5100rpm at 642ft-lbs.

Does any of this make sense to anyone?? I'll post the excel spreadsheets later on when i get home.

Nelson - I'd be interested in your spreadsheet. I've got a speadsheet that I've made up for analyzing different gearing, but I've never put the time into working up a shift point analyzation.

Ok, below is the link for my spreadsheet.
torque-speed.xls (http://www.nelsonpapel.com/temp/torque-speed.xls)

Obviously you'd have to enter whatever tire size, gear ratios, and torque/rpm numbers.

The transmission ratios I used were from an e36 ZF tranny (328/M3) with the 6th ratio from the euro (I don't have a 6spd, I was just curious).

so can you find a case where redline isnt optimal?

so can you find a case where redline isnt optimal?

The only time redline isn't optimal is when the powerband has fallen so far off by redline that the next gear wins. So if 2nd gear is half the ratio of 1st gear, your shift point is basically where the torque has fallen to half of the peak torque value. The only time redline is like that is if some moron at the factory set it a bit too high.

ThinAir, your math is correct. That's how you figure shift points.

thats my point. i've never seen a case where redline isnt optimal

In pretty much every case, the shifts to 4th and 5th are optimal at redline. The shifts to 2nd and 3rd are a few hundred above redline in a lot of cases, which is where a chip comes in handy.

thats my point. i've never seen a case where redline isnt optimal
I just pointed out above that the 3-4 and 4-5 shifts in my case aren't optimal at redline. The engine in my car seems to have breathing issues above 6000rpm, both the hp and torque drop off significantly. Althought it's probably due to the lack of displacement, it's only a 2.8L, so maybe there just isn't enough volume being pushed around above 6k.

And redline isn't optimal on many cars, especially larger displacement or long stroke engines with a lot of low end torque.

shifting at redline = optimal? perhaps you missed the first post.

shifting at redline is optimal in first gear only. 4-5 in the original poster's case was optimal at 6400.

So hp has no bearing on shift points, just torque? I know hp is just the relation of torque to rpm, it makes sense to me, I'm just seeing if any one else chimes in with info/feedback.

Thanks

Also, I put in the torque numbers from a 328 with a twin screw to see how it affected the shift points. With that car the shift points are all at redline because the torque is so linear and doesn't drop off at the top.

thats my point. i've never seen a case where redline isnt optimal

See the first post in this thread.

What do you guys think of this? http://www.geocities.com/MotorCity/Speedway/6172/m3shift.html

I think you are correct for this matter, I did the same for my car (actualy somewhat shorter, as I calculated for redline only, which seems optimal in my case anyway.

I did something else as well, recalculate the diffrent RWTQ's for each gear with a diffrent final drive, and compare by speed, and see in which parts you are actually gaining.

What do you guys think of this? http://www.geocities.com/MotorCity/Speedway/6172/m3shift.html
Seems to be a more graphical version of what I did, just with different units (Newtons instead of ft-lbs). Good find though, it's nice to see how my 328 compares to an M3 (appearently pretty well).

I just converted the spreadsheet to imperial units, this will make it easier to read for you MPH users.
http://www.nelsonpapel.com/temp/torque-speedmph.xls

On a side note, an e36 with a 6spd and a 2.93:1 diff can hit 200mph, in theory (aka a stupid amount of power, traction, and some downforce to reduce lift at that speed)

I tried shifting to drop to maximum torque for several runs at drags, was nearly a full second slower than shifting at close to max hp (change just before rev limiter)

You don't want to shift to drop into max torque though. You just don't want to lose actual wheel torque when shifting from gear to gear. I found I was actually a few tenths faster shifting at my specified shift points then just hitting 7000rpm in every gear.

If i was to shift to drop me to max torque in every gear it'd be something like 7000, 6700, 6000, 5500. That would really slow me down as you found out.

The way I see it hp really doesn't mean anything, torque does. For example, if a car produces a flat torque line from 3000-6000 at 200ft-lbs. The car will pull just as hard in a given gear at 3000 or 6000rpm. Yet, at 3000rpm the engine is only producing 114hp, but at 6000rpm the power now doubles to 228hp, even though the car pulls the same.

You don't want to shift to drop into max torque though. You just don't want to lose actual wheel torque when shifting from gear to gear. I found I was actually a few tenths faster shifting at my specified shift points then just hitting 7000rpm in every gear.

If i was to shift to drop me to max torque in every gear it'd be something like 7000, 6700, 6000, 5500. That would really slow me down as you found out.

The way I see it hp really doesn't mean anything, torque does. For example, if a car produces a flat torque line from 3000-6000 at 200ft-lbs. The car will pull just as hard in a given gear at 3000 or 6000rpm. Yet, at 3000rpm the engine is only producing 114hp, but at 6000rpm the power now doubles to 228hp, even though the car pulls the same.
That is absolutely correct.
The one thing that HP does show is where the torque is made. just using the example above, if you made the same torque at 3k and 6k, yes, it does pull the same at both points and make more power at 6k, but why would you be at 3k? downshift and take advantage of gearing. :)

HP is important in that it lets you take advantage of gearing to get more effective torque at the wheels.

If 2 cars both made perfectly flat torque curves of 200 ft-lb, but one made it from 3k-5k, and another from 7k to 9k, we know which one would be faster even though they would pull equally as hard given equal gearing.

Intersting topic, pretty much what I has intuitivly surmised just trying to optimize use of the torque curve. Shifting a little short also has the added benefit of avoiding the engine cut.

The real world impact of how a track layout impacts where you shift also matters. Just about every track I've been to (not many) has at least one corner where you need to short shift so you don't run out of RPMs in the middle or track-out of a corner.

I got bored earlier today and gave a shot at it. I actually did it the same way the thread originator said without even seeing this thread first. Anyways my dyno only went to 6300, but I wanted to figure out what would happen if I changed my redline to 7000(shark) so I made up some fake numbers from 6300-7000 guesstimating using other 330 dynos that go to 7000.

For the 1-2 and 2-3 shift 7k rpm was optimal, but for the 3-4 shift 6850 was best, and for 4-5 6650 was best. Once I get my oil pump I'll shark it and dyno it again.