Hirochi Sunburst

After a series of test driving I think the DCT in the Sunburst has serious mechanical issues. It's almost unnoticed on racing but it's very noticeable during gentle driving. The program (in the "controller" folder) might be part of the problem but I think the main issue is the physical model (in the "powertrain" folder). The test result is replicable

Realistic or Arcade gearbox (irrelevant), drive a Sunburst with DCT. Bring up the "Engine Info" UI app and observe the gear index, flywheel torque and wheel torque. Accelerate very gently at around 15% throttle. When the RPM reaches 3000 the DCT will upshift. During the upshifting, notice the large delay from the gear index change and the engine/wheel torque change, even after the large delay the wheel torque fluctuation doesn't make sense, it changes to -700 to -1000 Nm momentarily as the vehicle decelerate a whopping 3~5 km/h. You can use 4x slo-mo to observe clearly.

Now compare to manual gearbox, use Arcade gearbox so both the DCT and the Manual are shifted automatically by the game (control variable). Accelerate gently and wait for the upshift. During the upshift the vehicle is effectively coasting, the deceleration is very small, ~1km/h, the wheel torque changed to -30~-40 Nm (due to gearbox and differential friction) then back to positive again as the clutch pedal being released.

Let's analyze it. When the car is moving forward gearbox is in gear (i.e not shifting) both the wheel torque and the flywheel torque should increase (accelerating) or decrease (engine braking). If the clutch is disengaged (during shifting), the flywheel should be at or very close to zero while the wheel torque should be a small negative number due to gearbox friction. If the clutch is engaged (clutch pedal is released) abruptly while the rev is not matched, both the flywheel and the wheel torque should jump up (engine RPM too high) or jump down (wheel too fast). There's absolutely no condition for the flywheel torque to be positive while the wheel torque jumps negative during upshifting.

Use M mode to upshift has the same problem as D, lag and negative wheel torque while positive flywheel torque. Looks like the physical model programming confused it with downshift without throttle.

There's no reason for the vehicle to decelerate more on DCT compared to MT during, since the engine is always loaded (15% throttle translate to 60~120 N/m torque) maybe the gear behavior during shifting has calculation error.

Downshift also has problem, you can use M mode, drive gently with about 10% throttle and control the engine at around 2000~2500 RPM 2nd gear, use paddle shifter downshift to 1st, the engine will blip 25% to 50% depend on the model. However after shifting to M1 the throttle channel is blocked for about 1 second, during that time you can't accelerate, the gas pedal input is being ingnored.

 

Oh yes, and the window frames seem to turn into floppy spaghetti noodles a little too easily.

 

Thats a real thing, its cuz the glass gives away.

 

That's not how it works in real life. Window frames do not flop around like noodles in crashes like they do on the Sunburst and some other vehicles. They are generally much more structurally solid and can help keep the window intact in crashes. So it's rather unrealistic in a somewhat modern car. Not to mention the glass is extremely fragile on the Sunburst for some reason, which probably does not help. This was an issue affecting many more vehicles last year until eventually getting fixed.

 

I agree its over exaggerated but it is realistic. This type of thing should only happen on a real bad rollover.