Bumptravel Reboundtravel and Bumpstops

lordpantsington

Been seeing all sort of different comments on this. As I understand it sets the range of motion of the suspension.

Currently I am using these formulas:
BumpTravel= // suspension travel upwards ( = Free bump travel - minimum static ride height)
ReboundTravel= // suspension travel down ( = Free rebound travel - maximum static ride height)

Are the travel distances the wheel motion up and down? Or are they compression/rebound along the damper (which in my case is angled 19 degrees from vertical)?

Are the travel distances to the hardstops internal to the damper?
Or are they to Bump rubber?

Seems as though the BumpStopSpring and BumpStopDamper values are astronomical.
If you look at page 39, those do not seem to match the values in the HDV.

http://www.zf.com/na/content/media/all/all_corporate/products_services/motorsports/catalogs/36043.pdf

TIA
lp
 
Last edited:
I'm following W_Bradshaw interpretation - those are the limits for wheel travel, where bump limit is where you hit a bump stop (so no hard stop in rF, unfortunatelly :( ), so you have to recalculate them from your damper stroke and angles.

Remember, all stiffness values are expressed in N/m so they are 1000 times more than you would expect from real world data (typically expressed in N/mm). You see this with spring rates (a 200.0N/mm spring will be a value of 200000.0 in rFactor) and with bump stops.

btw. Page 39 of what document? :)
 
Hah, I use one of those in my mod :)
But as you can see, there are so many bump stops with different stifness, you have to pick the right one for particular car or racing series. It can't be the first one you've found.
 
Is this from the missing section as posted by Bradshaw?

[FRONTLEFT]
BumpTravel=-0.005 // Travel to bumpstop with zero packers and zero ride height (5mm compression)
ReboundTravel=-0.057 // Prevents rebound travel (for example, when upside-down), 45mm max front ride height plus 12mm leeway

Defines the limits of distance over which the wheel can travel vertically, relative to the reference plane.

BumpTravel is the extreme upward extent to which the suspension can move, relative to the "reference plane", which is usually the bottom of the body tub, before the suspension runs out of travel.

ReboundTravel is the extreme downward extent over which the suspension can move, relative to the tub, before it runs out of travel.

Do not misinterpret these values as the distance the suspension can move up or down. They define the location of a limit, not a distance.

Suspension travel up (a positive number) is RideHeight plus BumpTravel (Algebraic sum). For this car, Suspension travel up is 0.038 + (-0.005), which is 0.033

Suspension travel down (a negative number) is RideHeight plus ReboundTravel (Algebraic sum). For this car, Suspension travel down is 0.038 + (-0.057), which is -0.019

Total suspension travel is the algebraic sum of BumpTravel minus ReboundTravel. That total is 0.033 - (-0.019) or 0.520 (metres).

Packers will reduce the available bump travel before running out of travel.

This car has the BumpTravel value at -0.005, which means the suspension limit of travel is when the body tub is 5mm off the ground.

If the suspension runs out of travel during a corner the car behaviour becomes erratic and it can spin out of control. Bump stops may be leaned on during cornering, but their values have to very carefully matched to the spring values so there is a smooth transition.

In F1C, Other symptoms of inadequate suspension travel or ride height values outside the available travel range occurred. They were rapid side-to-side head movement by the driver and a reduction in frames per second displayed. This was an ISI engine issue that can be quite severe to the point of making a car unusable in groups. I do not know if it happens in rF.

BumpStopSpring=150000.0 // Initial spring rate of bumpstop
BumpStopRisingSpring=7.00e6 // Rising spring rate of bumpstop (multiplied by deflection squared)
BumpStopDamper=2400.0 // Initial damping rate of bumpstop
BumpStopRisingDamper=7.00e5 // Rising damper rate of bumpstop (multiplied by deflection squared)
Spring and damper properties of the bumpstop
Click to expand...
 
LesiU said:
Hah, I use one of those in my mod :)
But as you can see, there are so many bump stops with different stifness, you have to pick the right one for particular car or racing series. It can't be the first one you've found.
Click to expand...

Good thing didn't randomly pick a bump rubber, as I surely would have gotten it wrong. I measured, based on a photo, set to a scale, based on the full extension length of a damper, and then compared shape to mfg data.

But, I was speaking generally about them and if you look at page 39 you basically get this for the first one:

Deflection vs Force
(0,0)
(5,200)
(10,600)
(15,2000)
(20,5000)

Changing that into Rate [N/mm]:
0,40,60,134,250

[N/m]
0,40000,60000,134000,250000

As I see it: as the amount of deflection approaches the full height of the bump rubber, the force approaches infinity.

Values from the ZR:
BumpStopSpring=160000.0 // initial spring rate of bumpstop
BumpStopRisingSpring=1.20e7 // rising spring rate of same (multiplied by deflection squared)
BumpStopDamper=2000.0 // initial damping rate of bumpstop
BumpStopRisingDamper=9.00e5 // rising damper rate of same (multiplied by deflection squared)
Click to expand...

Ok, looking at BSS value: 160N/mm, increasing rate of 12N/mm/mm.

So If I were to place that rubber in the ZR, why wouldn't the initial spring rate be equal to zero?

If the Bump travel is set to limit the motion at bump rubber, how does the rubber ever come into play?
 
I will comment on the values later.

lordpantsington said:
If the Bump travel is set to limit the motion at bump rubber, how does the rubber ever come into play?
Click to expand...

Bump stop is that elastic limit of damper stroke. If you put very soft bump stop, then the damper's stroke will be long as hell. If you put bump stop stiff like a stone, then the stroke will end just after hitting it. So, in other words - bump stop stiffness it the ONLY limiting factor of total damper's stroke. With BumpTravel, you only show to rFactor, at what point it hits the bump stop.
 
LesiU said:
I will comment on the values later.



Bump stop is that elastic limit of damper stroke. If you put very soft bump stop, then the damper's stroke will be long as hell. If you put bump stop stiff like a stone, then the stroke will end just after hitting it. So, in other words - bump stop stiffness it the ONLY limiting factor of total damper's stroke. With BumpTravel, you only show to rFactor, at what point it hits the bump stop.
Click to expand...

I think i understand now, the BumpTravel=value is not the absolute limit of travel. There is more going on after that point, as dictated by the rubbers themselves.

This is good news for someone that wanted to create bump rubbers as upgrades.
 
Holy ****!!! An informative thread. Good stuff!!


bumpstops limit the amount of spring travel while still keeping the spring rate. The heavier the bump stop the less travel in that spring. Here's an excellant setup guide. It's for netkar which models chassis flex but still valid for alot of sims. Gives you an idea of how complex the physics are in netkar. ah taking too long to upload.
 
Last edited:
lordpantsington said:
Good thing didn't randomly pick a bump rubber, as I surely would have gotten it wrong. I measured, based on a photo, set to a scale, based on the full extension length of a damper, and then compared shape to mfg data.

But, I was speaking generally about them and if you look at page 39 you basically get this for the first one:

Deflection vs Force
(0,0)
(5,200)
(10,600)
(15,2000)
(20,5000)

Changing that into Rate [N/mm]:
0,40,60,134,250

[N/m]
0,40000,60000,134000,250000

As I see it: as the amount of deflection approaches the full height of the bump rubber, the force approaches infinity.

Values from the ZR:


Ok, looking at BSS value: 160N/mm, increasing rate of 12N/mm/mm.

So If I were to place that rubber in the ZR, why wouldn't the initial spring rate be equal to zero?
Click to expand...

EDIT: If we look at the formula for calculating bump stop stiffness, it is clear you will always have 0 for 0 deflection.

Personally, I don't like the concept of having bump stop stiffnes as a limiting factor (suspension geometry at some point probably will also limit wheel travel, but at much moree extreme wheel bump, than what we discuss here). I really hope, in rF 2 we will be able to set a proper damper stroke hard points and within them we will be using bump stops. From a dev point of view, having hard stops might be considered as a boundary condition so it has to be treated in a special way... I don't know. I'm only guessing.

So, in ZR we have initial spring rate of 1.6e5 N/m and second value 1.2e7 N/m (which will be multiplied by deflection^2)
For 1mm (which is 0.001m) we have (160000*0.001) + [12000000*(0.001^2)] = 160+12=172N/mm
For 2mm => 320+48=368N/mm
... for 5mm of bump you get 800+300=1100N/mm... so it's probably very stiff and very small bump stop... or just something that exists only in ISI's imagination. Who knows! ;-)
 
Last edited:
I'll start with a note that these things usually confuse the heck out of me (resistance, rate, ...).

BSS and BSRS are N/m.

For 1.6e+05 and 1.2e+07, if you multiply BSS by deflection and BSRS by deflection squared, you get
1mm => 1.6e5 x 0.001 + 1.2e7 x 0.001^2 => 172
2mm => 1.6e5 x 0.002 + 1.2e7 x 0.002^2 => 368
10mm => 1.6e5 x 0.01 + 1.2e7 x 0.01^2 => 2800
Above, we seem to be assuming that the result is N/mm, making 1mm deflection at least half as 'stiff' as usual maximum spring settings.

At 0mm deflection the stiffness will be 0. At 1mm it's 172000 N/m, 2mm is 368000 N/m.

So what we might refer to as the 'initial resistance' is 0, since deflection is 0 when we first make contact with the rubber.

What puzzles me about these figures is that I've seen quite high figures for BSS and BSRS in F1 mods (let's take CTDP06 as an example, with 2.1e+05 and 7e+06), giving:
1mm => 217
2mm => 448
5mm => 1225
10mm => 2800
15mm => 4725

... which, to me, says that you don't want 10 or 15mm deflection while cornering (it's actually more sensitive to low deflection values compared to the ZR); yet in practice it's often beneficial to lower the ride height by ~10-12mm on a setup that is just about hitting the rubbers (initial ride height - suspension position - packers + 'BumpTravel') during cornering, even when you adjust for aero effects (so, aim to avoid running on rubbers until you're doing ~120km/h, but rely on downforce from there even if it means hitting the rubbers).

In trying various things (and because I misinterpreted the BSS as a constant, rather than scaled by deflection) I tried increasing BSRS to ~2e+08, which seemed to better penalise hitting the rubbers as I'd hoped. Unfortunately it made AI, and my unoccupied car when joining the session, bounce around with better-than-elastic collisions.. usually ending up bouncing on the far side of the track and taking 10+ seconds to land each time (often with a 'reset' halfway through).

Anyway, I'll admit to being over my head in this area. Just pointing out that 0 deflection yields 0 'stiffness' from the rubber (albeit quickly increasing), and that rubbers don't seem to penalise you as much as you'd expect from something designed to stop you running out of suspension movement completely, rather than a handling-adjustment-tool :)
 
Lazza said:
BSS and BSRS are N/m.
Click to expand...

If BSRS is the rate that BSS is increasing, the units would be N/m per m (or N/m^2). As in: the BSS increases 12 N/mm every mm.

How much does it increase in total for deflection of 10 mm?

120 N/mm

In order to get the correct units for force, you need to multiply out deflection squared which checks nicely with the comment.
 
Last edited:
I edited my previous post, about BSS value used by ISI. As Lazza pointed, there will always be 0 for 0 deflection.
 
Thank you lordpantsington, that's exactly the sort of oversight I knew I'd make ;)
 
Been looking into "Micro-cellular progressive jounce damper" again/further. :D :D :D
It is very easy to find FvD diagrams for bump stops. I've seen nothing to describe what damping is involved.

When trying to synthesize a F(D) curve, BBS(D)+BSRS(D^2) doesn't cut it. According to all of the FvD graphs I've seen the change in the slope of FvD is not linear. Therefore, force with respect to distance should not be a quadratic equation.

http://www.eibach.com/cgi-bin/start.cgi/eibach/applications.html?APPLIVAR=bumpstop For a bunch of examples.

Seems as though an actual curve starts linear then takes off. For the synthetic either you have too much build early, or not enough nearing the full compression. Playing around with a spreadsheet, the curve that best fits is exponential F=(b+a^D).

According to a post here:
http://www.eng-tips.com/viewthread.cfm?qid=100307
The equation is cubic.

This makes me wonder if our understanding of how stops are implemented is flawed, or is it the implementation that is flawed?

Looking @ rf2 for some clarity will find you none. The bumpstop on the FR35 have completely linear spring rates! Numbers for other vehicles are still pretty astronomical.
 
Last edited:
Lazza said:
In trying various things (and because I misinterpreted the BSS as a constant, rather than scaled by deflection) I tried increasing BSRS to ~2e+08, which seemed to better penalise hitting the rubbers as I'd hoped. Unfortunately it made AI, and my unoccupied car when joining the session, bounce around with better-than-elastic collisions.. usually ending up bouncing on the far side of the track and taking 10+ seconds to land each time (often with a 'reset' halfway through).
Click to expand...
[N][m][N/m]3.54E+0071.46E+006[N/m]
Force
Displacement
Rate
D^2
D
Total
000000
1420.0052840088572808165
348.013480035401456018100
5900.01539333.379652184029805
8810.0244050141602912043280
12380.02549520221253640058525
18190.0360633.3318604368075540
26730.03576371.429433655096094325
46140.0401153505664058240114880

I just experienced this bouncing with my front bumprubbers. The real-world values for the stop are in the first 3 columns. The BSRS, and BSS are listed in the first row, column 4,5. This is as close to the real rates as I could get, and don't understand why it causes bouncing.
 
I was confused a long time ago too, made the rate error. Its really simple, its not rate but FORCE. So with only a progressive bump rate of 2.88e6 you get 4608N force at 0.04m displacement. (force = deflection^2 * progressivebumprate)

Your bump rubbers are hardened steel :)

As with my experience with real data, rFactors bump rubbers tend to not be progressive enough. Instead of ^2, ^3 or ^4 would work better, but this might cause too extreme forces when jumping / bouncing, possibly causing instabilities..
 
And they felt like it!
Thanks.
That is great info.
 
Last edited:
You must log in or register to reply here.

Back
Top