Internal shifting refinements

[MD]

I have experimented a lot with all sorts of refinements on these transaxle cars including developing an above tunnel shifting arrangement that reduces the gate and fore and aft shift travel. I must admit that I have not yet gone into the bowels of the gearbox to improve the shift mechanism itself.
There is nothing in this world that cannot be improved upon. Even small incremental efforts can make a difference. Forget sequential and hydraulic conversions that's way too much for my goofball but if anyone has any useful info on this aspect of the transaxle please share.

I am looking at you Dukster. No point hiding in the corner. 

[Vne165]

MD, Good discussion starter.

I know that there are mods that can be done in my open wheel race car Hewland LD200, such as radiusing the shift fork corners where they interface with the shift lever (hockey stick). Supposed to knock tenths of a second of the shift times and less chance of incorrect selection.

I'm sure there are mods and blueprinting that can be done.
Interested to hear any insights too.

Vne

[Duk]

What did I do? 

Any comments from yours truly, would only be theoretical and/or repeats of what others have done.

But here goes.

A light weight, solid centre clutch disc. Maybe a small diameter twin plate for motorsport use. But get that rotational inertia as low as possible, to make life a bit easier for the synchros.

Back cut gears.
As above, but you're at much less reduction of inertia, for a lot more cost/work.
I wouldn't cross drill them, unless you've got access to a CNC machine and can radius the drilled holes to the back cut sides. Sharp edges are stress rises and will probably be a point of failure.
Have gear sides shot peened after machining.

Nothing new in those comments.

But after that, I think you're really fighting the awkward design of the gearbox, with it having all of the syncros on the output shaft.
Even dog ring gear sets don't seam to be able to shift that quickly. I've never seen any videos of people doing clutchless upchanges, let alone having a flatshift cut and clutchless changes (like V8 Supercars were doing with H pattern Hollinger boxes, back in the day).



Mad Scientist Idea............... :

I had ideas about implementing a brushless DC motor on the input shaft, that would also do double duty as an electromagnetic brake.
This is similar to what Koenigsegg created for 1 of their transaxles, where they implemented an internal wet brake, to slow the input shaft speed faster than what the syncro could. They claimed up changes as fast as a dual clutch gearbox, but with a much lower weight penalty.
No mention on how that system could help down changes (I don't think it could).

The logic is/was to have a micro controller monitor the vehicle speed and knowing all of the gearbox ratios, it would know how fast the input shaft would need to be spinning in order for any given gear to be selected (obviously without over revving the engine.....).

Other inputs for the controller would be a gear lever position sensor, that would tell the controller what gate the lever was in and a strain gauge would tell the controller what direction the lever was being pushed or pulled.

So if the driver was going from first to second gear, the inputs would show the lever in the 1/2 gate and when the lever is being pulled to select second, with the clutch disengaged, the electric motor would act as a brake and slow the input shaft of the gearbox, to match the vehicle speed that it would spinning at, if second gear was already selected. Obviously the second gear synchro would also be trying to do this, but the motor (well, brake in this instance) should be able to do it with more authority.

The real plus would be performing down changes!
Where the electric motor could do a MUCH(!!!) better job of accelerating the gearbox internals and the clutch disc, than the poor little synchro ever could.
And the cool thing is (in theory), because the controller would be constantly monitoring the vehicle speed (the output shaft speed), it would be constantly changing the required input shaft speed to match the gear being selected, while braking.

[MD]

Haha, you get the Gong Award straight up !! I am visualising you in the white coat, monical and a smoking pipe touring the development lab.

I do like the supplementary synchro braking idea and doing lateral thinking on the subject, I can see a substitute for the motor because at the end of the day, electric motors are just a bunch of windings and electromagnetic fields. So if the field windings were actually formed on the gearbox shafts and casing, opposing fields could be deployed for a "no contact" braking of the gear sets just relying on opposing magnetic fields. Much like the same poles on magnets oppose each other. The greater the current flow in the "braking windings" the greater the rate of synchro slow down. How the current magnitude could be varied at will is perhaps for another discussion.

Getting back to a less exotic deviation, I got to thinking on the existing shifting mechanism. I am disappointed to see so much clearance in the gate activating the pull and push of the selectors. A 2mm gap would result in a substantial fore and aft gear lever travel for no gain. The design of the "hokey stick" selector could be changed from what it currently is to a substantial needle roller type bearing taking up the wasted space doing nothing. If this space is for any heat expansion provision, I do not think it is required because all the components would heat in the same area to maintain the relativity.

Do they serve coffee at your lab Duk...?