Sorry for lowly graphics, I had mostly everything off and low. Anyway there is a vehicle in picture flying at mid air, wheels are hanging down and should represent rebound settings that they have been set to, if I understand things correctly. For FRONT: For REAR: For added challenge I like to mix imperial and metric system units randomly. (yes, car is Firebird, but 1st gen Camaro is F-body and firebird shares Chassis with it, so it should be reasonable close in terms of wheel travel). Anyway, from my opinion, rear wheels should drop more than front wheels. Wheel rate is bit softer at front (real car is made that way) so in rFactor softer springrate wheel travels more than stiffer springrate wheel, which in my opinion causes this issue. It is not only limited to wheels hanging differently than what one excepts, weight transition is also affected, as front end has now more travel or rear less travel than should, that causes difference in handling. What I probably need to do is jump the car, check logs created by data aquisation plugin, adjust rebound value, jump check again and so on, but that will still leave bump travel, it is fairly difficult to tell bump travel when there are bump stops affecting movement, so I would need to make bumpstops act like steel and then I might be able to get movement range to be relatively ok, except bumpstop stiffness set to lower it might cause again more bump travel than hoped, which results randomly adjusting of bumpstops. Do I do this totally wrong or is this the method how to make wheel travel match reality? It is quite important bit that affects handling and somehow bit frustrating bit, but there might be better way and it would be nice to hear about it
What are your min and max ride heights? The last week I've been examining travels again, so there are at least two of us that are looking for some clarification. I've seen so many different comments on the travels, I'm not sure what to think anymore. The one that rings in my head is something about travels not being travels at all, but rather the location of the limit of vertical wheel motion. That would match what I've found. The only way I can get additional rebound is by softening the spring. No need to jump the car, just look at the first .05 seconds of your DAQ log. RF drops the body from a fully extended suspension position (as the pm is supposed to made) to a static loaded position. Playing devil's advocate a bit here: How can you limit the amount of rebound travel? How is it that springs do not dictate the amount of droop? *edit: Looking at what you are using for an equation, I think I see what is going on. I do this: I figure out the travel @ a certain Ride Height. So @ 76.2mm I know I should have 5.85mm of travel to the stop, and 44.11 mm of droop. So 5.85-76.2= BT (-0.0703486116) At my Mimimum RH (76.2) I have 44.11 mm left for rebound travel. As I raise ride height I use that. The max RH is I'll have something like 120.2. At that RH, I have .011 left for rebound. -0.011-0.1202=RT (-0.1203132363) How K! is handling travels. Looking at the analysis tab travel slider. It is taking the current ride height and simply adding it to the BumpTravel and ReboundTravel values. For the above example, since my suggested RH is 86.2mm I am getting an additional 10 mm to the 5.85 I already have. That should be up to 15.85. K! is showing 14. FreeRT also works the same way 86.2+-120.3...=-34.11... K! is showing -33. OR 44.11...-10mm=-34.11...
Ride height is not adjustable at all, spring rate is also fixed. For front, I think I had 178mm and for rear 180mm with that test, but those I need still refine as in rF ride height is to ref plane and ref plane is 0 position for suspension, level which car rests at. 5.5" is from frame to ground according to factory's specification. front suspension was over 9", oilpan is lower than ref plane should be and fuel tank is perhaps bit higher, but it is not quite easy to perfectly pinpoint exact level. With Volvo it was simple as I could take measurements, water scale from wheel center rearwards and then distance from bottom of car floor to ground at front edge of front door location. Then in 3ds max need to set body to have 0 at bottom of body, that very same measurement point, not edge of sill etc. After that in hdv set rideheight to distance to car bottom to ground + difference to wheel center. Sounds much more complicated than what it is. But real measured ground clearance, in my opinion, can't be used unless that matches to 0 level/ref plane, it does not always match. Undertray is then, in my opinion that sets aerodynamic car bottom, feelers and undertray together make physical car bottom. If my understanding is correct, you could set your bump/rebound amount partly with adjusting your ride height, so that car bottom is in correct position related to wheel center level, then set how much movement there is and it, at least in my mind right now, should result more or less rebound and bump. Wheel travel adjustments do limit movement of wheels, but only if one sets them enough low with enough soft springs. However if I set rebound to be 5cm and I change spring rate, movement amount changes so only way that I can know that I get 5cm is do lot of desting and adjusting after testing, then test again, values in rebound and bump travel seem to be imaginary units or something like motion range, not exact movement amount, but fraction of all movement that current wheel rate can offer. Well, kind of as it still has meaning in meters too, because you can't set rebound lower than ride height. I must say it is very confusing how it currently works and there is very big chance that I have not been able to figure it out correctly. But for street cars this method has worked quite ok. For ride height to affect amount of rebound, I would do it via upgrades. There is correctedInnerSuspHeight or similar, which would keep inner side of suspension at place and maybe in theory would also affect to rebound/bump amount, but that is bit cryptic for me and K! sets that to -1 always, imo. I don't know if that inner susp height is height from ref plane and don't know if measured from middle point between lower and upper arm, lower arm, center of spindle or what location, so setting it so that you would indeed get inner suspension to correct height is bit trial and error. For some reason K!'s analyze section says that it can't find contact patch and something about suspension geometry being so weird that it can't find roll centers, but it is suspension that K! made for me, so I haven't had luck with analyze section again, from there I could see that setting inner suspension height to 0 was way too high, 0.1 was lower and more right, but -0.1 was even higher and that is oppisite of what my mind tells me how it should be. But as I don't have adjustable suspensions, I leave setting to -1 and use upgrades, .pm and .hdv to make different setups. There are few test I have meant to be doing, but haven't had too much of time for them. One is to figure out if bumpstops affect rebound also, this could be done by setting bumpstops to 0 or close and test jump, then very stiff and test jump, after that comparing logs. Oh and I do jump because it's fun part of the testing and as I made ramp which is not upsetting the car, also when car comes down I can see bumpstops and can even tune suspension damage etc. Luckily I don't need to pay all the cars I wreck at testing What I would like to hear is how ISI gets values for bump and rebound for example when they measure car wheel moving 3.5" up and 3.5" down, making 7" movement range, maybe then this would be more clear to all of us, well at least two of us that have been talkative about subject
So I just did a test, and lost my post because rf locked everything up. I'll warn you now about going too high with your bump rubber rates. I previously had bump rubber set to actual values the test that locked things up was 1000x those values. My new test was conducted with 10x those values. One test was done @ minimum RH, the other @ maximum. What I found is that indeed you can limit body sag with Ride Height. There is a nice flat spot where I run out of travel. For certain at the lowest RH, I was running on the stop, and the stop allows for ~ 5mm of compression. The Formula I was using was the one I listed previously: "Freebumptravel@minRH"-minRH=BumpTravel, -1*("FreeReboundTravel@minRH"+minRH)=ReboundTravel Using your numbers: Front If your RH was 0.178 and your FreeBumpTravel was 0.096774 I get BumpTravel=-0.081256 If your RH was 0.178 and your FreeReboundTravel was 0.108966 I get ReboundTravel=-0.286966 Rear If your RH was 0.180 and your FreeBumpTravel was 2.32"(0.058928) I get BumpTravel=-0.121072 If your RH was 0.180 and your FreeReboundTravel was 5.43"(0.137922) I get ReboundTravel=-0.317922 CorrectedInnerSuspensionHeight, I've set this to WheelRadius-RideHeight. Both my front and rear RH are different, and my Radii are also different, the result is that the rear body side suspension points are 1.8mm too high. CISH will not affect travel, only the geometry. Since you don't have adjustment I don't see how using -1 would do anything detrimental.
Sometimes I might overcomplicate things, most of the time I feel confused, anyway this is ISI declaration of ref plane: View attachment 3234 One of cars under work: View attachment 3235 I can set refplane there to be distance from bottom of body to ground, or maybe from frame to ground, I guess, but with different setting and adjusting model height, feelers and undertray, I would imagine you could get more rebound, but it gets bit confusing again. There was more to it however, but now my memory is blank again what it was. For rebound and bump travel, I don't think that one can calculate correct value to use, because amount varies with stiffness of springs, so one just need to do lot of testing.
I hear that! If I find a new bit of info, I end up checking out my previous calculations, the number of times my head has spun is infinite. I think this has gotten slightly OT so I will attempt to steer it back. You started off the thread with "rebound travel is wrong", and a pic of the car in air. You feel that rear wheels should drop more than the front. You are correct to think the wheel rate is controlling the rebound travel. The question I have is why do you think that is an issue? When you unload a damper it extends fully until the springs are at their free length (if not precompressed). When you load a spring, it compresses according to the spring rate and the load. This is what determines rebound travel. Thus the way rfactor is built is correct. Changing the ReboundTravel= value in the HDV is only going to change how the available travel changes when the ride height changes, it is still up to the spring to determine where the actual split is. I'll put it this way: In order to get my car to a higher ride height, I need to keep the damper extended. Raising the ride height sacrifices available rebound travel by dedicating it to bump travel. At my max ride height the damper is required to be fully extended. The only place for the direction of travel is in compression, and the test I did shows very little rebound travel, if any. It is much easier to see what is going on with stiffer bump rubber. At this point my travels are correct to realworld numbers. I just need to adjust the stiffness of bump rubber to allow for 90% compressibility. Is currently too stiff (will compress maybe 6 mm max, should be more like 40mm). You are correct that the only way to get that to a target number is through testing. <time passes> I just tried to adjust my bump rubber to soften things, and what I found was interesting: It seems to affect both the bump and rebound position. Where I was getting very little rebound at max ride height before, I now get somewhat normal travel.
So bumpstops are like how I secretly have been thinking, at both ends and only method limiting suspension movement so that wheel does not go around and around and around after enough large impact or when jumping. I think that when I tune suspension via logs I get travel right visually too, but I would imagine that it should be possible via direct setting instead of finding 3rd party tools and software which is not exactly meant to be used by simracers, but buyers of Motec hardware. edit: What I found from logs is rather interesting, despite movement being visually clearly more at front, logs claim that rear has had more rebound travel, this is calculated from suspension positions. Crappy table attempt: ____________FL____FR____RL____RR rebound_____119.8_119.8_128.4__128.4 bump_______173___173___102___102 total________292.8_292.8_230.4__230.4 total in specs_205.7_205.7_199.4_199.4 Of course bump is tested with around 11g's acceleration according to logs, so it is rare when it reaches that amount. Some reason my logs don't start from beginning, but from random point, sometimes I have already 80kph speed when log starts and I hardly ever start driving immediately, engine has to warm a bit first.... With this car front has tiny bit more travel than rear as how it is designed, but as I did set my wheel travel it is clearly wrong, rear should have around 59mm of bump travel, front around 97mm bump travel, rest should be rebound. I think it is more of understanding way rfactor does things than rfactor doing things wrong, title rebound being wrong refers to this car, not rfactor. Anyway, I think that I will end up to similar issues you are having soon as I start to tweak rebound and bump to read correctly in logs. edit2: I end up with 128mm rebound travel no matter what, bump was easy to get right, but rebound should be 140mm, not 128mm, whatever that 12mm is from 199mm total and 140mm rebound, 5-10% and is it's effect some 2% then? Maybe bit softer bump stop would help to get that maybe 2% back, need to do front also, maybe that is as simple. I just calculated difference between what I see log and what bump should be and calculated that to bumptravel, then it was right. --- adjusting --- I got front to work even closer to specs. Now of course problem is that in typical usage only part of movement range is utilized, even at racing conditions and bumpy tracks, spinning and doing burnouts as quite some part is restricted by bump stops, perhaps it would be reasonable to accept too large movement range at extreme conditions and adjust movement to match what movement you get at typical racing conditions, I doubt that both can be achieved.
It might depend what DAQ pluging you are using, there are two, three if you count the one for rf2. All of my comments are related to rf1, but I see no difference in rf2. The plugin I'm using has a switch in the ini, something about starting logging after movement, I have that off.
I did not see such option, only options when to start building logs that I assume are stored in memory if not set to build immediately, but all my options were set so that logging starts immediately, automatically and logs are being build while still in car. It might have to do with rf2 being active or inactive, but I haven't had time for finding the reason for that misbehavior. I ended up soften a bit of bumpstops, then I added bit to movement range, it is rare that when driving around the track I get full motion range utilized, even when doing high speed spins and running over rumble strips and sand areas, but when crashing so that car flies trough the air, I get some more than movement range. Maybe I add bit more to range, but front/rear ratio should remain unchanged. Engineers of Firebird probably made rear bump travel so small amount to prevent car dragging it's rear when launching hard, I just haven't found anything from bumpstops so I started with hammer values from rF1, however can't say if those are even close, however that is another problem. I would like to choose at which point my bumpstops start to affect motion and how small they can become, but of course clever ones can calculate that with some clever formula using bumpstop values and total mass, me can't, me stoopid
rf2 was built differently than the plugin I'm using for rf. Heh, great minds think alike? Last night I got to thinking about what the force required to bottom out a bump stop is. I've got 260N/mm springs, they compress until they hit the bump stop at travel of 39.870268558 mm. From mfg data I see the bump stop can travel 50 mm. This is less than the point at which my spring binds, so I'll use that as the limit. In order to compress to the stop I need a force of 10,366.274N. From that point on the distance both spring and bump stop compress are equal. The Force required to compress them both is simply F(eq)=k1x+k2x. I would need an additional force of 13,000 N just to compress the spring the addtional 50mm. To compress the bump stop I need to use BumpStopSpring and BumpStopRisingSpring. The force required to deflect the Bump Stop is: DeflectionBS*(BumpStopSpring+((DeflectionBS-.001)*BumpStopRisingSpring)). DeflectionsBS units are in m. At max I was able to only get about 5.5mm out of my bumpstop. The this meant about 11,800N to the spring, and a force to the bump rubber of about 1046N, a total of 12,847 N. Using the same BSS, and spring values, I would need a total force of ~64,500N to get to a deflection of 50mm of Bump Stop. This is unlikely to happen. The remaining piece is the damper aspect of the bump rubber working to remove energy from the system.
My springs have this kind of characteristics: Solid 77.35mm, force 6528.27N Length 214.8mm installed to strut. front bump ~76.2mm front rebound ~60mm There may or may not be bump stops, I would guess that there are, but as I got solid length and force from spring design software, measured bump and rebound by experiments, Force I really can't know for certain, maybe somewhere in range of 13 000N was used in experiment, or bit more. If there are bumpstops, they are around 50mm very soft rubber and they compress to ~30mm with two fingers and start really resists after that, I need still measure top of shock housing and lower spring cup distance to know when bump stop should be active. Decimals have only for amusement factor, they came when I converted inches to mm and I did measure in inches only as ruler had bigger numbers on that side. Measuring was done with always so reliable 'bricks to hold ruler, flower pot to act camera stand, record video while pushing corner of car down several times' method. Rebound was measured with help of jack, neutral level is somewhat uncertain because lack of perfectly level ground, but range is there, finding neutral position just needs some car moving, measurements, averaging etc. This car I know best so it is better to use it for finding unknowns, car in 1st pic is inaccessible, but manufacturer specs reveal some of the things, just not all needed.
So I've been examining the FR3.5. The geometry of the suspension.pm matches that of Renault's documents. The origin of the suspension.pm is at the height of the reference plane. I don't understand why you would offset inner points when they are correct.
I believe posting links to copyrighted data is against forum policy, with the vehicle being licensed I certainly wouldn't want to put ISI in a predicament. I will drop a few breadcrumbs: 10 page preview links to a couple sections were already posted, the entire thing is available, all you need to do is figure out how to link back to the original documents.
