I like the DCT in BeamNG but in real life, when your car stopped and you shift in to Drive(D) the vehicle will start to move forward when you release the brake, which I dislike. In a car with a hydraulic planetery automatic transmission(AT) the torque converter replace the clutch. When you shift into D at zero speed. The transmission control unit(TCU) shifts the planetry gearset to first gear and the torque converter "slip" to transmit power so the engine won't stall at zero wheel speed. However, you have to brake in order to hold your vehicle still. The brake is used to counter the engine torque, multiplied by the TC Note that torque converter transmit power by fluid momentum, so there's no friction wear. Dual clutch transmission transmit power by clutches, like manual transmission. In real life, however, when you shift to drive and release the brake, the vehicle will still moving forward like AT. In BeamNG.drive you have to press the gas to move forward, releasing brake does nothing to the transmission control unit(TCU). In frequent low speed stop-go situation such as traffic jam. Clutches in DCT tend to wear much faster than in MT. This is because drivers use brake to control creeping speed. At half brake, the TCU make the clutch partially engaged, that makes the clutch and brake pads working against each other, wearing out faster. In BeamNG DCT you press gas to accelerate, press brake to decelerate even at very low speed. What's your opinion? have a discussion.Thanks.
CVTs can creep forward too in drive. My car does that, and every automatic I've driven does that. When designed well, CVTs are nice and smooth transmissions. I'm not 100% sure but I think CVTs may also use a torque converter. I've never experienced a DCT, but I want to at some point.
CVT use torque converter so creeping forward is perfectly normal. They also have a planetry gearset to select neutral and reverse. The torque converter locks up much earlier than ATs. I only dislike DCTs that creep forward in drive, because this is not due to torque converter (DCTs have no torque converter). Is that the TCU was programmed to partially engage the clutch in drive at zero speed even with no throttle to emulate AT-like handling, which is bad for the clutches, especially in traffic jams.
Would you happen to know why some CVTs make a slight whining noise (barely audible) in reverse? My car has always done it, and its normal for it but its interesting.
Most CVT in cars use a planetry gearset to select reverse, some of them use straight cut gears. When in forward(drive) all of the gears are rotating at the same angular speed so there's no relative motion thus no noise. When in reverse the output are rotating at the opposite direction relative to the gears from engine output shaft, thus make noise. Also unlike BeamNG in reverse there's a gear reduction so the topspeed is not as high as in drive.
I wonder if there could be adjustment in yours? Behavior sounds really bad design as in city traffic such will wear clutch out quickly, but haven't never driven such so can't really tell.
I don't have a DCT car so I can't tell really well, but many people do point out that a potential flaw. IMO that a DCT should disengage both clutches at stop even you don't press brake, only to engage them when you press the gas. There's a video in Youtube that after I watched it I realize there's a design flaw. A comment from that video Tim Gray one year ago I was caught in snow and stop and go traffic, lot of breaking and going slow, all the sudden got a warning tone and message the my transmission was over heating. Called the dealer they told me to leave space between me and the car in front of me to fully engage the clutch or to pull the car over and let it cool down for 20 minutes. Say what? That just seems to be a design flaw, did they not take in to consideration a lot of consumers buying these are city drivers in lots of stop and go traffic. Had I know this I would have thought twice about buying the car. Love it otherwise but this is a big issue. I worry what kind of damage this causes over time. Adjustment? Not at all. I never heard that you can enable/disable the creep in DCT, but battery-electric cars do can enable/disable creep, which I prefer disable. Torque converters don't "wear out" like clutch, so AT is best for busy city driving.
Welcome to why you're meant to shift to neutral in a DCT, hence why people hate them in city traffic. People moaned at VW that the DCT broke so quick when they started putting it in regular cars, yeah, because people abused it
I wonder if there is any technical reason behind that, I mean even Wartburg managed to have centrifugal clutch setup that did not creep, shouldn't DCT have the same kind setup just engage clutch at rpm X instead from the idle? Curious a bit of why they have chosen creeping setup.
The technical reason is behind the software not hardware. When creeping in traffic jam very slowly: MT: 1st gear, apply throttle and partially engage clutch to accelerate, fully disengage clutch while coasting or breaking on deceleration. AT: Drive gear, use brake to control speed. Slight acceleration(creep) is caused by engine idle torque, multiplied by the torque converter and the gearbox. The torque converter is a type of fluid coupling, which don't wear out like clutch. CVT: Same as above. The torque converter is connected betueen the engine and the belt drive. DCT: At stop or very slow speed in drive, when the brake pedal is significantly depressed both the clutch are fully disengaged. When the brake is released, the 1st clutch is partially engaged even the throttle is not depressed, creeping the car. The problem is if the driver only depress the brake very slightly(to control speed as above). both the brake pad and the clutch are partially engaged, which waste energy and increase wear significantly. In BeamNG.drive, however, at stop or very slow speed in drive, the clutch is always fully disengaged as long as zero throttle input,regardless of the brake position. The clutch only engage when throttle is pressed. Thus in traffic jam, the driver use throttle to accelerate and use brake to decelerate. The brake and clutch never engage at the same time. DCT is essentially two paralleled MT mechanism controlled by the TCU
I would design DCT software differently, closer to how it is in BeamNG.Drive, but that is just me Hmm, so to improve realism, we should have clutch target of DCT set to close to idle speed, clutch start to idle speed or bit below, then to set brake to engage clutch after certain percentage of braking. I wonder if that would flood support much about burned clutches, but then again BeamNG DCT Clutch seems not to be VW kind, I can't remember being successful at overheating it even once.
The reason DCT have creep is to "mimic" an AT driving feeling which, at least IMO, unnesseary. The solution is to adjust brake pedal "deadzone" at very low speed. For example 0%~20% brake pedal TCU control clutch partially engage. 20% pedal no clutch and no brake. 20%~100% pedal apply 0~100% braking force. At high speed 0~100% pedal is 0~100% braking. Of course I really wish that car manufactures do implement a setting to allow user enable of disable creep function. If you look at the Powertrain Visualization UI app, you'll notice only MT has a "clutch unit". MT has both a "clutch unit" and a "gearbox unit", while DCT only has a "gearbox unit". Thus BeamNG don't implement DCT clutch thermal simulation. Also IRL the torque converter can get damaged if the ATF(automatic transmission fluid) became too hot. Torque converter dissipate waste energy as heat to the ATF. In BeamNG vehicles with AT, you can hold brake and gas all day which only overheat the engine itself. ps. some CVTs(most in motor scooters) use centrifugal clutch, these vehicle have no creep.
CVT can has a creep, when starting scooter from the cold, it tends to creep a lot, well then again my scooter was tuned to max and after I tuned clutch to engage earlier and harder for faster launch from the lights, it kinda creeped on constantly, but it was fast too Why on earth anyone would want that creep on a car though, I mean car is not launching too well from idle anyway, one needs some rpm, so it must be as you say, which indeed sounds bit silly thing to do, but car makers do and sell. DCT clutch is in LUA files, you can find dctShiftLogic and there is clutch simulation for that.
Its the same reason why some CVTs try to emulate the behavior of a normal automatic transmission (which I hate... its a CVT and should behave like a CVT...). People tend to want something that feels familiar more than something different, even if it means forcing a different technology to behave in such a way.
So we all hate how they work, but it is what people want, thus that brings a question, are we people? Being different from majority has it's perks and drawbacks I guess.
But there's no clutch thermal simulation in DCT. And there's also some weird "manual mode"in CVT that emulate a 7-speed transmission. IMO the manual mode in CVT should be +increase engine RPM -decrease engine RPM, like the speed governor in constant-speed propeller aircraft.
That would be a cool idea. At least with the CVT in my Civic, it acts like a CVT and its S mode just makes it keep the RPM a bit higher (which is quite useful for passing as well as engine braking).
This is getting creepier! Did you know that some electric cars such as Nissan Leaf and the Chevy Volt have idle creep that cannot be turned off.
