Driveshaft in relation to the general layout

[checkered]

I was reading

through FIA's latest transcript exposé (WMSC's Renault hearing) and a number of things caught my attention. Among them was Paddy Lowe's description of the so called "screen grab" document and a few notions he offered on fundamental design principles:

This sheet mainly details prime dimensions for the car. In the submission, there are eight headings, showing the key dimensions covered. If I were to summarise the main utility of those dimensions, they would have to do with knowing how the key components in the car are positioned relative to the cockpit and the wheels. The main issue there is weight allocation in the car; positioning of the main components says a great deal about this. Details such as the drive shaft position and angle are crucial. (a blacked out, censored statement follows) This information would make it possible to understand how McLaren has positioned the key components and achieved a certain weight distribution.

Now, I've been under the impression that the current default design of the engine (and crankshaft), driveshaft (a.k.a input shaft) and gearbox output shaft is such that they're parallel, crank- and driveshaft being lined up one after the other. The driveshaft goes from the clutch plates into the gearbox. Whether they're all together at a very, very shallow angle in "neutral" conditions in relation to the "level" chassis itself is one thing, but Lowe's choice of words has me wondering whether the driveshaft (or gearbox input shaft, terminology being somewhat relative in this context) angle might actually be very different to the crankshaft angle in the MP4-22. This impression is by no means alleviated by the notion of the positioning and angle being somehow "crucial". (Then again, what isn't in F1 design?)

Conventional CV joints can be run only at an angle of a couple of degrees without friction losses that'd undoubtedly be unacceptable in F1 terms. Given the additional packaging problems I guess the use of a conventional CV joint in this purpose hasn't been all too easy to justify, but could it be that McLaren has been making a use of the Thompson coupling already? They'd do so only if there was a significant advantage to it. In theory the engine could be lowered still with radically different shaft angles, but given the severe design limitations on the engine itself is there actually any room to do so currently? Have the radii of the gears been determined in part by the "level" geometry and if so, could the gearbox be packaged in a smaller space still if the shafts within were placed at a certain upward angle towards the differential? (The CV joint being neatly "wedged" between the clutch and the gearbox.)

Is there some advantage to suspension geometry that could be attained in this way? And what about aerodynamic packaging, with the V8s it's already been pretty amazing just how "empty" the rears seem but hey, I guess every spare mm to optimise the flow can and will be used. I hope my thoughts on this haven't come over as completely senseless, this was just stream of thought sorta affair as I was wading through the document. I also wouldn't mind some discussion as to how one starts to put the fundamentals into place once designing a F1 car "from the scratch". Lowe seems to use the cockpit and the wheels as original references (in addition to general design geometry rules, I have to presume).


Image linked from cvcoupling.com

Thompson Couplings Limited - link

[zac510]

I got the impression from that short quote that they were talking about the shafts from the differential to the wheel hubs..
This would show the location of the gearbox mass relative to the wheelbase.

[checkered]

I got the impression from that short quote that they were talking about the shafts from the differential to the wheel hubs..
This would show the location of the gearbox mass relative to the wheelbase.

A much more

likely explanation. For a while there I was speculating on some very unlikely arrangements. Perhaps I was thrown off by the use of the singular form of "shaft" ... at least that's my story and I'm sticking by it. Until my next post or so. I'd like a ballpark figure on the "currently fashionable" diff to hub driveshafts' angles, though.

[donskar]

I assumed the reference was to the half shafts. I recall a few years ago when Ferrari lengthened their wheelbase by the simple expedient of angling the half shafts toward the rear (and making appropriate related changes of course!)

[Belatti]

I got the impression from that short quote that they were talking about the shafts from the differential to the wheel hubs..
This would show the location of the gearbox mass relative to the wheelbase.

Yep

angles between cranckshaft / clutch / gearbox should be 0

[DaveKillens]

I spent some minutes doodling some force vector diagrams, and to me it appears you can alter ride height of the rear depending on the angle of the halfshafts. If the wheel hubs are in front of the differential, there should be a tendency for the rear end to rise under acceleration. And the reverse, if the rear wheel hubs are behind the differential, the rear end of the car should squat under acceleration.

[manchild]

If you know the position of drive shafts than you probably can tell if gearbox is longitudinal, transversal or something exotic like planetary or so. I think that detail tells you more than anything else suggested.

[zac510]

If you know the position of drive shafts than you probably can tell if gearbox is longitudinal, transversal or something exotic like planetary or so. I think that detail tells you more than anything else suggested.

The most efficient layout of the gearbox has been established for about a decade. They're not going to venture away from that in a hurry.

[manchild]

If you know the position of drive shafts than you probably can tell if gearbox is longitudinal, transversal or something exotic like planetary or so. I think that detail tells you more than anything else suggested.

The most efficient layout of the gearbox has been established for about a decade. They're not going to venture away from that in a hurry.

So what has Williams done in 2006?

[Carlos]

Often a team can alter wheelbase and weight distribution by lengthening the gearbox. I've read this several times over the last ten years. In the past, while testing layout, teams sometimes used spacers between the engine and gearbox for this purpose.

[DaveKillens]

I suggest referring to Scarb's excellent web site and technical descriptions.
http://scarbsf1.com/transmissions.html
Getting back to the topic, in modern F1 cars, the gearbox dimensions are determined by the location of the engine output shaft, the centerline of the wheels, and additional factors such as location of the suspension locations and associated gear. Ideally, for maximum power transmission the centerline of the differential and the centerline of the wheel hubs should be in a straight line. But those reference points may be altered, as pointed out earlier by Ferrari's simple act of just changing the rear suspension mounting arms. An increase or decrease of the wheeelbase can easily be accomplished, but there are other reasons for displacing the centerline of the differential/wheel hubs.
My personal theory is that the movement of the suspension may be altered when torque is applied to the drivetrain. My belief is that you can get the rear end of the car to rise or squat under acceleration by moving the locations of the rear wheel hubs by moving them fore or aft relative to the centerline of the differential.

[donskar]

Very interesting:

"My personal theory is that the movement of the suspension may be altered when torque is applied to the drivetrain. My belief is that you can get the rear end of the car to rise or squat under acceleration by moving the locations of the rear wheel hubs by moving them fore or aft relative to the centerline of the differential."

Is it not true that suspension pickup points also play a large part in anti-dive (to reduce nose dive under hard breaking) and anti-squat (to reduce nose rising/rear squatting under acceleration)? For a classic double wishbone system, I thought that having the forward lower a-arm pickup point higher than the rear pickup point would reduce both dive and squat.

Looking forward to someone clarifying my fuzzy understanding on this one.

[riff_raff]

With the very limited amount of suspension travel on current F1 cars, I don't understand why they bother with CV joints at all. I design rotorcraft drivetrains and we use flex couplings to deal with angular mis-alignment/shaft flexing. No moving parts, splines or lubrication issues to deal with, and they are light and reliable.

http://www.fus.goodrich.com/productline ... Index.html

http://66.70.211.189/kaflexpage1.html

[shelly]

IIrC there have been some trials in the past with angled driveshafts.

With the engines being shorter in order to have a clean coke aero teams can run a spacer between engine and gearbox or they can put the gearbox forward and run an angled drive shaft in top view. This would have also the advantage of freeing up some space above the diffuser (if you can make a thin enough rear crash structure and make a smart transition for the rain light).
You could also maybe exploit some magnus effect (inboard component) instead of having to shield the shaft like in 2012-13.
Does any body know the mechanical disadvantages of the angled shaft set-up? How much would it be possible to angle it?

[Moxie]

I spent some minutes doodling some force vector diagrams, and to me it appears you can alter ride height of the rear depending on the angle of the halfshafts. If the wheel hubs are in front of the differential, there should be a tendency for the rear end to rise under acceleration. And the reverse, if the rear wheel hubs are behind the differential, the rear end of the car should squat under acceleration.

So is the converse true? Under braking the rear end rises or falls depending upon the orientation of the halfshafts? Nothing like having more rake/downforce when the breaks are on.