New Z owner - Curious about Optimal Gear Shift

[Bubble1982]

Just bought my first zed a couple of weeks ago (e89 s drive 35i manual) - must say I'm absolutely loving it!

Because I live in central london I haven't needed / owned a car for a while (last one I did own was a 1.0 litre metro while at uni 7 yrs ago!) meaning I'll happilly admit i'm far from a pro when it comes to getting the most from all that power. (To be honest I'm still at the stage of uncontrollable grinning when I accelerate hard in 2nd!)

So I've been thinking - if I want to accelerate as quickly as possible, with it being a manual, what is the optimal rev count to change gear?

Theoretically, I would have thought I need to change gear at the point at which the new gear will give me more torque than the current gear. If I had torque data across the whole rev range for each of the six gears I could calculate the optimal shift in each gear (I think!)

Would welcome any thoughts you guys have on a) whether this is the right method, b) where I could get the torque data for each gear across all revs for my model and c) if I'm completely barking up the wrong tree, how I could find the best change point another way?

Thanks in advance for any help you can offer. Looking forward to posting on here more often : )

 

[Maniac]

You can usually feel the best point to change.

The n54 runs out of turbo boost around 6000 rpm (despite it going to 7500) so my shifts are usually around 6200 rpm as you can feel the pull drop off after that. Turbos are at full tilt so it drops down to around 5000 kicks you in the back and off you go again..

 

[Wondermike]

Without doing the science I can't see any situation where the drop off in torque towards the redline would result in lesser acceleration than changing up a gear, just because of the gearing.

 

[Quadracer]

Without doing the science I can't see any situation where the drop off in torque towards the redline would result in lesser acceleration than changing up a gear, just because of the gearing.

I agree, red lining would be the best in a straight line however sometimes you can benefit from short shifting coming out of a corner.

 

[ronk]

Is there not a graph showing the torque curve against rpm?

 

[Wondermike]

There is, but what Bubble1982 is requesting would require the torque at the wheels to be defined for each gear, rather than at the crank.

 

[ronk]

Keep the RPM at the +ve side of the torque curve?
I think thats what Maniac meant as well.

 

[Bubble1982]

There is, but what Bubble1982 is requesting would require the torque at the wheels to be defined for each gear, rather than at the crank.

That's right - you can get standard torque curves pretty easily like this one from AC Scnitzer...

http://preisliste.ac-schnitzer.de/acs_kpl/hp_kpl/bmw_engl/preisliste_ausgabe_motor_daten_engl.php?Modell=Z4-E89&Teile_Nr=110089320&ID_NR=5580

But I think this is only showing the position across the rev range at the crank.

Theoretically with curve for 0-7000 revs for each of the gears (6 curves in total), showing torque at the wheels, combined with the gear ratios to calc how much the revs drop on each gear shift, you could tell almost exactly where the optimum changeover is from 1st - 2nd, 2nd - 3rd by ensuring that each changes drops you to equal or higher torque at the new revs level.

Interestingly, there's a video of an autos35is accelerating from 0-262 kmh. Set in sport mode it changes pretty much bang on the redline for every shift - no reason for it to hold to the redline unless it gives the best acceleration.

http://www.youtube.com/watch?nomobile=1&v=fRVs1iCitzo

For all my theorising it could well be that simply holding to the redline gives the best outcome! - I need to get out more! Lol

Thanks for all your thoughts.

 

[bcworkz]

Yes, for most street legal cars, the optimal shift point is redline. It seems only cars where the torque falls off significantly and somewhat early, combined with relatively close gear ratios does the optimal shift point occur earlier, and typically not that much earlier. Also usually only for the higher gears, which would be illegal speeds in most locations. So even with cars where the optimal shift point is not redline, the acceleration loss from running to redline anyway would be quite small.

 

[Maniac]

While I understand all the whys and wherefores my arse tells me to shift earlier as it just feels strangled after 6000rpm and there's a hell of a kick when the next gear engages (DCT)

 

[ronk]

While I understand all the whys and wherefores my arse tells me to shift earlier as it just feels strangled after 6000rpm and there's a hell of a kick when the next gear engages (DCT)

I have not seen the graph but would say it is producing less torque at 6k than at 5k .

 

[ralphfcooke]

I generally find that the best change point, going up, is at the revs that drop you onto the peak torque in the next gear.
This may vary depending on the gear ratios.

 

[ronk]

I generally find that the best change point, going up, is at the revs that drop you onto the peak torque in the next gear.
This may vary depending on the gear ratios.

Thats what I said
ie change up at the top of the curve and when the result is the rpm drops into the strong bit of the torque curve.

In reality tho, you will not be able to measure the difference!

 

[wparente]

I generally find that the best change point, going up, is at the revs that drop you onto the peak torque in the next gear.
This may vary depending on the gear ratios.

Thats what I said
ie change up at the top of the curve and when the result is the rpm drops into the strong bit of the torque curve.

In reality tho, you will not be able to measure the difference!

I do road racing in a very tightly controlled spec class (not a Z4) where I have experimented with shifting at different points on the torque curve while along side my competitors going down a straight. Although I can say I can see a miniscule difference in acceleration for 500 to 1000rpms in the upper part of the torque curve it sure isn't much. The overwhelming difference is in braking and the transition through the corners and acceleration on the exit of the corner prior to the straight .

 

[bcworkz]

Yes, most car's torque falls off precipitously the last 1000 some rpm before redline, or even earlier, enough that it is noticeable to nearly anyone. However, when considering torque at the wheels, despite the falling torque, unless your gear ratios are fairly tight, you still are getting more torque to the wheels in the lower gear. You will gain a minuscule bit of extra acceleration by holding off shifting beyond what feels right.

Intuitively, it makes sense to shift so the next gear falls into the peak torque range. It certainly feels good. But if the peak torque is fairly low down, you end up severely short shifting, and there will be a noticeable difference in a timed run. Basically, you always get better acceleration in lower gears, so you should use it as long as you can. It gets more complicated because torque output varies, but as long as you use a good part of the available rev range, exactly where you shift probably makes little difference. I suppose it wouldn't be too difficult to calculate the difference from shifting at say, 5000 instead of 6000, but I'm just too lazy. I'm confident it will be small. It may matter in a drag race, but for anything else, other factors are more important, as wparente points out.

 

[ChuckB]

Just to add more to the discussion, the 35i make max torque around 2,400 RPM and is fairly flat until around 4,000 RPM where it starts to fall off a bit, but levels out until around 5,600 RPM where it tapers off quickly[Orange line], however the horsepower [Red line] peaks around 5,600 RPM. I think shifting up so that you drop to around 4,000 RPM would run pretty good. the blue line is horsepower at the wheels after parasitic losses [Green line] of almost 60 horsepower take their bite out. the chart is from an "untuned" Z4.

 

[Maniac]

That graph bears out the 'feel' of power I get in mine. Hence early shifts. But as others say, dropping to more power early could be short sighted given you're then in a taller gear.

 

[bcworkz]

The graph is useful, thanks. I'd developed a calculator for this, so finding shift points based on this is easy. The hard part was finding gear ratio data. I used 4.71, 3.14, 2.11, 1.67, 1.29, 1.0. Looks a little odd, if someone has better data, I can easily redo the calculation.

Believe it or not, the optimal shift, despite falling torque, in 1-2 and 3-4 is redline. The optimal shift in the remaining gears is 6500. This is based strictly on mathematical calculation of relative torque at the wheels. I would probably shift at a bit over 6000 and call it good. I hate bouncing off the rev limiter.