Seamless Gearboxes Find The Gap

[thisisatest]

auto,
are you saying that, when in gear and accelerating, shifting to neutral and continuing to accelerate, but at a lower rate, will disengage the drive? i dont think so, until the drive torque reaches almost zero.

[autogyro]

I will repeat my questions to you then:

You claim that a manual stick/clutch/foot combo is more efficient than a 40ms current F1 box? Would you agree that teams use the (when rules allow) systems and technologies that are the best to reach the goal of the fastest chassis possible? If a quickshift/zeroshift box is the slower torque transferring and lest efficient of the two, then why don't the teams use a manual dog box???

Speed rules.....if a manual were faster the teams would run those boxes. I give you the benefit of the doubt as much as I can, but your beef seems to be hostility due to your own box.

I do not believe the actual shift 'overlap' (that is the time for torque to change its path through the geartrain), is faster in a 'seamless' shift system.
A power on racing shift with a dog ring mechanism and a skilled driver is faster IMO. 40ms is easy to achieve with a good racing dog ring shift. In fact the mechanism within the dog and hub assembly is simpler than the 'seamless' mechanism which is far more complex in its dynamics.

However, even if the 'shift overlap' is faster in a particular 'seamless' mechanism, to achieve a gear shift that does not result in a sledgehammer bang of engagement in the 'seamless' type, the input torque from the engine has got to be modulated and the rear powertrain damped.
This essential requirement is achieved by de-clutching the engine from the input shaft, controlling the engine's torque output through the EMU and by placing a radial damper in the rear powertrain.
All these control methods in various application, reduce the available torque at the output shaft of the gearbox and are inefficient. Lap time will suffer.

The only logical reason for F1 cars to use the 'seamless' gear shift systems is to patch up the use of inefficient layshaft stepped gearboxes from a transmission concept not improved upon since the 1890's.
With the domination of high downforce aerodynamics,it has become almost impossible for the driver to use skill to change gear with these obsolete gearboxes with so little time available under deceleration. without upsetting the car and the energy recovery systems.
Calling these systems for gearchanging 'seamless'is incorrect and masks the fact that layshaft technology is ancient and essential development in powertrains is being held back by this obsession with downforce aero.

[autogyro]

auto,
are you saying that, when in gear and accelerating, shifting to neutral and continuing to accelerate, but at a lower rate, will disengage the drive? i dont think so, until the drive torque reaches almost zero.

No.
I am saying that it is the reduction in the torque applied to the lower ratio gear that disengages that gear.
This torque reduction can be achieved from either an application of torque to the higher gear (with the resultant torque reduction from the lower gear), resulting in a bang of engagement at the higher gear (as postulated by those promoting the systems as all that occurs), or from any method that reduces the torque input to the gearbox and therefore the torque applied to the lower gear.
The torque applied to the lower gear does not have to reach zero for the sprung loaded selector fork operating on the low gear selector/dog ring to spring this dog ring out of engagement.
It simply has to overcome the torque load from the gear on the dog ring at a pre-set level (a balance of torque reduction and the selector forks spring rate and position).
The 'gap' in drive torque through the machanism will then occur between this low gear disengagement and the high gear engagement, so the system cannot be 'seamless'.

I know this all to be true because I have built and tested such mechanisms as early as the late 1970's. At that time the fine control using electronics was not available, the shifts were crude and in anycase we considered the idea of controlling the shift mechanism of a layshaft stepped gearbox to be a big step backwards. I visited Marenelo in the late 1980's and in the early 90's and spoke with Ferrari's gearbox people. I mentioned the so called 'seamless' shift concept to them (we called it an automatic shift mechanism for manual layshaft stepped gearboxes) and the idea appeared in Mansels Ferrari shortly afterwards.
Notice, I have not mentioned any other gearbox concepts.

[Lurk]

or from any method that reduces the torque input to the gearbox and therefore the torque applied to the lower gear

Maybe I missed the post when someone talked about that but how you reduce torque at gearbox input while gear engagment is done in only 1 or 2 ms - and not 40ms like I read ? (maybe it is even faster nowdays)
Is it enough time to shut down a cylinder?

[autogyro]

or from any method that reduces the torque input to the gearbox and therefore the torque applied to the lower gear

Maybe I missed the post when someone talked about that but how you reduce torque at gearbox input while gear engagment is done in only 1 or 2 ms - and not 40ms like I read ? (maybe it is even faster nowdays)
Is it enough time to shut down a cylinder?

Where do you get your figure of 1 or 2 ms?

Engagement of any gear is zero time at the point of full engagement.

It is the time taken to move the mechanism to disengage one gear and engage the next (the shift overlap)that needs to be guoted.

Even this is not the complete time taken for the gear shift if the system also modulates the torque input from the power unit and damps the rear drive train to absorb any kind of fierce shift.

All the 'time' that this is going on there is a loss of potential torque transfer from the engine to the drive wheels.
All you hear is the engagement not the rest of the shift before this occurs, or the continued modulation of drive torque and damping that occurs afterwards.

[Lurk]

It is the time taken to move the mechanism to disengage one gear and engage the next (the shift overlap)that needs to be guoted.

Sorry if I do not use the correct terms but this is what I was talking about: It takes 1 to 2ms to disengage a gear and engage another.

[autogyro]

It is the time taken to move the mechanism to disengage one gear and engage the next (the shift overlap)that needs to be guoted.

Sorry if I do not use the correct terms but this is what I was talking about: It takes 1 to 2ms to disengage a gear and engage another.

I would say that such a time would be unlikely with the so called 'seamless mechanisms' I have seen.
I certainly would not expect it with the zeroshift shown.
In anycase a manual dog box can be shifted almost instantly.
For a start the shift disengage and engage motion is a strait line with a dog mechanism and not through two right angles like the zeroshift.

[thisisatest]

it is my understanding that one way to lower the applied torque, and to soften the "hit" during an upshift, is to partially disengage the clutch, allowing it to slip a little, just for a moment.
this is all what ive read, i have not built and tested one, or even seen one working in real life...

[autogyro]

it is my understanding that one way to lower the applied torque, and to soften the "hit" during an upshift, is to partially disengage the clutch, allowing it to slip a little, just for a moment.
this is all what ive read, i have not built and tested one, or even seen one working in real life...

Yes, this will reduce the input torque to the gearbox and AFAIK also always has to be accompanied by control of the engine through it electronics to prevent an increase in rpm. This also adds to reduced input torque.

This will indeed reduce the engagement 'bang' as the higher gear engages.
It will also allow the lower gear to disengage before the higher engages.
There is therefore a 'gap' between disengagement and engagement, where there is NO torque transfer from input to output of the shift mechanism.
The shift mechanism is therefore NOT seamless or zeroshift.
It does not matter how small the gap is, it is still a gap.

The clutch disengagement and engine control is NOT needed with a skilled driver race shifting a dog ring manual layshaft stepped ratio gearbox.
The dog ring box can be shifted at least as fast as the so called 'seamless' systems, if not faster.
Therefore the so called 'seamless' gearbox must be less efficient in changing the torque path from input to output on a layshaft stepped ratio gearbox.
In the 1970s we stopped development of such systems because we considered them a backward step.
It is only the domination of downforce aerodynamics that has masked these truths.

[Steven]

Same old, same old