# Physics spreadsheet - measuring suspension

Discussion in 'Car Modding' started by thoraxe, May 3, 2020.

1. ### thoraxeRegistered

Joined:
Dec 25, 2016
Messages:
128
10
I've rented a CMM arm in order to plot the 3D coordinates (with mediocre accuracy) of all of the suspension points on my car in order to put them into the physics spreadsheet. I've got some measurements but the relative location in 3D space is what's putting me off right now, and the fact that my car has:

* a lopsided / asymmetrical lower front control arm
* a multi-link rear lower suspension

Some observations:
* The X=0 in the suspension tab appears to be something inboard -- is this the vehicle center line?
* The Y=0 appears to be the ground?
* The Z=0 appears to be the vehicle center (front to rear)?

Do all my numbers need to be in relation to these observations above? If so, what's the best way to determine the X=0 and Y=0 and Z=0 for my particular vehicle? I kind of have X=0 figured out and I think I can figure out Y=0 but Z=0 is the one that's more or less giving me pause. Where should Z=0 be, or does any of this matter?

Here is a picture of the front lower control arm:
https://photos.motoiq.com/MotoIQ/Project-Cars/Project-Lexus-SC300/i-tQSKtD6/A

You can see that its asymmetrical. The front inboard mounting point is practically in-line with the ball joint but the rear inboard mounting point is far behind and isn't a straight line to the same point. For the purposes of modeling with the spreadsheets, should I just treat this like it's a pure A arm?

Here's a (not very good) picture of the rear suspension:
https://photos.motoiq.com/MotoIQ/Project-Cars/Project-Lexus-SC300/i-P6WW4zC/A

There is a single lower control arm that's directly in-line with the spindle/ball joint. Then there are two other lower front arms. One controls the toe curve and the other basically operates in a straight line. If you scroll down here it appears to look a lot like the Art Morrison multi-link:

Here is a blown up parts diagram:
https://jp-carparts.com/toyota/partlist.php?maker=toyota&type=321140&cartype=4&fig=4804

The spreadsheet has "toe link" as a measurement -- I think that is exactly what I would do here?

Other questions:

* Inner steering is where the "ball" of the tie rod screws into the steering rack, I assume when the steering is zeroed?
* Pushrod is the damper, and "inboard" is where it attaches to the chassis, and "outboard" is where it attaches to (in my case) the control arm, right?
* Spindle is the center of the spindle? Would this be where the center of the wheel bearing is or something else? On the sample spreadsheet the spindle is offset from both the control arm points and the wheel center...
* Wheel is where the wheel center is?
* Turns lock is clear, but what is steer ratio?
* 1/2 rack stroke - this would be the stroke of one side of the rack from full lock-to-lock?

Vance Le Garde likes this.
2. ### davehenrieRegistered

Joined:
Jul 6, 2016
Messages:
4,587
2,537
Wow! quite the project!

3. ### LazzaRegistered

Joined:
Oct 5, 2010
Messages:
9,442
4,458
I'll answer what I (think I) know, this will be a little condensed as I think quoting separate parts will expand to about 3 pages.

0 X is centreline, 0 Y is the 'floor' of your car (roughly undertray, sideskirts hang below; somewhat arbitrary), 0 Z can effectively be whatever you want as long as everything uses the same reference but I find equidistant between front and rear to be easiest (this may have other consequences, I'm not sure). Then yes, everything should be on the same page (haha!) for things to work. I personally prefer to add spreadsheet pages if I'm entering in from another coordinate system or doing some calculations for the benefit of (or from the result of) other sources or analysis, and then use formulas to get back to the 'vehicle' coordinates in the Susp tab. Set it up once and forget; try to keep adjusting/translating manually and you'll end up with errors everywhere.

Treat the asymmetrical arm like an A-arm (effectively that's what it is; the fancy shape is to fit around other stuff). Steering inner/outer are with centered steering, I'll leave the exact mechanicals to your figuring or someone else's expertise.

Spindle is the (basically vertical) axis that the wheel rotates around when you steer - not 'normal' wheel rotation when the vehicle is moving. The centre of the spindle bolt or its equivalent, in other words. Wheel is the centre of the wheel, yes.

When you turn the steering wheel to full lock (from centre) you may need to turn it 540°, while at the wheel (the one with a tyre on it) the actual steered angle might be 30°. That's a steering ratio of 540/30 = 18. If you measure your wheels when fully turned just take the average to negate ackermann effects.

1/2 rack stroke I take to mean literally half the full stroke - so centre to full lock. In practice I make this my adjustment figure (rather than using an exact reference) to get the desired full-turn angle.

The multi-link rear, you may be better off finding out its characteristics and replicating those with the standard double A-arm for rF2 (or measure everything up, then work out the A-arm equivalent). I haven't yet tried to calculate and implement a multilink, I think rF2 is probably capable of doing it but if you don't get the positions exactly right you'll get strange behaviour, and you may need to find/buy some analysis software to verify the correct positions.

Hope that helps a little, bearing in mind I'm coming at this from a maths/spreadsheet/physics perspective rather than a mechanic's one.

4. ### John R DenmanRegistered

Joined:
Mar 18, 2018
Messages:
420
485
Use the pTool for inputting dimensions, it gives a visual perspective.
It takes a bit of reading and playing with to figure it out and well worth it.
It opens with "default.ini which is empty. pTool is the working folder and when its done it saves a new file to the ModDev/User folder - so when you're ready to save, type over the default.ini with the newname.ini.
https://www.studio-397.com/modding-...n-to-physics-tool-ptool-and-flexible-chassis/

Here are the data points:
// [Subsystem]
// Name="inter"
// Constraint="Chassis_BJ"
// Constraint="Chassis_OJ"

// [Subsystem]
// Name="front"
// Constraint="FL_BJ"
// Constraint="FR_BJ"
// Constraint="FL_HINGE"
// Constraint="FR_HINGE"
// Constraint="FL_FORE_LOWER"
// Constraint="FL_REAR_LOWER"
// Constraint="FL_FORE_UPPER"
// Constraint="FL_REAR_UPPER"
// Constraint="FL_STEERING"
// Constraint="FR_FORE_LOWER"
// Constraint="FR_REAR_LOWER"
// Constraint="FR_FORE_UPPER"
// Constraint="FR_REAR_UPPER"
// Constraint="FR_STEERING"

// [Subsystem]
// Name="rear"
// Constraint="RL_BJ"
// Constraint="RR_BJ"
// Constraint="RL_HINGE"
// Constraint="RR_HINGE"
// Constraint="RL_FORE_LOWER"
// Constraint="RL_REAR_LOWER"
// Constraint="RL_FORE_UPPER"
// Constraint="RL_REAR_UPPER"
// Constraint="RR_FORE_LOWER"
// Constraint="RR_REAR_LOWER"
// Constraint="RR_FORE_UPPER"
// Constraint="RR_REAR_UPPER"

5. ### JokeriRegistered

Joined:
May 17, 2014
Messages:
191
29
@thoraxe Can you show the CMM arm results please?

6. ### thoraxeRegistered

Joined:
Dec 25, 2016
Messages:
128
10
My first attempt likely created just mostly garbage data that doesn't make sense. I have a friend coming over this week that has more experience with this equipment and I spent today trying to find various center lines of the vehicle. We are going to try to play with taking better measurements.

That being said, here's a CSV equivalent of the garbage data that I took with my first attempt which is just guesses at points in 3D space with the CMM arm at some unknown origin.

Code:
```Units: in
Decimal Indicator: .
Date: 5/2/2020 5:46:13 PM

Origin_World;0.0000;0.0000;0.0000
Hub Center - Actual;-15.21401;-13.6428558843071;40.5919058834837
Upper Front RE - Actual;-18.5260495;-10.1333647769479;52.6331884
Upper Rear RE - Actual;-11.2819257651541;-10.7627583608782;52.12594705
Upper BJ - Actual;-15.2922322052808;-8.4603;44.64863
Lower Front RE - Actual;-14.206337225;-19.3280361648614;55.6373188861779
Lower Rear RE - Actual;-1.0987719128894;-19.0797;55.53246647
Lower BJ - Actual;-16.37675;-18.0309442662561;42.3701
Tie Rod RE - Actual;-21.388356;-15.912167;42.6370561
Lower Shock Eye - Actual;-15.63191227;-15.679206828;47.548445218
Shock Upper - Actual;-15.3891882208715;1.06745316261846;52.2623615
Swaybar arm mount - Actual;-17.14072;-15.64635;46.06086086487
Swaybar inner mount - Actual;-27.2876685142879;-13.3953328;56.57317
Brake Lower Bolt - Actual;-12.49945;-15.48787073;42.74564366
Brake Upper Bolt - Actual;-13.4633614406058;-10.4554117623388;43.4043746767406
Shock upper top location - Actual;-15.5130725442256;2.92378840507261;53.0177787846251
Outer rear shock bolt - Actual;-15.5926450819438;4.079771;50.7806917933
Inboard shock bolt - Actual;-13.4473489447694;3.19622916238641;54.1536
Outer front shock bolt - Actual;-17.5551316434353;3.17612;54.25457```

7. ### thoraxeRegistered

Joined:
Dec 25, 2016
Messages:
128
10
Thanks to everyone else that replied as well!

8. ### EmeryRegistered

Joined:
Oct 24, 2010
Messages:
2,800
1,395
If you have the choice, switch to meters for measuring rather than inches, otherwise you'll have to convert them.

Brake bolt locations aren't needed for building the suspension (unless I've missed something). Swaybar mount locations aren't needed, either, but they'll be useful when calculating swaybar spring rate.

9. ### lordpantsingtonRegistered

Joined:
Oct 5, 2010
Messages:
835
66
I converted those points to metric and placed them in 3d space. They appear to be in (Fore-/ Aft+, Up+/Down-, Left-, Right+) notation, if this is a Rear Left.

edit: I think the labels are off. If rear is actually front it would make more sense to me. (Which would swap X to Fore+/Aft-)

You are missing the Inner Tie rod location.
At what ride height were these locations taken?

Last edited: May 4, 2020
10. ### JokeriRegistered

Joined:
May 17, 2014
Messages:
191
29
@thoraxe did you measure the suspension again??

11. ### thoraxeRegistered

Joined:
Dec 25, 2016
Messages:
128
10
Yes I'll post the CAD files and measurements again in a bit.

12. ### thoraxeRegistered

Joined:
Dec 25, 2016
Messages:
128
10
@Jokeri
https://drive.google.com/open?id=1mLHyFXuB2JPSLyHuOwpDsrF_veEOwQuA - IGS file with the combined front and rear suspensions on the passenger side of the car, assumes vehicle symmetry.

https://drive.google.com/open?id=1izbg5b6jVat3sjyzpVYVGNWyRKMulDqz - CSV file with the lists of the various coordinates as they should likely appear in rFactor

I still have to measure the steering stroke and turns and etc, and a few other nits.

Then I have to look at my shock dynos and plug in those values.

Then I have to look at the specifications for my transmission and see if I can figure out the gear teeth counts. I've got the ratios set right, but I pretty much fudged it.

I have the engine figures plugged in but I have them plugged in as if it were an NA engine in spite of being a turbo engine. I don't have sufficient data, really, for the turbo -- boost graph, RPM, etc.

I also have to go measure a bunch of stuff on the brakes.

So many things left to do...

13. ### JokeriRegistered

Joined:
May 17, 2014
Messages:
191
29
Great, i'll look in to it

14. ### thoraxeRegistered

Joined:
Dec 25, 2016
Messages:
128
10
Anyone know:

* What's the difference between bumpstop spring and RisingSpring ?
* My dampers do not have independently adjustable high/low speed compression/rebound. My understanding is that there is one adjustment that affects the entire curve. I suppose this just needs to be user behavior in-game (eg: make changes to both high/low speed at the same time and make sure the amount of adjust is relatively accurate)?
* Does the actual tooth count for gears/layshaft in the trans really matter, or as long as the ratios are right it's good enough? The ratios are easy to find (https://www.speedwaymotors.com/TREM...Magnum-6-Speed-Manual-Transmission,67175.html) but the teeth counts not so much.
* Same question about the differential / ring&pinion

Probably many other questions...

15. ### lordpantsingtonRegistered

Joined:
Oct 5, 2010
Messages:
835
66
IMO the only way to get the true, correct, ratios is to do the work and get the tooth counts. Sure with rounding, lazy ratios match to the 2nd or 3rd decimal, but more often than not these will not match ratios when doing the math from tooth counts. A few months back I invested some time into researching tooth counts for Tremec T-56 & TR6060, specifically for the ViperCC. Just make sure you use bevel to compensate for the input shaft multiplier and it will be spot on. Service manuals and replacement part searches were very helpful. What ratios are you running? I'm sure I've got toothcounts or can find them.

BSS is the Initial springrate of the bumpstop.
BSRS is the additional rate (Multiplied by deflection squared).
Bumpstops were still too linear the last I looked at them. IRL the Force v Displacement curves for my application would have been better with ^3 or ^4 ramping. As a result there is extra travel, but I'm not sure the forces needed to get there are going to be achieved.

ApplySlowToFastDampers=1 //* whether to apply slow damper settings to fast damper settings
LimitFastDampers=1 //n Whether to limit the fast damper rate to be less than or equal to the slow damper rate (actual rate, not numerical setting)

Emery likes this.
16. ### John R DenmanRegistered

Joined:
Mar 18, 2018
Messages:
420
485
Bumpstops are rubber (or polyurethane). Polyurethane is compounded in different hardnesses with different compression values, and tend to have built in dampening where rubber is usually one hardness with little dampening.

Rising Spring would apply to a progressively wound spring common on sedans. The spring rate increases as its compressed where a standard wound spring used on most race cars has a steady spring rate through the compression range.

Different dampers are different in construction. Generally there is a check valve that inhibits fluid flow for each return path that the pintle valve controls, but not always.

On a rack & pinion the tooth count doesn't matter; its the Pitch Diameter that matters. However in rFactor they tend to be expressed as tooth count. Another factor with those gears is Teeth Per Inch, TPI. For example a 10 tooth pinion of 10 TPI is twice the Pitch Diameter of a 10 Tooth Pinion of 20 TPI, and therefore the 20TPI only moves the rack 1/2 as far.

As for ratios vrs tooth counts, ratios are the quotient of ring divided by pinion so as long as the ratio is the same the actual tooth counts don't matter.

On your tire questions, the TBC file is used to define the tread and physical dimensions, while the TGM file defines the carcass. Despite rF2 have the best tire algorithms going tire physics is extremely complex and somewhat chaotic to model. Pick a close example from the pre-defined models and tweak the TBC files, its a lot of trial and error for a newbie.

17. ### thoraxeRegistered

Joined:
Dec 25, 2016
Messages:
128
10
I have the TUET11009 (close ratio):
https://www.grannasracing.com/colle...c-t56-magnum-gm-6-speed-close-ratio-tuet11009

I kinda don't think I'll ever hit the bumpstops, so I'll just do what I can to measure them.

Also thanks to @John R Denman -- good info.

18. ### thoraxeRegistered

Joined:
Dec 25, 2016
Messages:
128
10
Sorry, I don't entirely understand LimitFastDampers=1 //n Whether to limit the fast damper rate to be less than or equal to the slow damper rate (actual rate, not numerical setting)

What does that setting do?

19. ### LazzaRegistered

Joined:
Oct 5, 2010
Messages:
9,442
4,458
@John R Denman John, you obviously have a wealth of real world knowledge and experience, but if you're going to answer questions regarding rF2 you really need to check what it does.

The only 'rising' parameters are the bumpstop spring and damper ones. These define the rates per deflection squared, as lordpantsington said. Nothing at all to do with progressive coil springs.

The only rack in rFactor is the steering rack, and the pitch diameter is used there as part of the 'new' steering system. (not by name, but defined by coordinates)

The TBC needs to have a radius that matches the TGM radius (which is the node geometry data, multiplied by the SizeMultiplier parameters in the realtime section), but the rest of the data there applies only to the AI physics except the treadwidth which is only used graphically. The entire physical modelling of the tyre is defined in the TGM.

It does exactly what it says on the tin. If you set that to 1, when you're in the garage and reduce the slow damper rate to a point where it would be lower than the current fast damper rate, the fast damper will reduce also. If you set it to 0 people can set the fast damper rate higher than the slow damper rate. (as the comment says, this refers to the actual applied rate - not the 'setting number' which is defined by the range parameters)

20. ### lordpantsingtonRegistered

Joined:
Oct 5, 2010
Messages:
835
66

The thing that obfuscates the tooth count for each ratio given is the input shaft ratio. What I had listed for the Viper T-56 was 36/31 (1.161). If I use that it makes R,1-4 easy. And this is what I was talking about in my previous post, while it might look like 1st is 2.66, It actually is a horrific number and goes out to 14 decimal places (2.66413662239089). In the older rf style gears INI format it would look like this, I'll leave it up to you to format to rf2's HDV/special style format if you want to. I left the Dana/Ricardo M44-4 rear ratio as an example so the rear wouldn't be forgotten.

[GEAR_RATIOS]
ratio=(14, 35) // R: 2.903 35/23*23/14*36/31
ratio=(17, 39) // 1: 2.664 39/17*36/31
ratio=(28, 43) // 2: 1.783 43/28*36/31
ratio=(33, 37) // 3: 1.302 37/33*36/31
ratio=(36, 31) // 4: 1.000 31/36*36/31
ratio=( , ) // 5:
ratio=( , ) // 6:

[FINAL_DRIVE]
bevel=(31, 36) // 1.161 Input shaft
ratio=(14, 43) // 3.071 Primary Dana/Ricardo M44-4

edit: If this uses the Aston Martin 5th, 6th. The toothcounts are:
ratio=(54, 37) // 5: 0.796 37/54*36/31
ratio=(61, 33) // 6: 0.628 33/61*36/31

Last edited: May 25, 2020