E36 Diffs

[Sankekur]

Just keep in mind that swapping the diff will only make the car's in gear acceleration faster, it will have next to no effect if you accelerate from a standing start or low speeds and you have to change gears.

have to disagree with you on that one... unless traction is a problem it will make the car quicker... yes you may have a slower 0-100 time because of an extra change before 100 but this in unimportant unless you specifically racing from 0-100 :rollsmile:

We have tested this on 540i's with 2.93 and 3.15 diffs and it made pretty the change from 2.93 to 3.15 made almost no difference.
Removing 60kg or so from the car made more difference (this is on cars the weigh around 1700kg).

tests where done from 20-220km/h and on uphill downhill and level roads.

You would be surprised how little of an effect torque has on the outcome, for the most part it it will be the car that revs higher and generates more power that will come out on top.

 

[Loom]

So would you say it's worth it if you're only interested in faster acceleration in town, say 0 - 70 km/h, meaning an earlier change only to second gear? Let's say it's a swap from a 2.93 to 3.15.

 

[moranor@axis]

Just keep in mind that swapping the diff will only make the car's in gear acceleration faster, it will have next to no effect if you accelerate from a standing start or low speeds and you have to change gears.

have to disagree with you on that one... unless traction is a problem it will make the car quicker... yes you may have a slower 0-100 time because of an extra change before 100 but this in unimportant unless you specifically racing from 0-100 :rollsmile:

We have tested this on 540i's with 2.93 and 3.15 diffs and it made pretty the change from 2.93 to 3.15 made almost no difference.
Removing 60kg or so from the car made more difference (this is on cars the weigh around 1700kg).

tests where done from 20-220km/h and on uphill downhill and level roads.

You would be surprised how little of an effect torque has on the outcome, for the most part it it will be the car that revs higher and generates more power that will come out on top.

2.93 to 3.15 is only about 5% difference :fencelook: do the same with your 330i and the 3.38 diff :rollsmile:

there are definitely limits to the gain for instance the e46 M3 with a shorter ratio does not seem to gain much acceleration...

if you have some 540 torque curves i can do some simulations so we can see the effect...

 

[Sankekur]

Just keep in mind that swapping the diff will only make the car's in gear acceleration faster, it will have next to no effect if you accelerate from a standing start or low speeds and you have to change gears.

have to disagree with you on that one... unless traction is a problem it will make the car quicker... yes you may have a slower 0-100 time because of an extra change before 100 but this in unimportant unless you specifically racing from 0-100 :rollsmile:

We have tested this on 540i's with 2.93 and 3.15 diffs and it made pretty the change from 2.93 to 3.15 made almost no difference.
Removing 60kg or so from the car made more difference (this is on cars the weigh around 1700kg).

tests where done from 20-220km/h and on uphill downhill and level roads.

You would be surprised how little of an effect torque has on the outcome, for the most part it it will be the car that revs higher and generates more power that will come out on top.

2.93 to 3.15 is only about 5% difference :fencelook: do the same with your 330i and the 3.38 diff :rollsmile:

there are definitely limits to the gain for instance the e46 M3 with a shorter ratio does not seem to gain much acceleration...

if you have some 540 torque curves i can do some simulations so we can see the effect...

It is actually a 7.5% change :mmm: I have actually done tests with the 330i, but unfortunately I do not have a spare 330i around so had to compare to something else (and as interesting as the results were I doubt the meaningfulness)

What I can say from my experiences (as limited as they may be) is that around 160km/h seems to be the transition point between a torque and power dominated performance. That being said the difference that more power makes is much more noticeable than a difference in torque does. (And of course RPM is king) The most extreme would be a case where two cars had close to the same power output but the one had a more that 60% advantage in terms of torque, but this advantage only translated to around a 1 to 1.5 car length lead.
On the other end of the scale would be two cars where the one has around 10% less torque but just over 10% power advantage, where this advantage translated to around a 4 to 5 car length lead.

Unfortunately I don't have the torque curve for the 540i, but you know what they say about simulation don't you?

 

[moranor@axis]

How do you know one 540i was not healthier than the other? if you use the same car and swap diffs on to do the tests it will be more accurate...

from the tests I have done the simulator seems pretty accurate and the butter zone for the e36/e46 2.5l-3l seems to be about 3.38... shorter seems to give diminishing returns...

If i can find some 540i torque curves I will PM you some graphs :rollsmile:

the relationship between peak torque, peak power and performance is not simple to explain... but peak torque should always be produced in the RPM band where the engine is being used...

For racing this is generally in High RPM and a Torque peak at High RPM will produce a High power figure...

So then generally a high Peak Power is more important for Racing then Peak Torque...

peak power is also the determining factor for top speed as when you run out of gears you will need to use your RPM to reach higher speed... if you producing your Torque low down you wont have enough to push the car faster when you go into high RPM...

 

[Sankekur]

True that the one might be healthier than the other, but this is why the test was done before and after the diff swap. It was done 3.15 vs 2.93 then 3.15 vs 3.15 and in both cases the outcome was the same. So for instance the one in the first test with the 3.15 was running worse than the one with the 2.93 casing them to be the same, then the stronger one would be faster when both had 3.15 diffs?

 

[moranor@axis]

same driver? different driver? 7.5% is very small so many things can affect the outcome...

dont know why you so intent that diff ratio does not work when it is one of the most common mods done to track cars... if it did not bring results why would they do it?

 

[328ii]

The time taken to 'run' my E36 328iM from 2000 rpm to 6000 rpm
on an inertia dynomometer was a wopping 2 seconds FASTER
with a
3.15 ratio diff and 235x40x17 tires
than with
2.93 ratio diff and 235x45x17 tires..!

Converting/calculating vehicle speed from above results
using a 3rd gear ratio shows that, theoretically...

[edit]

2.93 ratio with 235x45x17 in 3rd gear
2000 - 49.9 km/h
6000 - 149.6 km/h
6750 - topend @ 279.4 km/h

3.15 ratio with 235x40x17 in 3rd gear
2000 - 44.7 km/h
6000 - 134.1 km/h
6750 - topend @ 250.3 km/h

8?>

 

[Sankekur]

Same drivers

Racecars are different matter (assuming NA) most changes made to a racecar engine will increase output, but the increase will be at the top end, ie torque is moved from the mid range to the top end, so increasing the diff ratio will firstly get the car to where it makes peak output quicker it will also keep the engine running there more often.


Where a diff will enhance "performance" in a road car is if you do a test over 1 km, drive the car in top gear at 120km/h, and floor just as you enter the 1 km, then measure your speed as you exit the 1 km, this is where you will see an improvement by changing the diff.

The time taken to 'run' my E36 328iM from 2000 rpm to 6000 rpm
on an inertia dynomometer was a wopping 2 seconds FASTER
with a
3.15 ratio diff and 235x40x17 tires
than with
2.93 ratio diff and 235x45x17 tires..!

Converting/calculating vehicle speed from above results
using a 3rd gear ratio will show, what..?

8?>

This is exactly what I am talking about the car will be faster, if you don't have to change gears

 

[Serano]

Thanks for all the feedback guys.

After reading all the posts and speaking to some of the forum members, i will either be installing the 3:15 or a 3:23. It just a matter of what i can find now

I would even consider a 3:38...But after getting a 7k quote i was scared off

If i cant get the 3:23, i will try the 3:15. Lets see what comes up..

Needless to say, if anyone has or knows anyone who has these parts, please let me know..

:ty:

 

[moranor@axis]

7k for a diff :roflol: look somewhere else...

Sank the thing is a shorter ratio also makes the gears closer together this means you can keep the car at the correct RPM more easily so it does not really matter where you are producing your torque a shorter ratio will help you keep it there... yes there is a limit to how short you want to go because eventually it will involve too much shifting but its not like we talking super short here...

with my 3.38 from tests I have done it is most definitely quicker with the diff...

 

[Fordkoppie]

same driver? different driver? 7.5% is very small so many things can affect the outcome...

dont know why you so intent that diff ratio does not work when it is one of the most common mods done to track cars... if it did not bring results why would they do it?

Same drivers :fencelook:

They change ratios on track cars to eliminate mid corner shifting and/or having to upshift just before you enter the braking zone, just to downshift again.
Another thing is that if you exit a corner in 3rd gear for argument's sake with both a longer and a shorter ratio, you will have more torque available at the wheels for that short blast till your next upshift when using the latter........ And you know how crucial corner exit speed is for the entire following straight.

However, in a run from 20 - 200 it makes no difference as Sank said.
It just feels faster with the shorter ratio

 

[moranor@axis]

then how do you explain that cars that my car used to struggle with are now easier? am I imagining that they are behind me?

I have been test driving alot of 325i because I am looking for another e46 and they all complete donkeys compared to my 323i...

 

[InFrA007]

While on the topic, looking for a 3.46 LSD for my track car, anybody with spares lying around? :rollsmile:

 

[Fordkoppie]

Case in point - a certain 350 ci Chev V8 with a 3speed slushmatic managed 13.6s and 167km/h on the 400m with a 2.75 LS 9" diff. (just had to select 3rd gear a few meters before the trap)

Swapped the ratio to a 3.0:1 and the 1/4mile time was 13.6 seconds again on the same dragstrip with an exit speed of 169km/h. (had to select 3rd a fair distance before crossing the trap)

Now this particular engine is a torque monster, so revving it higher, you actually lose time.

 

[moranor@axis]

I will concede that a shorter ratio under some circumstances will not make a difference or even loose time (e46 M3 does not gain much from ratio change as i mentioned earlier) but the thread is about a 2.5l not a V8 doing a 1/4 mile... some cars do faster 1/4 mile times because of traction and one less shift with a longer ratio this does not prove much... only that it works for 1/4 mile times...

I have already said there is a limit to the gain to be had with a ratio change and obviously the more torque you produce the limit is lower... a 1.6l might be best with a 4.50 where a 4l will probably be best with something around 3.00...

for the 2.5l (3l results are pretty similar according to the simulations) in question here I have no doubt a 3.23 to 3.46 works well... from simulations and my own real world testing...

maybe i should find some 540i torque curves and see where the sweet spot is for that car... maybe it came out with an awesome ratio to start with?

 

[Muzhar H]

http://www.dragtimes.com/gear-ratio-calculator-mph-speed.php

Just to see howmuch you stand to lose top end!

 

[Sankekur]

If you have two cars up against each other that are identical save for different diff ratios, the one with the higher ratio will deliver more torque to the wheels, but it will also have to shift before the other car meaning it will now be delivering less torque to the wheels until the other car shifts.

 

[moranor@axis]

oh crap...

I just rechecked my simulations on the 323i and Im wrong... my 'stock' 323 in the simulator had the wrong torque curve

going from a 3,07 diff to a 3,38 is not nearly as big of a difference as before now with it corrected... the 3,38 pulls ahead before the change then the 3,07 edges lightly before it has to change... while the 3,38 has a slight advantage its really way to close to call and is more of a drivers game...

Now this result makes no sense to me because my car is most definitely faster since i did the swap so there must be some other factor at play here...

 

[Fordkoppie]

Here is a torque curve for the 540