[audioholic]

Originally Posted by Sutripta

Actually it would be great if you (or anyone for that matter) could give definitive information on the sequence (algorithms)Without this one can bandy about anything which sounds reasonable. eg. why not keep the clutch completely disengaged when at standstill with foot just on brake (enough to operate the brakelight switch)

Definitive information is something I unfortunately dont have with me. I get to interact a lot with a guy who has spent a few years calibrating transmissions for various cars but of a single manufacturer. I chat with him whenever possible and know only a few details on what is what. For this certain dual clutch setup he calibrated, the creep would be disabled in the following conditions:
1. Brake pedal pressed for more than a certain duration irrespective of pedal pressure
2. Auto hold or hill hold active, which also refers to a firm press of the brake pedal.
3. Clutch pack overheat.

Now as to your question on why not disengage clutch for a slight press of the pedal, it will get annoying. Take my own Celerio AMT for example. The clutch disengages as soon as the brake pedal is pressed which will make low speed crawling such as parking getting used to. So much that I have improved my skills of parking the car with least usage of brakes so that the speed of the car is consistent. Same thing for the DSG. If this is implemented then at low speeds the behavior will get annoying and will be undesirable. Hence they let you creep at low speeds and maybe for a few seconds before completely disengaging the clutch. Also in dry clutch setups this is done more carefully like I said before to avoid clutch wear and overheating.

[Sutripta]

Originally Posted by audioholic

Now as to your question on why not disengage clutch for a slight press of the pedal, it will get annoying.

Disengages completely only when the vehicle has come to a complete halt.

Now the question on brake line pressure sensor. Which cars have it? Does its reading show up on the OBD? Error codes?
As for frying the clutch is concerned, AMTs and DSGs would be similar. And I'm sure Marutis don't have a brake line pressure sensor.

Regards
Sutripta

[Vid6639]

Originally Posted by Sutripta

Actually it would be great if you (or anyone for that matter) could give definitive information on the sequence (algorithms), and more importantly why a particular one was chosen and not (equally reasonable sounding) others. Without this one can bandy about anything which sounds reasonable. eg. why not keep the clutch completely disengaged when at standstill with foot just on brake (enough to operate the brakelight switch)? And start engaging the clutch when accelerator movement is detected.

Regards
Sutripta

In my Skoda dual clutch I can feel the car fighting the brake when brake pedal is lightly pressed, when I press a little more the car stops fighting. This might be data from from either pressure or position sensor. I do not know this definitely.

As audioholic mentioned it is also time based and it stops fighting after 2-3 seconds.

It immediately engages the clutch when you lift off the brake pedal even before you press accelerator.

If you are inching forwards in slow traffic you are usually fighting the car as it never disengages the transmission.

The Skoda has auto hold so that disengages transmission when I press the brake pedal firmly. A light press is not enough to engage "auto hold".

[Doo_Dev]

Originally Posted by Sutripta

Disengages completely only when the vehicle has come to a complete halt.

Now the question on brake line pressure sensor. Which cars have it? Does its reading show up on the OBD? Error codes?
As for frying the clutch is concerned, AMTs and DSGs would be similar. And I'm sure Marutis don't have a brake line pressure sensor.

Regards
Sutripta

Almost all the cars with DCT will have a brake pressure sensor because the clutch engagement /disengagement is based on vehicle speed and brake line pressure.
Case 1: During the phase of coming to a standstill, the driver presses the brakes. When the vehicle comes to a standing stop, the driver may not shift to N. In this scenario, if the clutch is not disengaged, it will result in a clutch slip. So the ECU looks at vehicle speed and brake line pressure and then disengages the clutch
Case 2: During the launch, when the user lifts off the brake, the ECU reengages the clutch as the pressure in brake line drops.

However, in many vehicles, instead of having a physical sensor, there is a possibility of having a mapped sensor. In this setup, a predefined current vs brake pressure table is stored in the ECU and the brake pressure is estimated based on the current to brake valves ( inlet and outlet solenoid valve).

The smoothness of the creep depends on how accurately the stored table is representing the real-life scenario.

[Doo_Dev]

Originally Posted by dinu2506

I have a slightly different view here. One of the best CVT boxes I have used (and still using) is the one that is there in the previous gen Audi A4. This gearbox has proven to be one of the most reliable CVTs globally and it has dealt with the 380Nm torque of the 2.0 TDI engine consistently. The rubber band effect is not noticeable and if you are in the mood for some fun, the sport mode also performs like a TC with noticeable gear changes. I don't know how they do it but in spirited driving in Sport mode, you can see the RPMs dropping during each upshift. So, I feel, CVTs are probably better suited high powered/torque vehicles. My sister has a CVT Nissan Micra, there you can really feel the rubber band effect.

Originally Posted by manjunathkl

Thanks for this info, I always thought all Audis and VWs are DSGs. I now know even CVTs can be built for better performance and torque. CVT as tech itself is not the problem, it is the design of the particular CVT box that can determine its performance. Premium segment cars like Audi can plonk this well made CVT. All TC boxes aren't the same, similarly, all CVTs aren't the same.

Mass-market cars will go for the run-of-the-mill CVTs which doesn't have all these goodies.

One of the main difference comes from the type of belt used and the control setup. In the case of Nissan, the belt is a metal belt whereas all Audi came with a chain belt CVT. A chain belt is capable of transferring higher torque with fewer losses.

Additionally, the rubberband effect can be reduced by maintaining a fixed ratio during acceleration. However, with CVT, maintaining a fixed ratio goes against the basic premise of CVT. However, by tracking the engine speed and vehicle speed, the Audi Engineers could have set up a control strategy in ECU to control the pulley movement and maintain a fixed ratio only during hard acceleration.
Since this additional control needs more precisely controlled pulley movements, the costs are prohibitory in a mass-market car compared to a premium car.

[Sutripta]

Originally Posted by Doo_Dev

Almost all the cars with DCT will have a brake pressure sensor

Some more details of the pressure sensor would be nice.

However, in many vehicles, instead of having a physical sensor, there is a possibility of having a mapped sensor. In this setup, a predefined current vs brake pressure table is stored in the ECU and the brake pressure is estimated based on the current to brake valves ( inlet and outlet solenoid valve).

Now this is getting complicated. What controls the solenoid valve current? Servo or proportional? A line diagram showing the hydraulic system, and the electrically actuated components in it would be helpful.

'Almost all cars' and 'many vehicles' seems to be contradictory.

Regards
Sutripta

[Doo_Dev]

Originally Posted by Sutripta

Some more details of the pressure sensor would be nice.


Now, this is getting complicated. What controls the solenoid valve current? Servo or proportional? A line diagram showing the hydraulic system, and the electrically actuated components in it would be helpful.

'Almost all cars' and 'many vehicles' seem to be contradictory.

Regards
Sutripta

'Almost all cars' and 'many vehicles' are not contradictory as they are to be interpreted in different contexts.
'Almost all cars' .... refers to the usage of a pressure sensing system
'many vehicles' .... refer to the type of pressure sensing system

From my profession, I have not come across a design without a pressure sensing system. However, it would be wrong to make a general statement that all cars will come with a pressure sensing system.
The question is whether the pressure sensing is based on a physical sensor or a mapped sensor.

Note: I changed the name from pressure sensor to pressure sensing system to avoid confusion.

[sandygordon]

Day 7 Update

Kia has accepted a faulty gear actuator assembly and the part will be replaced under warranty.
They have placed the order and the service center is waiting for arrival of the same.
Hope it won't take an eternity to get my car back

[beast_within]

Originally Posted by sandygordon

Day 7 Update

Kia has accepted a faulty gear actuator assembly and the part will be replaced under warranty.

Does this imply that the earlier software updates done during the round 1 and subsequent rounds were not relevant and they failed to do a proper root cause analysis initially?

[audioholic]

Originally Posted by Sutripta

Disengages completely only when the vehicle has come to a complete halt.

This type of a direct logic will be the annoying part when we are trying to park the car by moving small distances at a time and stopping. During this scenario, the clutch will still be kept at creep position so that it will mimic the behavior of a torque converter automatic which fights against the brake during creeping. It will allow you to practically move at any speed from 0. Whereas in my AMT, every time I press the brake, speed will drop to 0 and when I release it will get back to creeping speed of some 6-10kmph which requires the need of repeated braking. In comparison, when you always have the clutch at its bite point in the DCT, you can just increase or decrease brake pressure and get a smoother parking behavior. In fact, the initial Maruti AMT software was even worse and made parking a pain especially during reversing. After a software update, they made reverse creep slower and more gradual.

Originally Posted by sandygordon

Day 7 Update

Kia has accepted a faulty gear actuator assembly and the part will be replaced under warranty.

Looks like they finally realised what could be a bigger culprit than just the software. The actuator is what I was always doubting than the software. Its an easy way out for the dealership to blame software since all they have to do is to reflash the software be it a new version or the same and try to pass it off.

[Salnv]

Originally Posted by sandygordon

Day 7 Update

Kia has accepted a faulty gear actuator assembly and the part will be replaced under warranty.

How long they are expecting to get it fixed? I had a DCT issue with my ford ecosport and they changed the actuator assembly but the problem didn't get fixed ( there the issue was just a sudden jerk at random gear shift ) and finally they replaced the complete gearbox

[Sutripta]

Originally Posted by Doo_Dev

The question is whether the pressure sensing is based on a physical sensor or a mapped sensor.

So a derived value. Derived from what?

Now the enlightenment on valves and currents please.

Regards
Sutripta

[sandygordon]

Originally Posted by Salnv

How long they are expecting to get it fixed?

The parts will arrive from AP in 3 to 4 days.
Hoping to get the car early next week.

[ecenandu]

Originally Posted by Sutripta

So a derived value. Derived from what?

Now the enlightenment on valves and currents please.

Regards
Sutripta

As the Kalman filter, using different sensor values to deduce another measurement.

What he is referring to here is a simple map lookup, for example, a single-dimensional map.

[vDragon3]

Originally Posted by sandygordon

Day 7 Update

Kia has accepted a faulty gear actuator assembly and the part will be replaced under warranty.
They have placed the order and the service center is waiting for arrival of the same.

This is what should have been done by KIA when you had initially reported the issue! Software update stories just seem a reason not to replace a faulty part under warranty.