Poll & Discussion: Spring-Rate Tuning - "Real" values vs. Custom values.

[Kurei]

As the title shows, the focus of this thread is a poll & general discussion of Spring-Rates, and the tuning/selection of them by me, you & fellow GTP Members.

---------------------------------------------------------​

The reason I've decided to open this thread & poll (besides searching and finding no recent relevant threads on spring rates specifically), is part discovery & part curiosity.

I've recently re-vamped the entire suspension set-up on 1 of my favorite Japanese 'Sport' Cars in GT6, The MK.IV Toyota Supra (same as in my avatar).
The re-vamp included using 'Real' spring-rate values of actual, publicly available KW Suspension Variant III Coil-overs for the JZA80, converted from found lb/in ratings to GT6's 'most likely' kg/mm value, which were then 'punched into' GT's car tuning menu, and then tested for (subjective) handling through my DFGT at Nurb' up to the T1 marker.
Needless to say,the Supra's overall handling improvement from the previous rates (which must've been too weak, especially in the rear) completely changed the behavior of the car for the better.

Unfortunately, I didn't find anything (other than some function descriptions) to indicate what damper values (Compression/Extension) I should use to simulate the V.III coil-overs in GT6, and have been slowly tweaking here and there to find that 'perfect' balance for my preference (which with the actual KW's, is part of the adjustability).

---------------------------------------------------------​

Besides that initial experiment, I've been trying the same approach with a few other cars and seeing similar results. But that got me thinking, "How many others have thought of this ?, especially with the inclusion of KW's tech in GT's suspension model.?"

All of which eventually led me to making this post & thread. I'm curious how many GTP'ers use actual 'real' values (KW or otherwise), how many don't & just 'wing-it', and so-on. I've tried to include as many options/variables in the poll answers that I could think of.

I know we have a few tuners who do use actual 'real' values, along with the Replicas Thread full of... well, Replica cars, but I'd like to see the more individual basis of the GTPlanet community. (Tuners feel free to represent your individual garages)

And for those curious, I found the MK.IV Supra KW Variant III spring-rates here

Feel free to share & source other 'values' found!
​

 

[rams1de]

In GT5 and the first couple of months of GT6, I used weight distribution to determine front/rear balance and then increase/decrease the spring rates in proportion depending on vehicle weight, downforce and tyre grade. So option B, originally.

Because the effect of springs changed with the new game and GT6 suspension settings seem to have a more subtle effect than GT5, I've since took to using springs more and more to address understeer/oversteer balance. So now I use the stock settings as my starting point and test softer/stiffer, in proportion, trusting feel to determine what I think is the best rate for the track before making final adjustments to help address understeer/oversteer if needed. From the options that's probably winging it?

I've tried real life track settings that have been recommended to me for a BMW M1, Integra and S2000. The real world settings felt pretty good but were better after a few tweaks. The best example of why, would be the settings I was given for the S2000. They were described as 'aggressive' and based on an option available from Spoon with high spring rates and high damper settings. The car was fine round Tsukuba but as you might imagine, very challenging to drive at the Ring - unsuitable in my opinion.

The thing to consider with real world settings is that due to cost and regulations, the tuners may not be able to set the spring rates exactly as they would like in an ideal world. Therefore some real life spring set ups are going to be a compromise that require other means to address imbalance in the car.

 

[F1Racer68]

I selected the "random" option on the above poll, but in reality my technique is not really random. There is a logic behind it and it is very much what @rams1de described above.

I will start with the default or stock values and determine how the car feels on those. Once I have formed an opinion on how the car handles on the default settings I will begin to adjust the spring rate on one end or the other to address specific issues and to achieve the feel I am looking for from the car.

 

[Kurei]

Hmm, I wanted to change "Increase/Decrease rates randomly" to "Increase/Decrease by feel" (sounds more professional), but it seems I can't do that. Or I just haven't figured out how.

I've voted my answer, which is "Mix", of B/D & now A.

Hoping to see more input & votes yet !

 

[Ronald6]

While I love Replicas (shoutout to @Ridox2JZGTE ) I don't have the time/skill/motivation to lookup real world settings. Like most people on GTP (esp. In the tuning section) after a few years (hah, few, more like a dozen+) all I need to do is a lap or two, often just a corner or two and I can adjust the car from GT-stock to decent. A few hours late I might even have something better than decent.

I don't think it's random or by body weight, I'll use the elusive term "experience. " I start off with what experience has shown me and then go by feel and lap times. Sometimes I might even break out the Data Logger.

In my "experience" real world settings can produce a wonderful feeling car, but "tweaking" can produce a fast car.

Good poll, I'm looking forward to the results as they come in. 👍

 

[Ridox2JZGTE]

While I love Replicas (shoutout to @Ridox2JZGTE ) I don't have the time/skill/motivation to lookup real world settings. Like most people on GTP (esp. In the tuning section) after a few years (hah, few, more like a dozen+) all I need to do is a lap or two, often just a corner or two and I can adjust the car from GT-stock to decent. A few hours late I might even have something better than decent.

I don't think it's random or by body weight, I'll use the elusive term "experience. " I start off with what experience has shown me and then go by feel and lap times. Sometimes I might even break out the Data Logger.

In my "experience" real world settings can produce a wonderful feeling car, but "tweaking" can produce a fast car.

Good poll, I'm looking forward to the results as they come in. 👍

Real life values and custom tune both works well 👍 I'm not only strictly using real life spring rate, often the real life spring rate that I'm going to use ( stock spring or coilover springs setup ) is not reachable, either too low or too high Why no full range PD for all cars ? Afraid of funky bouncy / jiggly cars ?

When fine tuning custom spring rate, I set the value based on many factors, from tires used, the conditions of the road surface ( smooth or bumpy ), the car weight distribution, the car handling balance and real life values that works as reference. Some great example are GT3 and GT4 cars, Camaro GT3, McLaren MP4-12C GT3, 370Z GT4, M3 GT4 and some others

Sometimes I have to improvise, when the rate that I set as target ( replica ) are too high ( over 20kg/mm for example ) and not available in GT6, I used the ratio ( between front and back ) and set the highest possible value as starting point.

After building so many replicas, I can say GT6 often replicates the real car spring rate reaction quite well. Some of the good example are Gallardo SL, Nissan GTR, MK IV Supra @Kurei, KPGC10 GTR and Alfa Romeo 1600 Giulia Sprint. The KPGC10 GTR grip/drift replica that I built uses similar approach in spring rate as IRL, and it drives as expected in GT6. Many of the coilover setup ( rate - front higher, rear higher or same rate ) that I used on the same car model gives similar effect as in real life in terms of handling balance, a small example are Honda NSX, R32 GTR, E92 M3 and some of FF Honda like Del Sol, Civic, Integra and CRZ., as well as VW Scirocco.

In the past 6 months I have been building HKS cars ( very detailed specific build ), some are demo cars, some are stock cars with HKS signature Hipermax coilover. I tried to be as close as possible, getting as much data I could, from test car corner weight, alignment ( both sets - test car and OEM ), ride height range ( highest, lowest and test car ). These gives great challenges as some cars are setup with lower front, which means understeer with GT6 reversed effect of ride height, and with the usual Japanese tuner tendency for high spring rate and aggressive rate setup ( front higher or rear higher ), I have to make the best use of other settings like damper, ARB and LSD.

The HKS Hipermax Accord Euro R CL7, Toyota bB NCP31, and HKS CT200MR ( Evo IX CT9A ) are nice example of real life spring rate that works. Check them out, and the CT200MR is very good for a car with front heavy distribution and lower front height and uses no custom LSD ( AYC rules )

 

[Thorin Cain]

I've voted B ( Calculations based on In game weight/distribution.) although on a lot of the earlier tunes I did would come under D (Ramdom increases/decreases.)

I found a thread by @BlueShift a while back (Sorry, cant find it to link.) which explained the formula for calculating the spring rates as (Car weight x { weight distribution/100}) divided by ride height.
Example: Car weighs 1000kg, weight distribution is 45:55,
ride height is set at 75mm front and back.
For front springs (1000x{45/100})/75 or (1000x0.45)/75
=450/75 =6Kg/mm
and the rear (1000x{55/100})/75 or (1000x0.55)/75
=550/75. =7.33Kg/mm

I used this from then on. I find it really helpful any generally quicker to find the results I'm seeking that way rather than randomly changing values until I'm happy.

I usually use the values calculated as a base and also use factors like track type and tyre selection to make adjustments, I will typically calculate 10% increments upto 200% and write them down and use these values when stiffening or softening the springs until I find the figures that work for me.

It is also worth noting that I have also had some decent results with some rear heavy cars by switching the results so the higher spring rates are up front and similarly some front heavy 4WD, FF and a few FR cars with the higher spring rates at the rear.

 

[Ridox2JZGTE]

It is also worth noting that I have also had some decent results with some rear heavy cars by switching the results so the higher spring rates are up front and similarly some front heavy 4WD, FF and a few FR cars with the higher spring rates at the rear.

This is also used IRL Honda NSX-R NA2 factory spring rate is softer/lower at the rear, even with close to 40/60 weight distribution and downforce balance.
Some Japanese coilover maker uses higher rear spring rate for FF cars ( mostly for street / track duty ), the Toda Racing coilover kits are good example.

The E92 M3 aftermarket coilover also often use higher rear spring rate

 

[rams1de]

I've voted B ( Calculations based on In game weight/distribution.) although on a lot of the earlier tunes I did would come under D (Ramdom increases/decreases.)

I found a thread by @BlueShift a while back (Sorry, cant find it to link.) which explained the formula for calculating the spring rates as (Car weight x { weight distribution/100}) divided by ride height.
Example: Car weighs 1000kg, weight distribution is 45:55,
ride height is set at 75mm front and back.
For front springs (1000x{45/100})/75 or (1000x0.45)/75
=450/75 =6Kg/mm
and the rear (1000x{55/100})/75 or (1000x0.55)/75
=550/75. =7.33Kg/mm

I used this from then on. I find it really helpful any generally quicker to find the results I'm seeking that way rather than randomly changing values until I'm happy.

I usually use the values calculated as a base and also use factors like track type and tyre selection to make adjustments, I will typically calculate 10% increments upto 200% and write them down and use these values when stiffening or softening the springs until I find the figures that work for me.

It is also worth noting that I have also had some decent results with some rear heavy cars by switching the results so the higher spring rates are up front and similarly some front heavy 4WD, FF and a few FR cars with the higher spring rates at the rear.

I used a similar formula in GT5 except for the ride height figure which was unavailable. In GT6 however, I would always flip the results for reasons similar to your own, as most cars seem to turn better if the balance of the springs are softer at the end where the weight is.

I also found the formula above can throw out numbers outside of the adjustable range when going near to the extremes of high or low ride height.

 

[Thorin Cain]

I used a similar formula in GT5 except for the ride height figure which was unavailable. In GT6 however, I would always flip the results for reasons similar to your own, as most cars seem to turn better if the balance of the springs are softer at the end where the weight is.

I also found the formula above can throw out numbers outside of the adjustable range when going near to the extremes of high or low ride height.

There have been a couple of times where this problem has come up, normally on race cars with high minimum values for spring rates.

On most of the race cars I tend to end up using between 150-200% of the values calculated anyway which will put a lot of them within the adjustable range, if not then I'd just use the min or max value available and recalculate the other spring rate to match it.

 

[Lionheart2113]

I've voted B ( Calculations based on In game weight/distribution.) although on a lot of the earlier tunes I did would come under D (Ramdom increases/decreases.)

I found a thread by @BlueShift a while back (Sorry, cant find it to link.) which explained the formula for calculating the spring rates as (Car weight x { weight distribution/100}) divided by ride height.
Example: Car weighs 1000kg, weight distribution is 45:55,
ride height is set at 75mm front and back.
For front springs (1000x{45/100})/75 or (1000x0.45)/75
=450/75 =6Kg/mm
and the rear (1000x{55/100})/75 or (1000x0.55)/75
=550/75. =7.33Kg/mm

I used this from then on. I find it really helpful any generally quicker to find the results I'm seeking that way rather than randomly changing values until I'm happy.

I usually use the values calculated as a base and also use factors like track type and tyre selection to make adjustments, I will typically calculate 10% increments upto 200% and write them down and use these values when stiffening or softening the springs until I find the figures that work for me.

It is also worth noting that I have also had some decent results with some rear heavy cars by switching the results so the higher spring rates are up front and similarly some front heavy 4WD, FF and a few FR cars with the higher spring rates at the rear.

That is another interesting way of doing the math, and to be honest I've never really considered ride height into my equation probably due to that being more for CoG (Center of Gravity) and that opens up a mathematical nightmare with all sorts of missing numbers from the game.
Same scenario 1000Kg car with weight distribution of 45:55
1000Kg (x) 9.8Nm = 9800Kg/Nm
Front 45%- .45 (x) 9800 = 4,410 (then you can simply change the comma into a decimal or divide by 1000) 4.410
Rear 55%- .55 (x) 9800 = 5,390 (replace comma with decimal or divide by 1000) 5.390
So it seems that we both have logical reasons for our math, but come up with different numbers. And like you I go up or down in 10% increments and even flip the numbers if it suits. At the end of the day, like others have said, it comes down to feel. I really don't think there is ANY method for finding spring rates, including my own, that can be used 100% of the time with confidence.
It is interesting seeing some of the other strategies though. Good stuff!

 

[Thorin Cain]

That is another interesting way of doing the math, and to be honest I've never really considered ride height into my equation probably due to that being more for CoG (Center of Gravity) and that opens up a mathematical nightmare with all sorts of missing numbers from the game.
Same scenario 1000Kg car with weight distribution of 45:55
1000Kg (x) 9.8Nm = 9800Kg/Nm
Front 45%- .45 (x) 9800 = 4,410 (then you can simply change the comma into a decimal or divide by 1000) 4.410
Rear 55%- .55 (x) 9800 = 5,390 (replace comma with decimal or divide by 1000) 5.390
So it seems that we both have logical reasons for our math, but come up with different numbers. And like you I go up or down in 10% increments and even flip the numbers if it suits. At the end of the day, like others have said, it comes down to feel. I really don't think there is ANY method for finding spring rates, including my own, that can be used 100% of the time with confidence.
It is interesting seeing some of the other strategies though. Good stuff!

Oh no, he's brought Gravity into the equation. Better watch out, the last time I brought that up around here....it didn't go down too well. (Sorry, couldn't help it.)

The ride height bit was new for me too, but it did make a good bit of sense (to me at least) as the results would always give higher spring rates for lower ride heights.

I quite agree. There is no 100% sure-fire ways of doing it. At the end of the day there are always going to be the personal preferences and driving style differences which will effect the choices we make while tuning. But I am a firm believer that using a bit of logic and math is usually a good way to start when considering any physics related problem, real-life or virtual.

 

[Lionheart2113]

Oh no, he's brought Gravity into the equation. Better watch out, the last time I brought that up around here....it didn't go down too well. (Sorry, couldn't help it.)

The ride height bit was new for me too, but it did make a good bit of sense (to me at least) as the results would always give higher spring rates for lower ride heights.

I quite agree. There is no 100% sure-fire ways of doing it. At the end of the day there are always going to be the personal preferences and driving style differences which will effect the choices we make while tuning. But I am a firm believer that using a bit of logic and math is usually a good way to start when considering any physics related problem, real-life or virtual.

Oh no, I'm not diving into the CoG debate either...even though that formula is the first half of it!
Back on springs.....
I am very interested in the results to this and if others have formulas to try out. For me it's mostly set and forget, but always looking for that little bit extra.

 

[DolHaus]

Option 4 was the closest to the method I use but I wouldn't call it random/winging it.
Once the car is built to spec (tyres/power/weight), I start by removing all other suspension variables by zeroing the alignment and setting dampers/ARBs to 1. I will then test drive the car and pay attention to how the car is rotating at various speeds and under various conditions, this data will form my base test point.
First modification will be to adjust the overall stiffness/softness of the car overall by moving both sliders together as equally as possible. I will test drive the car after each adjustment and pay attention to how it is affecting the car under various circumstances and speeds and note down the pro's and cons.
Once I feel I am in the right general area I will start to adjust the springs independently to adjust how the car rotates by altering the relative stiffness between front and rear (I believe this is known as the roll axis but I could be wrong ).
I am looking for the point where the car is starting to rotate slightly too much as I know that when I start increasing the damping the weight transfer will be reduced and stop the tyres being overloaded.

No formulas or maths, just a bit of common sense in regards to weight reduction and distribution and some good old fashioned trial and error 👍

 

[brian wolf]

Option 4 was the closest to the method I use but I wouldn't call it random/winging it.
Once the car is built to spec (tyres/power/weight), I start by removing all other suspension variables by zeroing the alignment and setting dampers/ARBs to 1. I will then test drive the car and pay attention to how the car is rotating at various speeds and under various conditions, this data will form my base test point.
First modification will be to adjust the overall stiffness/softness of the car overall by moving both sliders together as equally as possible. I will test drive the car after each adjustment and pay attention to how it is affecting the car under various circumstances and speeds and note down the pro's and cons.
Once I feel I am in the right general area I will start to adjust the springs independently to adjust how the car rotates by altering the relative stiffness between front and rear (I believe this is known as the roll axis but I could be wrong ).
I am looking for the point where the car is starting to rotate slightly too much as I know that when I start increasing the damping the weight transfer will be reduced and stop the tyres being overloaded.

No formulas or maths, just a bit of common sense in regards to weight reduction and distribution and some good old fashioned trial and error 👍

Thats pretty much how I do it.👍 I leave arbs and dampers at 3 though so I have a negative movement available once springs are set.

 

[F1Racer68]

@Thorin Cain and @Lionheart2113

Do either of you know how the calculations need to altered in order to account for downforce? Like weight and ride height this will have an impact on spring rates.

 

[DolHaus]

@Thorin Cain and @Lionheart2113

Do either of you know how the calculations need to altered in order to account for downforce? Like weight and ride height this will have an impact on spring rates.

In theory, yes.
In practice finding the multiplier for G would be extremely difficult (impossible even) as we have no accurate way to measure spring compression and expressing it as a number.

 

[Lionheart2113]

@Thorin Cain and @Lionheart2113

Do either of you know how the calculations need to altered in order to account for downforce? Like weight and ride height this will have an impact on spring rates.

Well, there is a "real life" way of figuring out how much weight is added due to downforce. I'm afraid to even bring up "real world" because it always leads to disagreements, which is ok, but some people take it personally and then the hair pulling starts. You have done a lot of testing with the data logger and I'm guessing the new MoTeC, how confident are you with the ride height numbers given with those? I haven't done too much testing with either one...yet!

 

[Thorin Cain]

@Thorin Cain and @Lionheart2113

Do either of you know how the calculations need to altered in order to account for downforce? Like weight and ride height this will have an impact on spring rates.

In the thread I found before, the same formula was used but with the downforce figure included in the weight part of the equation.

Say you had a car running downforce of 350/600, when calculating the spring rates just add the front or rear figure to the total weight before taking the weight distribution into account. So for the same car as above it would be ((1000+350)x0.45)/0.75=8.10kg/mm for the front.

No idea if this would be the proper way to do it or not, but it works out ok for me. I'll have a look around and see if I can find that thread again and link it here.

EDIT: Fixing the maths, thanks for the spot @TurnLeft

 

[DolHaus]

Well, there is a "real life" way of figuring out how much weight is added due to downforce. I'm afraid to even bring up "real world" because it always leads to disagreements, which is ok, but some people take it personally and then the hair pulling starts. You have done a lot of testing with the data logger and I'm guessing the new MoTeC, how confident are you with the ride height numbers given with those? I haven't done too much testing with either one...yet!

Neither the Motec nor the data logger gives quantifiable data unfortunately, you could make an educated guess but that's all it would be

 

[Lionheart2113]

Typical! Of course they're not going to give us what we need to figure it out entirely...we might just stumble on some more "oh 🤬" if they did. Theres gotta be some way of getting close...in this game I'd be more than happy with close.

 

[DolHaus]

Typical! Of course they're not going to give us what we need to figure it out entirely...we might just stumble on some more "oh 🤬" if they did. Theres gotta be some way of getting close...in this game I'd be more than happy with close.

Yeah it was a bit disappointing but its better than nothing. There is one tool called the "Friction Circle" which should help in finding out conclusively what's going on with camber 👍
I have a plan for working out downforce effects but I'm not sure if it will work yet

 

[Thorin Cain]

I've had a good look and can't find the original thread...well I found the link and it didn't work, and I can't see it on the forums. Possibly been changed
It was "Spring rate formula." by Voodovaj. This is where I found the information about the formula and taking aero into account but most of it was covered here...
https://www.gtplanet.net/forum/threads/mostly-all-my-dirty-secrets.296571/
Edit: Link to OP, sorry guys.

 

[TurnLeft]

I've had a good look and can't find the original thread...well I found the link and it didn't work, and I can't see it on the forums. Possibly been changed
It was "Spring rate formula." by Voodovaj. This is where I found the information about the formula and taking aero into account but most of it was covered here...
https://www.gtplanet.net/forum/threads/mostly-all-my-dirty-secrets.296571/
Edit: Link to OP, sorry guys.

Just checked it out, I guess I won't be sleeping for a few days , thanks for sharing.

 

[Thorin Cain]

No problem at all my friend, happy reading.

Best advice, use bite-size pieces and plenty of Advil.