Physics The Physics discussion thread

[PhilS13]

@William Levesque @Avo77

Everything is fine on my side. I'm pretty sure the theory that a symmetric car should develop equal front and rear slip angles during steady state cornering is only valid at ridiculously low steer angles.

If I input the typical wheelbase, track, turn radius and accelerations that I get on the skidpad into this simple 4 wheel model here : http://www.gipsa-lab.grenoble-inp.fr/~moustapha.doumiati/ACC2010.pdf (equation 1 in particular), the model says the front lateral forces should be around 6.5 % higher than rear to reach equilibrium. The smaller the turn radius, the higher the steer angle, the bigger this number gets, more than 10% if I bring the front tyres to their optimum slip.

I believe this 10% is what @Avo77 was seeing in his slip angle data. Nothing inconsistent going on imo. I see no self realignment in the data at typical driving yaw angles.

 

[kondor999]

Wow, read the whole thread and realized the lesson is:

The Corvette owner stereotype (crotchety, argumentative, opinionated old man) is kinda correct after all.

I got an i8. Gets you laid way better anyway, which is the real purpose of a nice car btw

 

[Brownninja97]

I got an i8. Gets you laid way better anyway, which is the real purpose of a nice car btw

Well you will be surprised with how effective my 04 Corolla has been. Might be because im still young but between the rich guys with their Porsches and i8s and one sausage with a Hellcat at my uni dont seem to have much of a one up over me.

 

[kondor999]

Yeah, you're right of course. But I'm 50. I gotta use every advantage I can find to loudly announce I've got some money. Sad but effective. You'll see

 

[William Levesque]

Meanwhile all I pull in my Miata are weird looks of confusion... Might have something to do with it having series stickers, aero and not much interior trim left. And the strange bearded guy listening to weeb music that's driving it.

Maybe.

 

[Brownninja97]

Meanwhile all I pull in my Miata are weird looks of confusion... Might have something to do with it having series stickers, aero and not much interior trim left. And the strange bearded guy listening to weeb music that's driving it.

Maybe.

Oh i can agree with you there, no one minds if i listen to anything remotely non english in doors but if you do in a car wow like you have been indecent in public and its not just japanese stuff, same reaction to bollywood. Genuinely have no clue are people that afraid from what they dont understand.

 

[garyjpaterson]

Can I borrow the expert hands of a physics guru to test and try to solve an issue I'm having with one of my cars?

Specifically its my Anglia hot rod - which physics wise has nothing to do with the standard Anglia, and is supposed to be completely custom built.

I'm having problems with the inside front wheel coming off the ground during just about any corner. That in itself doesn't seem too bad, but its seems to allow the car to do some weird disconcerting 'rocking' back and forth diagonally across the inside rear/outside front.
It has the effect of making a perfectly smooth bit of tarmac feel quite bumpy!

I'd like to think I have a reasonable idea of how to set a car up (evidently not ), but nothing I've tried was successful. I've tried lowering the front ARB, stiffening the rear (its oversteer city as it it...), played with bumpstops/packers... Probably something obvious but its eluded me.

Let me know if you're willing to give it a go - I'll send a download link. Apart from those issues, its actually a lot of fun.

 

[William Levesque]

Suspension geometry is my guess. Can't be more precise tho.

 

[ALB123]

I'm working on a Lamborghini Diablo model right now and I have enabled Brake Temperatures. I know exactly what PERF_CURVE & COOL_SPEED_FACTOR do and how to set them properly, but I am not certain just what exactly COOL_TRANSFER and TORQUE_K are and how they should be interpreted (are there different ways to approach this for a street car vs. race car?, for instance)... If anyone has any knowledge they could share I would be very grateful.

[TEMPS_FRONT]
COOL_TRANSFER=0.XXX
TORQUE_K=0.XX
PERF_CURVE=XXX|XXX|XXX|XXX
COOL_SPEED_FACTOR=0.XXX

 

[Stereo]

Torque K should be a heating coefficient (proportional to the amount of torque being generated by that brake) so essentially you set it based on the size of the disc, then cool_transfer is how quickly it'll cool off again.

I don't know of guidelines to set them reasonably, not enough data.

 

[ALB123]

Torque K should be a heating coefficient (proportional to the amount of torque being generated by that brake) so essentially you set it based on the size of the disc, then cool_transfer is how quickly it'll cool off again.

I don't know of guidelines to set them reasonably, not enough data.

Hmmm...I was thinking that COOL_SPEED_FACTOR was the parameter to determine how quickly the wheels cool off. Maybe I'm wrong...Hopefully some one will come by and know exactly what every setting does...and then gives an explanation of how these parameters interact.

Thanks.

 

[PhilS13]

Hmmm...I was thinking that COOL_SPEED_FACTOR was the parameter to determine how quickly the wheels cool off. Maybe I'm wrong...Hopefully some one will come by and know exactly what every setting does...and then gives an explanation of how these parameters interact.

Thanks.

This is only after 5 minutes testing. Don't take it as absolute truth.

TORQUE_K changes how much of the braking torque turns into temperature. See it as the bigger the rotor and caliper mass the smaller this value needs to be.

COOL_TRANSFER is basic cooling. Again the bigger the mass the smaller this needs to be.

COOL_SPEED_FACTOR is cooling factor dependent on vehicle speed. Setting it to 0 will make the brakes cool at the same rate regardless of vehicle speed. A car that has dedicated ducts to direct ambient air to the rotors will need this number to be larger.

 

[mclarenf1papa]

This is only after 5 minutes testing. Don't take it as absolute truth.

TORQUE_K changes how much of the braking torque turns into temperature. See it as the bigger the rotor and caliper mass the smaller this value needs to be.

COOL_TRANSFER is basic cooling. Again the bigger the mass the smaller this needs to be.

COOL_SPEED_FACTOR is cooling factor dependent on vehicle speed. Setting it to 0 will make the brakes cool at the same rate regardless of vehicle speed. A car that has dedicated ducts to direct ambient air to the rotors will need this number to be larger.

This is correct (though the speed factor is based on the square of speed).

 

[ALB123]

This is only after 5 minutes testing. Don't take it as absolute truth.

TORQUE_K changes how much of the braking torque turns into temperature. See it as the bigger the rotor and caliper mass the smaller this value needs to be.

COOL_TRANSFER is basic cooling. Again the bigger the mass the smaller this needs to be.

COOL_SPEED_FACTOR is cooling factor dependent on vehicle speed. Setting it to 0 will make the brakes cool at the same rate regardless of vehicle speed. A car that has dedicated ducts to direct ambient air to the rotors will need this number to be larger.

Thank you, guys... I'm really trying to get a grasp on car modding and I've never done any of this stuff before. So, if you read some idiotic questions from me, please go easy.

I'm curious to hear some opinions on this while I've still got my attention on the brakes. I'm "practicing" with a Lamborghini Diablo. What type of advice could you offer regarding the over-heating of a 1990's supercar's brakes? What I mean, I was running test laps on Red Bull Ring - full throttle, but responsible on the brakes (no ABS). I wasn't babying them, but I wasn't pushing them to the edge every single braking zone. Here is my: PERF_CURVE=(|0=0.8|200=0.98|250=1.0|600=1.0|800=0.7|) -- I was displaying Brake Temps in Sidekick so I could keep an eye on them...

Every now and then I would throw in some really, really heavy braking to push those temps over the edge (600 degrees) to induce fade. A Diablo is a performance car - but it's not a race car. Is this a realistic range for optimal braking? 200-600 degrees? A Diablo couldn't run laps all day at Red Bull Ring under maximum braking in real life. When I push the car over the upper range, it takes a few corners with intentionally easier braking to get them cooled into good operating range again. What should I be aiming for as a goal when trying to reproduce real car brakes?

 

[mclarenf1papa]

Thank you, guys... I'm really trying to get a grasp on car modding and I've never done any of this stuff before. So, if you read some idiotic questions from me, please go easy.

I'm curious to hear some opinions on this while I've still got my attention on the brakes. I'm "practicing" with a Lamborghini Diablo. What type of advice could you offer regarding the over-heating of a 1990's supercar's brakes? What I mean, I was running test laps on Red Bull Ring - full throttle, but responsible on the brakes (no ABS). I wasn't babying them, but I wasn't pushing them to the edge every single braking zone. Here is my: PERF_CURVE=(|0=0.8|200=0.98|250=1.0|600=1.0|800=0.7|) -- I was displaying Brake Temps in Sidekick so I could keep an eye on them...

Every now and then I would throw in some really, really heavy braking to push those temps over the edge (600 degrees) to induce fade. A Diablo is a performance car - but it's not a race car. Is this a realistic range for optimal braking? 200-600 degrees? A Diablo couldn't run laps all day at Red Bull Ring under maximum braking in real life. When I push the car over the upper range, it takes a few corners with intentionally easier braking to get them cooled into good operating range again. What should I be aiming for as a goal when trying to reproduce real car brakes?

600 definitely seems too hot for a street compound. Steel brakes peak early and then lose a lot of efficiency at higher temps. Even some carbon compounds lose efficiency by 600c. @Stereo gathered some street brake data so he should be able to help.

 

[William Levesque]

600C would be track oriented compounds (Hawk DTC30 + kinda deal). Still streetable in some cases tho. But probably much higher than what a 90s supercar could handle from factory. Around 400C-450C is the highest I would assume, but Im just guessing here. At that point you've already lost some performance, and nearing the cliff.

 

[ALB123]

Excellent!! Thank you... The way that I was driving....Full throttle, but smart driving for a street car, I was pushing around 400-450 degrees braking for Turn 1. Of course, I can still fiddle with the other parameters too which will adjust how much heat is actually generated. I'm happy with the amount of Torque I've given the brakes, so it looks like I'm off to a good start. I intentionally pushed that upper limit higher than I thought would be realistic (600 degrees) just so I wouldn't experience brake loss and I could push corner after corner. But truthfully, I didn't know what should be the upper limit. So thank you, both of you, for your help.

 

[garyjpaterson]

Who remembers this thread?

Question regarding dampers (for Ferrari 550 btw);

Left if the front dampers, right is the rear. Above the X-axis is rebound, below is bump.

After working out the slopes and applying the motion ratios, I'm finding extremely low bump values and extremely high rebound - anyone seen this sort of behaviour before for road cars?

Now bear in mind, the 550 has 'adaptive' dampers, where they can self adjust between 'open' and 'closed' (seen on the above graphs) based on various sensors - and this is not something we can do in AC. So instead I've just taken the average between the two, to get the best middle ground.

But even so, my slow Bump values are around 20% critical, but rebound is about 150% - is this acceptable or expected? I don't have experience with this so can't say really. Rear rebound is slightly more reasonable, at 110% critical or something like that.

 

[garyjpaterson]

Who remembers this thread?

Question regarding dampers (for Ferrari 550 btw);

Left if the front dampers, right is the rear. Above the X-axis is rebound, below is bump.

After working out the slopes and applying the motion ratios, I'm finding extremely low bump values and extremely high rebound - anyone seen this sort of behaviour before for road cars?

Now bear in mind, the 550 has 'adaptive' dampers, where they can self adjust between 'open' and 'closed' (seen on the above graphs) based on various sensors - and this is not something we can do in AC. So instead I've just taken the average between the two, to get the best middle ground.

But even so, my slow Bump values are around 20% critical, but rebound is about 150% - is this acceptable or expected? I don't have experience with this so can't say really. Rear rebound is slightly more reasonable, at 110% critical or something like that.

Seems this should be okay, posted it elsewhere and heard that this sounds reasonable for a road car, and that the bump/rebound ratio on road cars can vary quite wildy.

Just got to refine bumpstops as thats the only area I don't have data on.

 

[ALB123]

I have awakened the dead!! Well, a dead thread, that is.

How can I edit drivetrain.ini to make shifting older cars - Jaguar XJ13, for example - more realistic. I'm using a DFGT wheel with no external shifter or clutch pedal. So, I'm getting 240ms upshifts and 260ms downshifts which isn't possible with cars like the XJ13, Ferrari 250GTO, etc... I watch videos of people using H-Pattern shifters with these cars and there is that nice pause before the engine comes roaring back to life. Heck, even a Lamborghini Countach should have that pause while the gears are working themselves out.

Drivetrain.ini:

[GEARBOX]
CHANGE_UP_TIME=240                ; change up time in milliseconds
CHANGE_DN_TIME=260                ; change down time in milliseconds
AUTO_CUTOFF_TIME=240            ; Auto cutoff time for upshifts in milliseconds, 0 to disable
SUPPORTS_SHIFTER=1                ; 1=Car supports shifter, 0=car supports only paddles
VALID_SHIFT_RPM_WINDOW=400        ; range window additional to the precise rev matching rpm that permits gear engage.
CONTROLS_WINDOW_GAIN=0.2        ; multiplayer for gas,brake,clutch pedals that permits gear engage on different rev matching rpm. the lower the more difficult.
INERTIA=0.0265                    ; gearbox inertia. default values to 0.02 if not set

Is it just a matter of changing the parameter values of CHANGE_UPTIME, CHANGE_DN_TIME and AUTO_CUTOFF_TIME?
Using the gorgeous, sexy XJ-13 from mantasisg with aphidgod's drivetrain settings, I've increased the values of all three to 600ms and it does a pretty good job simulating what I'm looking for. I'm just curious if I'm forgetting anything or if there is a better way of doing it? I lift my right-foot when performing a gear change as if I were shifting a manual transmission...I just wanted to hear what the experts thought of my silliness... Yay or Nay?