asking about gearbox

[autogyro]

You could always reverse the compressor on the HERS

Ooops, I mustnt give them any ideas, forget I said that. (I have more )

[machin]

You still will not beat a power on dog box up shift with a good driver.
Makes a big jolt though.

...unless you replace the old style dogs with the seamless-shift dog arrangement in which case it will be faster and smoother.

Check out this graph from a car fitted with a seamless shift where there is no change in Accelerative G right through the shift point (the black line which is more or less horizontal through the centre of the screenshot), and therefore no "pause" in the speed trace (red line): that's impressive. And that's without any input from a KERS or HERS system, and neither is it fitted to a little F1 engine with minimal inertia: its connected to a 4.2 litre TVR V8!



I'd love to see a driver with a traditional dog box replicate that... HINT: due to the way they're manufactured, it is impossible!

[autogyro]

So what exactly does that mean?
It is certainly good marketing for zeroshift.

[machin]

Well the x-axis (horizontal along the bottom) is time, with time=zero (start of manoeuvre) on the left. The red line shows velocity increasing as time increases towards the right. As you can see it is constantly increasing with no steps as you would expect with a conventional gearbox. The dark line is acceleration. It starts off low (as the clutch is fed in and grip is limited), reaches a peak and then follows the shape of the engine's torque curve (multiplied by the overall gear ratio in use at the time), but is reduced as speed increases because aerodynamic drag increases. With a conventional gearbox which has breaks in the torque during gear changes we would expect to see a large momentary drop in acceleration followed possibly by a shock pulse as the next gear is engaged (depending on how smoothly the engagement is made).

The equipment required to generate such a trace is relatively cheap.... I would say even most amateur competitors have this capability. I have this equipment which I can mount on any car.

[jz11]

I think this is a good representation of the current discussion discussion:
edit: changed the wording a little bit

[flynfrog]

no need to get personal guys

[machin]

I've modified the above diagram to show what we would expect to see if the gearbox had a standard shift mechanism. There would be a very clear loss of acceleration during the shift (no engine torque being applied to the drive train downstream of the gearbox). The acceleration would be slightly negative (note Zero "G" is half way up the chart, the lower portion shows negative "G") due to the impact of aerodynamic drag, and rolling resistance. As the next gear is engaged we would expect to see a sudden increase in acceleration, most likely with a peak due to the engine inertia, assuming you hadn't rolled off the throttle sufficiently (you can feel that kick if you come on the throttle too early during a normal shift!).

On the speed trace (red line) the loss of drive torque exhibits itself as a slight dip in the speed during the gear change.

The faster the shift the shorter the window where no torque us applied and therefore the less speed is lost.



As I say, the equipment needed to produce such data is easy to get hold of, so anyone saying that "the data doesn't exist" either hasn't looked hard enough, or hasn't bothered to generate their own. I could post many graphs showing what a "slow" gear change looks like, but I know that whatever data I post Autogyro will say "Your driver is crap" or "the leverage ratio is incorrect"....

[autogyro]

Hahahaha one minute its velocity then its acceleration then its G.

Gad I must find out what kind of vegetables these super heroes eat.

[NL_Fer]

Without getting into details, Ross basicly tells us the how a seamless shift works.

http://youtu.be/FipsVzlcUL0

[machin]

Hahahaha one minute its velocity then its acceleration then its G.

Just checking, but you do know that "Speed" measured in one specific direction = "Velocity" (so when talking about acceleration of a vehicle in a straight line, velocity = speed) -That's the red trace in the two graphs.

And Acceleration can be measured in "G", where 1G = 9.8ms^2, and that equals a 21.9mph increase per second (i.e. 1G is quite a high acceleration; equivalent to 0-62mph in about 3 seconds). That's the black line in the two graphs.

...right????!!! I'm sorry, I didn't think I needed to explain this. My bad.

I'm guessing that if I say that the velocity curve is simply the integral of the acceleration curve it'll really blow your mind?!

[NL_Fer]

Machin. 100% correct. The over-running dog system can be totally seamless whereas the traditional dog box cannot.

NL_Fer. A DSG box can be totally seamless. During the handover period, torque is being transferred by both clutches and both ratios simultaneously. The difference between shifts on a DSG and the over-running dog is the DSG can prolong the "handover" for as long as the designer wants (at the expense of high friction loss). Handover time on the over-running dog system is very short and is a function of the compliance of the components transferring the torque.

Still disagree, high friction loss during an overlap period, wouldn't be very seamless. And some cannot even overlap, because of the friction plates configuration.

Btw i have driven DSG for three years, i wouldn't call it seamless at all. It doesn't really matter, they don't use dual clutches in F1, only dual clutch levers in the steering wheel.

Also i remember, already in 2009 they were using the kers unit, to smoothen out the shifts.

[machin]

Also i remember, already in 2009 they were using the kers unit, to smoothen out the shifts.

Remember; Seamless shift pre-dates KERS by about 2 or 3 years, and that TVR doesn't have any sort of Hybrid system. I'm pretty sure that what I read was they were using the KERS system to fill in the "gaps" in the output power that you get between the power peaks in each gear (i.e. even when the gear is fully engaged, and not specifically during the shift itself). This means that rather than getting a rise and fall in the available power as the car goes through the gears you can get a smooth power output... I'll draw what I mean.. give me a few moments....

[machin]

So the graph below shows maximum power available compared to road speed on the x-axis (not time in this case). If the car doesn't have a KERS system then the maximum power available at any road speed is defined by the top of the green curves... so the driver gets a sort of "rising and falling" amount of power available as the revs rise and as he goes through the gears.

With a KERS system they can use it to "fill in" the gaps between the ICE's power peaks (shown in yellow) and give the driver a constant amount of power (Orange line), regardless of Engine RPM and road speed, so you don't get the "lumpy" power delivery.



Also remember that the KERS motor in F1 is mounted "upstream" of the clutch, on the engine: so if a gear isn't engaged then the KERS motor can't supply power to the rear wheels any more than the engine can... Both the engine and the KERS motor rely on the seamless shift mechanism to allow power/torque delivery without interruption during gear shifts. A "non-seamless" KERS powertrain would still exhibit the gaps in power/torque delivery during gear changes.

.... You could mount the motor downstream (driven-wheels side) of the gearbox... then it could be used at all times, even if the gearbox and/or clutch is disengaged... The downside of that arrangement is that your motor cannot take advantage of the gearbox ratio, and therefore the motor has to be bigger and heavier to generate the same power (i.e. more torque at lower RPM).

[NL_Fer]

No what i mean is. Pre KERS they used a ignition cut, to reduce torque during shift. Before seamless boxes, the ingeruption was long enough to match rpm, before engagement, like Autogyro is telling us.

After the introduction of seamless boxes, the interruption could be very short, just to remove the torque spike on engaging the gear.

After KERS they started to shift with no ignition cuts, just the KERS unit acting as a brake on the crankshaft, just to smoothen out the torque spike.

I mean, wih a seamless shift, the rpm difference is actually introducing a torque spike in the driveline.

[machin]

After KERS they started to shift with no ignition cuts, just the KERS unit acting as a brake on the crankshaft

Ahhn Yeah... I see. Although by all accounts the Zeroshift demonstrators tested by numerous magazines were very smooth and they didn't use any sort of "auxiliary engine braking unit".... If F1 teams do use the KERS motor that way I would suggest that it is probably only a minor improvement over an ignition cut strategy and not at all an integral part of the seamless-shift technology; remember drivers were already saying that seamless shift was smoother than "the old way" back in 2007 (see quote a few pages back); i.e. before KERS was introduced.