[Question]need some more Gear ratio lecture please

[Pliskin856]

Hello everyone,

To start off my questions, i better mention that i have already read through the gear tuning section of Mr. Scaff's tuning guide, Mr. Grenadeshark's write-off and some other gear threads in GTplanet.



Now i have learnt 3 methods to set the gear ratio, first is the Mr. Grenadeshark's calculation method; second is the 'max final drive and lowest the slider' method; third is using any existed tuning programs. But because i want to learn how the stuffs actually works therefore i would like to choose the first two methods, but some problems poped up.

When i used the 'max final drive' method, it seems dosent works too well on cars which had peak hp located in the middle, or drop off rapidly at high rev. So, i go for Grenadesharks method.

To make up the scenerio, this is the data that i am using:

Image1 showed the car's power curve. sorry about the blur it is 258 ftlb@4300rpm and 250 hp@5900rpm.

Image2 showed the car's spec

Image3 showed the car's original gear ratio

Ok MY QUESTION IS [finally...] where will you, as an expert of gear tuning, decide the gear shift point AND what rpm should it be for the next gear?

I was thinking if it should be shifted up at 7000rpm? and locate the next gear rev at around 4300-4800rpm? But then the gear ratio became very ugly after it was calculated and have no way to get fit into the game no matter how i slided the slider.

The CORE QUESTION of this thread is i hope someone can expalin abit more to me, when the power band is not all the way up to the rev limiter, how should i set up the ratio?

I know it could be my bad didnt read carefully in other professional's thread. But i really hope that someone can tell me where i did get wrong.

Beside Mr. Grenadshark's method, any other theoritical methods and calculations are welcome, i am eager to learn and i appriciate you to give informations, thanks

 

[Master_gamer438]

Not sure what car your using, but at 250hp is fairly low, I would keep the gears closer between 5,000RPM and redline. In a car with such a low amount of power I wouldn't shift early. Simply not enough power to go through RPMs quick enough even with the decent amount of torque. This is just a guess as i'm usually between 500PP to 600PP (430HP to just over 700HP)

 

[esoxhntr]

those graphs are almost useless.. they represent the shape of the curve but other than that they don't tell you much as they are out of scale.

hp=tq*rpm/5252
tq=5252*hp/rpm

so:

car is making 250hp & 222lb/ft torque at 5900 rpm
211hp & 258lb/ft torque at 4300 rpm

we can guesstimate the car makes about 230hp & 237lb/ft torque at 5100. you've just plotted 3 points on the power curve.

with a little math, you can figure out the average power in between that RPM range and you can get a pretty close number for the power at any given RPM in that range. and remember at 5252 rpm, HP=TQ.

this is actually a fairly flat curve over that RPM range that falls off rather quickly past the RPM peak. you can utilize this 'wide' power band by allowing for more RPM drop between gears (more space between them) and using a higher FD then say... a honda civic whose narrow powerband would require more tightly spaced gears and a lower FD. remember, 6th gear 1.000 with FD of 3.000 is the same as 0.5000 6th gear with a FD of 6.

without doing the math, I would guesstimate the shift point to keep the car in the highest average power range would be about 6100 (& there's a bit of guesstimation with power numbers past the HP peak).

power to the wheels in each gear (or how quickly a car will accelerate) is tq*gear ratio*fd.. since FD is a constant, you can throw it out and just use tq*gear ratio. armed with 3 (ideally more points on the power curve) and excel, you can figure out how much power is going to the wheels at those 3 (or more points) and figure out when to shift and how long to make each individual ratio. you want to shift at the point where "power to the wheels" is greater in the next gear after the RPM drop than it is in the current gear. this should get you a pretty good gearbox for drag racing where the object is to go from 0-xxx. Circuit racing IS different because you are not accelerating from 0-xxx (other than the start) and sometimes you may want to make compromises to make best use of the gear box at the speeds you will be seeing. For example on Nurb you are primarily accelerating from 75-140 (or whatever). On shorter/tighter tracks, that may be 40-100.

Here's how I'd tune this car on Nurb:
I would tune 1st to not spin or bog at WOT (if possible). I would tune top gear to be at the HP peak (5900 rpm) just before you jam on the brakes on the back straight. After that, I tailor my gearing to where (speed) I spend most of time accelerating out of on a track. I would tune 2nd to drop to 4400-4500 and make each gear progressively shorter after that to take advantage of gearing because, other than the start and two 2nd gear corners, I will primarily be accelerating out of 3rd or 4th gears through to 5th-6th (if equipped) .

 

[esoxhntr]

double post

 

[Johnnypenso]

I generally ignore the torque curve when setting up the transmission. In this case you want to throw a blanket around the 6000 rpm mark and shift up around 6500-7000 rpm so the rpm drops back down to 5500 rpm or so, give or take. That way you are always around the peak of the power band. If you use the redline for shifting you could be losing significant power on many cars, but some prefer a redline shift for maxing power.

If this car is around 430 pp then you'll probably get what you want by setting the top speed to around 250 km/h.

 

[Pliskin856]

Sorry about late reply.Thank you and I appreciate the ideas that you guys provided.

Replying to Mr. Master_Gamer438, this is a Ford Focus from dealership.
Replying Mr. Esoxhntr, nur is the place that the car to be put in, so I would like the top speed to achieve around 240-250km/h (149-155mph).

Unfortunately the game could not provide more info for the graph (and I don’t think any game would), many errors and uncertainties occur in the following works. But anyway I think I should show some homework to attract people to correct my mistakes. So here we go:

Now the project is planned to have the car (data provided in Image1-3) to achieve 240-250km/h (149-155mph), while gears are estimated to be shift at about 7000rpm and drops back to about 5500rpm.

Takes Mr. Esoxhntr’s idea, I used the equation following equation to computer the horse power and torque at 4300rpm and 5900rpm (since the toque@4300 and the hp@5900 are provided by the game):

Hp = tq*rpm/5252

So I got the same result as Mr. Eso mentioned, car is making 211hp & 258lb/ft torque at 4300 rpm250hp and & 222lb/ft torque at 5900 rpm.

By using these results, I used Excel to plot a simple linear graph and the following data is obtained. Noted that in this stage an error has occurred as the power band was being ‘assumed’ as a linear slope (which is not). But it should be better than just guessing around. As you can see in the following data, hp @ 5252rpm is not equal to torque @ 5252 rpm.

rpm hp tq
4300 211.11 258
4400 213.55 255.75
4500 215.99 253.5
4600 218.43 251.25
4700 220.87 249
4800 223.31 246.75
4900 225.75 244.5
5000 228.19 242.25
5100 230.63 240
5200 233.07 237.75
5252 234.3388 236.58
5300 235.51 235.5
5400 237.95 233.25
5500 240.39 231
5600 242.83 228.75
5700 245.27 226.5
5800 247.71 224.25
5900 250.15 222


When looked at Image1 again, you may see a kink in the hp curve when is dropping. I simply used some basic, non-accurate method, which is, ruler plus right angle measures, to predict that point is about 6895rpm. After the kink a more rapid drop of hp was occurred. Therefore I set up the shifting point at 7000.

According to Mr. Scaff’s tuning guide, rpm in the next gear should produce more torque (and power) than current gear or else it is pointless to shift up. Therefore I estimated the shifting point at about 5500rpm, which should produce more power than in 7000, assuming the car stays in the average peak power band while shifting up.

Then the following part is to used the method mentioned by Mr. Grenadeshark, to calculate the gear ratio by using the final drive and the last gear ratio (in this case the 6th gear) provided by the default customised gearbox.

Default customised gearbox 6th gear: 0.859
Default customised gearbox final drive: 4.000
Estimated shift up point: 7000rpm
Estimated rpm in next gear: 5500

Gear high/ Gear low * shift point = new RPM in next gear

Before I keep going could someone please point out for me if I done wrong? Thanks.

If the information above is corrected, the following calculation would be resumed, template for the 6th to 5th gear ratio would be:

0.859 / 5th gear * 7000 = 5500
5th gear = 1.093

1.093 / 4th gear * 7000 = 5500
4th gear = 1.391

Is it correct?

 

[esoxhntr]

hp @ 5252rpm is not equal to torque @ 5252 rpm.

It is.. for every car, every motorcycle, 5252 is the point where HP=TQ. Don't ask me why, it just is. THe reason it doesnt in this case is because we are assuming the power/tq curves are perfectly straight/linear.. which they are not. Regardless, your result is close enough.

According to Mr. Scaff’s tuning guide, rpm in the next gear should produce more torque (and power) than current gear or else it is pointless to shift up.

It's not torque at the crank (engine) that determines when to shift, it's torque at the wheels. Torque @ wheels is Engine Torque & Gear Ratio & Final drive (there are drivetrain losses as well but lets assume they are the same for each gear). So multiply your torque figures at each RPM by the gear ratio for each gear.. (we don't need to use FD because FD is the same for each gear) then decide when to shift based on your new "torque at the wheels" numbers for each of your gears.

You're well on your way to figuring this out.

 

[Master_gamer438]

Not sure if you tried this, but sometimes I use the data analysis, it'll show your video replay, give you real time graphs of acceleration, deceleration, RPM, Torque, Lateral and longitude G-force. Then you can save another replay, upload it and compare graphs simultaneously. It might help you fine tune your gearing/handling. I use it sometimes, but for the most part I'm pretty well rounded at gears. I think data analysis is in arcade mode, and online/offline physics differ slightly but its not a crazy difference but will get you going in the right direction.

 

[esoxhntr]

Have a look at this spreadsheet

https://docs.google.com/spreadsheet/ccc?key=0Amn-vtcV_6oKdE5UNkUtRVlrOTl2OEs2czR0ZTdBTGc

I took your data above, added some gear ratios (they are not real ones just ones I made up) and plotted additional hp/tq past 5900rpm based on the assumption that HP decreases at twice the rate after peak than it increases before peak. ie. the line decreases at twice the speed then it increases.. it looks to be somewhat close to the graph you posted.

Let's take a look at 1st and 2nd gear. Immediately with the data we have, we can see that before 6300, 1st gear will pull harder (more torque at wheels) than 2nd at any RPM with these gear ratios.

You will see that shifting at 5900 in 1st (721 to the wheels) will drop you to 4538 in 2nd (~633 to the wheels). You could shift there but why? You still have 5900-6300 in 1st that will pull harder than anything in 2nd. You would want to go past 5900 rpm in 1st gear before switching to 2nd - 6300 rpm maybe even 6350. If you did that, you'd drop from 6300 (624.8) to 4846 (~615).

Copy the spreadsheet and play around with the gear ratios.. pay attention to the gear ratio(s) effect on torque at the wheels and RPM drop between gears.

Hope that helps you understand what i am trying to say. (hey you wanted a lecture)

 

[Pliskin856]

Just wondering how would you determine the shifting point in the spread sheet, like 1st is 6300 then 6200 6200 6200 then 5th is 6100? Because i also see something similar in Mr. Grenadeshark's thread, which he said 'the effect of gearing':

"Example. Car A makes 286 hp @ 7200 rpms with a 9500 redline. It has a flat powerband up to 8900 rpms. It has substantial drop off of power after 8900 rpms. This means that in reality, we will be working with 1700 rpms in our transmission tune. But, knowing the effects of gearing, we will have 2500 rpms (why...?)to use in 1st gear, 2300 in 2nd, 2100 in 3rd. This is because of the effects of gearing on the lower gears overcomes in most cases the power drop off. Your car may vary. If your HP line drops straight down after the peak, only go to the redline in 1st year and drop back to the power drop off in the next gears."

"So, basically our car is going to look like this:

1st-2nd shift will occur at 9500 redline
2nd-3rd shift will occur at 9300
3rd-4th shift will occur at 9100
4th-5th shift will occur at 8900"

If this is the case, which means the template of my gear ratio calculation is wrong, because i did not include this 'effect of gearing' things in... more lecture please.

 

[esoxhntr]

re-read my last two posts and look at the spreadsheet again, all your questions are answered there.

yes, you need to take the 'effect of gearing' into account.

 

[nomis3613]

Hi, I'm late to the party but just wanted to throw my 2 cents in.

The point of gearing is to keep hp as high as possible at all times. Lets say your peak power is at 6500rpm, then your gears and shift points should be aimed towards staying as close to 6500rpm as possible.

A simple way to set shift points is working out what rpm will have the same hp in the next gear after you shift (if you like being scientific, feel free to use rulers and set squares on your TV to work this out from the power curve!). Since your power curve has roughly the same gradient before and after peak power (ie it is not steeper as it drops off towards the redline), it makes working out the shift points easier: just make the difference between your revs and peak hp revs constant (eg if you shift at 7000 you should end up at 6000 in the next gear, shift at 6800 = end up at 6200). Basically, if there is rpm drop into the next gear, you need to rev out a bit more before shifting (to compensate). Other people will correctly point out that this is an approximation, but IMHO it's close enough for the purpose of circuit racing and spreadsheets are overkill.

My method is to set 1st all the way left and then reduce the final gear until rpm juuust stay above peak torque when launching from a standing start. Then set top gear all the way right and space out the gears evenly in between. There is some tweaking to fine tune it for each track, but that's the gist of it. For top speed tuning, you will want to make sure that the top speed is achieved at peak hp in top gear. Otherwise (ie for normal circuit racing) you will be more concerned about having an even spread of ratios and not needing to gearshift at an awkward point in a corner.

Hope this helps.

 

[esoxhntr]

nomis' approach is a good one, and a decent way to setup a tranny quickly. it will get you a decent tune.

where it falls short is taking into account the width of the powerband. the point of gearing / shift points is not to keep rpms as close to peak HP as high as possible but rather keep the average power to the wheels over time (RPM) as high as possible. There's a distinction there.

Also, 'evenly' spaced gears are not necessarily ideal (but again, they will get you close). You do want the gears to get closer together the higher you go to take advantage of gearing.

 

[Johnnypenso]

Has anyone ever tested out different theories on transmission settings on the Route X drag track? While I do spend a little time on tranmission tuning, I don't believe that beyond an approach like Nomis's, which is pretty much what I do, there is much to be gained beyond a tenth or two per lap.

 

[nomis3613]

Has anyone ever tested out different theories on transmission settings on the Route X drag track? While I do spend a little time on tranmission tuning, I don't believe that beyond an approach like Nomis's, which is pretty much what I do, there is much to be gained beyond a tenth or two per lap.

Good idea, I'll try to organise such a test with some drag buddies (take a setup of theirs, slap on my "rough and ready" tranny tune and then see how much slower it is).

Totally agree Johnny. I suspect that- aside from the launch- the difference will be minor in the context of lap times.

 

[nomis3613]

Ok, time for some numbers. I tested 0-400m times for a Mustang GT350R. With a supercharger, power is almost flat from 5000-6500rpm then it drops off quickly after this. Firstly I shifted at the optimum point (6500rpm). Then I repeated with shifts at 7200, this means I was running 700 rpm outside of the power band on an engine where this will have more effect than most other cars in the game.

6500 rpm shifts: 11.72
7200 rpm shifts: 11.73

 

[esoxhntr]

nomis, if you'd like send me a pm with the tune for the mustang (ie. what mods / hp / pp)

i'll then sent you the gear ratios/shift points that i would come up with for 1/4 mile racing. i dont profess to be a drag racing master but i am curious if i could top those times (or make it worse lol). if you are using the stock gear box then obviously just the shift points.

 

[CSLACR]

It is.. for every car, every motorcycle, 5252 is the point where HP=TQ. Don't ask me why, it just is. THe reason it doesnt in this case is because we are assuming the power/tq curves are perfectly straight/linear.. which they are not. Regardless, your result is close enough.

HP and torque do not always meet at 5252 in GT5, that is certainly a fact.


I'm well aware real world physics mean that they always do, but I've seen cars (can't remember, but always looking now) where torque and HP were the same at 7300 RPM, according to GT5 on the one car.

While I'm here, I'll tell you why: Because HP is torque X rpm, / 5252.
200lb-ft @ 2000 rpm = 400,000 / 5252 = 76.16HP.
200lb-ft @ 5252 rpm = 1,050,400 / 5252 = 200HP.


BTW based on that graph 6500rpm is probably around optimal shift point, depending on gear length of course.
5200-6500rpm should be about ideal there.

 

[esoxhntr]

HP and torque do not always meet at 5252 in GT5, that is certainly a fact.

I'm well aware real world physics mean that they always do, but I've seen cars (can't remember, but always looking now) where torque and HP were the same at 7300 RPM, according to GT5 on th one car.

I have never come across where they would not meet there in gt5. I'd love to be proven wrong. We can't just look at the lines because the two are not to scale.

 

[esoxhntr]

HP and torque do not always meet at 5252 in GT5, that is certainly a fact.

I'm well aware real world physics mean that they always do, but I've seen cars (can't remember, but always looking now) where torque and HP were the same at 7300 RPM, according to GT5 on th one car.

I have never come across where they would not meet there in gt5. I'd love to be proven wrong. We can't just look at the lines because the two are not to scale.

While I'm here, I'll tell you why: Because HP is torque X rpm, / 5252.
200lb-ft @ 2000 rpm = 400,000 / 5252 = 76.16HP.
200lb-ft @ 5252 rpm = 1,050,400 / 5252 = 200HP.

i'm well aware of the formula (i posted it above) and that HP is a measure of torque over time (RPM) but who or why the formula is what it is eludes me.

 

[Johnnypenso]

Ok, time for some numbers. I tested 0-400m times for a Mustang GT350R. With a supercharger, power is almost flat from 5000-6500rpm then it drops off quickly after this. Firstly I shifted at the optimum point (6500rpm). Then I repeated with shifts at 7200, this means I was running 700 rpm outside of the power band on an engine where this will have more effect than most other cars in the game.

6500 rpm shifts: 11.72
7200 rpm shifts: 11.73

Interesting:idea:

For anyone interested, you'll find an explaination of the 5252 constant in the TQ/HP formula here:

http://www.revsearch.com/dynamometer/torque_vs_horsepower.html

 

[nomis3613]

nomis, if you'd like send me a pm with the tune for the mustang (ie. what mods / hp / pp)

i'll then sent you the gear ratios/shift points that i would come up with for 1/4 mile racing. i dont profess to be a drag racing master but i am curious if i could top those times (or make it worse lol). if you are using the stock gear box then obviously just the shift points.

Esoxhntr, sorry but I didn't bother writing down the settings and I've messed with them since (for some online racing). Sorry.

Drag racing isn't my cup of tea, I only use it to test traction effects for circuit tunes. Thanks heaps for the offer though.

 

[CSLACR]

I have never come across where they would not meet there in gt5. I'd love to be proven wrong. We can't just look at the lines because the two are not to scale.

I'm sure I can find one, np.

I'm talking peak HP/TQ numbers that happen at mathematically impossible RPM's. The graph can be debated till the end of time, the only important thing to know about it, is that it's a goofball feature, being as inaccurate as it is.

I'm well aware of the formula (i posted it above) and that HP is a measure of torque over time (RPM) but who or why the formula is what it is eludes me.

The formula exists to provide an actual measure of power while driving, since the number "200 ft-lbs of torque" means absolutely nothing in terms of performance. I have more than 200 ft-lbs of torque in my arms alone.
The calculations exist because someone figured them out.
The purpose they serve is torque per minute.

I'm expecting to hear you know all this, but if you do, I'd love to hear why you're asking.