Inside the Black Box

Sim Tools allows us to designate signals from the "game" for SWAY, SURGE, PITCH, ROLL, HEAVE, etc. to certain axes resulting in movement of a motor or actuator.

My question is does it add/subtract these signals if designated to the same motor/actuator?

For example, if you assign SWAY and ROLL to the same motor--how does it handle the information? Add the two together? Same for SURGE and PITCH.

Then for Traction Loss--is that SWAY and YAW or just YAW?

Can you have SWAY signals sent to more than one motor/actuator?

It started occurring to me that in racing, ROLL might be a small movement and SWAY much larger in proportion. But in a flight simulator, ROLL might be large with no SWAY. But in both cases the seat would rotate or tilt. Same situation with SURGE and PITCH.

And, with SWAY in a race car, is that the sliding movement sideways or the G Force in a turn? If a car had a stiff suspension (no ROLL) and big tires (no SLIDE) but BIG G Forces (LIKE F1) would the software the difference between a fast turn with high G forces and a slower turn with no G forces?

I started wondering if one might design something that allowed motion cues to be different for ROLL and SWAY.

This becomes of interest when adding the concept of a G Seat to the platform.

Comments/Answers welcomed!

Wayne

PS. I have read the Sim Tools Documentation below. I can't understand how Roll is movement and Sway is acceleration are translated to movement of the motor especially since for the most part Roll would be caused by lateral acceleration in a racing sim.

Nomenclature of movements in SimTools
Pitch is the tilt of the car forwards or backwards in [°]
Roll is how much the car is dipped to the left or right in [°]
Yaw is the heading of the car (north, east, south, west) in [°]
Surge means the acceleration of the car in longitudinal direction [g]
Sway means the acceleration of the car in lateral direction [g]
Heave means the acceleration up and down [g]

Pitch, roll, and yaw are angles and surge, sway, and heave, are g forces. Roll and sway cancel out each other’s motion. Pitch and surge are added or merged together . Yaw and heave have no correlation.

So if roll and sway cancel each other, then in a right turn the seat tilts to left because of roll to the left and the g force of sway trys to tilt the seat the opposite to the right????

My intuition which must be wrong would say that in a right turn sway and roll should tilt seat in same, not opposite direction??? Or is this the famous centrifugal vs centripetal force concept where the real force making the car turn right is to the right???

Look at it this way. Why do Nascar's oval tracks usually have high banking turns? Because as the car is turning left and throwing the driver to the right, the track is highly banked at an angle to the left to counter this force. This makes for lower sway g forces applied to the driver as well as for more down force on the car keeping it from flying off the track at higher speeds.

Btw, in the scenario above, when you drive on a highly banked turn at just the right speed to equally counter the angular force, your rig should be level in its sway/roll axis.

http://www.buildingspeed.org/blog/2013/09/turning-g-forces-and-banked-tracks/?cb=08025246823672205

Please see above link. I don't know yet how the software works but I do have experience driving Nascar in real life. At my first Winston Cup race at Dover in 1987 the g force in the turns was enormous. The Sway acceleration force was over 2G. The car leaned over and springs on the right were bigger than on
left. The force needed to keep my head upright was so large that the strap connected to my shoulder cut off the circulation and I drove a number of laps with one arm.

But technically the centripedal force and acceleration is to the inside.

So if we are leaning the seat to the outside of the turn to simulate the forces I still seem to think the actuator should lean the seat to the outside. If that means positive roll and negative sway..then I get it. If not, then I am still in the dark.

THANKS FOR YOUR REPLY.

http://www.nascarhall.com/media/news/Scavenging-for-NASCAR-Goodies

Above is a link from Nascar indicating the neck strap I mention was first used by Richard Petty to counter the force pushing his head to the outside of the turn. Banking helps but not enough to counteract the higher speed. Banking that would eliminate sway would have to be much steeper than what tracks have. Cars still get loose and tight in banked turns.

Banking does reduce the sideways sway of the car...it does not eliminate it. The driver uses the banking to increase speed until the limits of the tires are met. My guess is the acceleration sway is pretty much a function of tires on a flat track or an oval..assuming the driver is talented enough to use all of the tires' capability.

My suggestion would be to try out all forces and movements in real life on the rig. As far I know and based on my personally experience and preference I would explain mot as technical as you would whish but perhaps it helps you to fiddle out some details:

Situation: I am driving a car at an even speed over a longer bump or curb (right side of the car) in a right corner. The corner is totally flat, no hill climbing.

I am using only "Roll" in Simtools: - in game the car is lifting on the right side and the seat moves to the left. Stopping the car on the bump means the seat stays in the new left position.
I am using only "Sway" in Simtools: - in game the car is lifting on the right side, and the seat is moving as long as I am driving to the left side. Stopping the car on the bump means the seat gos back to the basic position.

-> The two forces are correlating to each other but they do not annul each other. Sway do the job from the new position of the seat done by Roll.
-> Example: in the right corner I have to avoid an obstruction (moving to the left) but I am staying on the curb. The seat stay in left position done by Roll but Sway moves the seat to the right.

I am using only "Heave" in Simtools: in game the car is lifting on the right side, and the seat is moving as long I am driving accordingly to the surface condition. Stopping the car on the bump means the seat goes into basic position.

-> Heave and Roll COULD correlate to each other, so it is important to tailor Heave to the surface (textures, fast moves) before you will set Roll (slower moves).

I am using only "Surge" in Simtools: - in game the car is lifting on the right side and the seat does nothing. Breaking the car on the bump means the seat moves forward from the basic position and after stopping seat goes backward to the basic position.

I am using only "Pitch" in Simtools: - in game the car is lifting on the right side and the seat does nothing. Stopping the car on the bump means the seat does nothing. It will stay always in the basic position.
-> The corner is flat and or no hill climbing, so the car is always in a horizontal position.

>Surge and Pitch COULD correlate to each other, so it is important to tailor Surge to the accerleration (fast moves) before you will set the Pitch (slower moves).

Now i hope I did not lost in space...i am doing a break and come back later to see, if I am all in right ...

PS: Surge and sway can be used in the opposite direction. Very "special" immersion guaranteed! (but requiring getting used to do so). Hint: Using traction loss the opposite sway would neutralize it.

Best do each individual setting of the force to get a decent feeling
Yes Roll can counteract sway, but as @BlazinH says, if you run around a heavily banked track, the sim will lean with the bank angle of the track.
Its a balance you need to find once you build your ride.

Thanks. Your answer is VERY helpful.

This is what is fascinating about the science of simulation. We are trying to create the sensation of force when no force is present...essentially fooling the brain. In a left turn pushing on your right hip to the left would create sensation of the turn in a g seat yet leaning the seat to the right creates a sense of turning by making your neck, arm, stomach and leg muscles try to bring you back to the left.

Miltary training for fighter pilots have been studying this for years.

There is also studies regarding roller coasters.. Roller coasters typically roll into the turn. I find myself leaning into the turn while racing...there are those that believe that simulation motion cues should lean or sway the seat into the turn like you suggest trying.

I think the mind adapts.

I think I get it...BlazinH and bsft coment that on banked oval roll will offset dynamic body roll and sway...as pit mentions with his stopping on curb example. The banking create roll to left while force leans seat to right.

In real life I will say that there is considerable sway felt by driver so IMHO i would adjust so seat leans right on steep banked oval for more realism. But as bsft indicates..adjust to your liking.

When a car is motionless on a high banked left turn your body is pulled to the left. When you are moving at the speed that equally counters the force pulling you to the left your body feels level. As your speed increases past that point your body will feel an increasing pull to the right.

My English is not the very best, can you explain it for me a little easier? Thanks

Let me try by expanding using more words....(based on my understanding of your very nice previous explanation that was very enlightening.)

BlazinH says same thing you did essentially---when not moving....car stopped... roll determines whether the car is level or tilted. In the case of your example, with car stopped and right tire on the inside bump strip, car leans left. In the case of the highly banked turn (like Daytona with 31 degree banking) then the car when not moving is sitting leaning to the left 31 degrees. As speed increases angular acceleration causes sway force and car leans less to the left--at some point being level---and when at maximum speed, the acceleration sway would lean seat to the right. The amount the seat moves would be set when you adjust the Sim. So at O Kph on a banked oval==car leans left 31 degrees. At 160 Kph, the car would be level with sway movement equaling roll. The lean left sensation is equalized by the centrifugal acceleration force. At 280 Kph, sway force would have exceeded roll and driver would feel a large centrifugal force--meaning seat would lean right in most Sim's to simulate that. I can tell you from personal experience, the only time you feel that car is leaning because of the banking is after you exit the curve and are going down the straight that is banked 18 degrees.

If you drove around Daytona in street car at 100 mph (160 Kph) you would never sense any centrifugal force. You would feel neutral in the corners---the feel from the banking would be equal to the centrifugal sway force. But at 175 mph (280 Kph), the centrifugal force would have increased by 3 times while the banking force staying the same, so you would get the same sensation as when turning fast on a flat track--in fact, probably more side force on the banked oval.

The sway force or acceleration varies by the square of the speed and inverse of the radius. 75% more speed creates 3 times the force. (1.75 x 1.75 = 3.06) The reason the car does not slide UP the hill banking in the curve is the holding force caused by the downhill portion of the weight force of gravity because of the banking, plus the side force produced by the tires. In a winged sprint car, your could add the side force from the roof mounted wing. When and if the driver exceeds this total force by going too fast, then the car slides up the track and hits the wall.

By the way, your English is pretty good. My German, Italian and French are non-existent. I could have a simple conversation with a 2 year old Brazilian child--that the limit for me.

@stroutmail thank you very much for the very detailed explanation and in deed my English is better than a 2 year old child speaking English as the first language, to be exact say not much better than a 3 year child

My 8 year old son only has a 20 word vocabulary and thats in English.

Thanks for that great explanation @stroutmail! I posted from my cell phone earlier so I was trying to keep it pithy.

I don’t know if it is technically accurate but at Indy on my rig I can feel that the car is leaning slightly left before centrifugal forces throw me right as I enter into a turn as well. At least I did when I was using roll and pitch but I only use g forces atm.

Whether correct or not my theorizing on the above is as follows. When you are making a turn where there is no banking in the road at all then about 100% of the centrifugal force of that turn is felt in the sway g’s. However, if it were possible to have a 90 degree banked turn, and assuming you are traveling at least a fast enough speed to keep from sliding downward, then about 100% of the centrifugal force is felt as a downward heave force. Therefore, using a simple logic tells me that a 45 degree banked road would about split the centrifugal force equally between sway and downward heave. Using your Daytona example of 31 degrees would spit the force 31% to downward heave and the remaining 59% to sway more or less I would think.

On this topic Pit is right that using surge and sway in the opposite direction that most of us do is technically the accurate way to simulate g’s. But there are a couple of problems trying to use that method on a simulator the way I see it. First, this method would not feel correct on a chair mover with a pivot. It would require a surge and sway table or something of that sort. But the problem with those are that when you reach the end of their range, there is no further feedback from them. On a chair that pivots, while the initial motion of it is opposite real g forces, once it has tilted, the feedback feels correct and remains until it is returns back to center. So, imho, for surge and sway tables to be effective, they must be very large. They need to be able to continue moving in order to maintain the feedback or at minimum enough that they can effectively use a washout. The second problem is that, to be immersive using this method, the viewing device must move with you too. Something like the Toyota simulator comes to mind. .
Hopefully though devices like the Oculus Rift, once perfected, will help alleviate at least some of this challenge.

http://www.nascarhall.com/_files/pages/2012-Physics---Compass-Tracks.pdf


BlazinH Generally I agree with your thoughts. To reach a complete understanding of the physics requires a free body force diagram analysis similar to the Nascar seminar. It requires a general understanding of high school trigonometry. The concept of the Normal force pushing on the car, tire friction and gravity
as the only three forces acting on the car is key. Then calculating the force components related to sway and heave oriented parallel and perpendicular to the track would be next.

Then we must consider that these forces cannot be duplicated in a Sim..they must be simulated.
How best to do that is a matter of opinion, debate and probably experimentation. I am becoming convinced that it will take more than 2 or 3 servos. That is why many are interested in incororating a g seat in conjunction with prsent seat movers.

I think it not possible or even necessary to duplicate real world movement..it is important to simulate it realistically from a mental perception. So I think we are in the mind fooling business..and we want to do it affordably so as many as possible can enjoy it. IMHO

Fun stuff for the curious mind!!!!!!