Gear Set Design Considerations 1

[fhalperformance]

I am interested in designing a gear set for one of my cars and although I have a good understanding of how transmissions work and what ratios I am looking for, it has come to my attention that there may be other factors concerning the design that I am not considering. I was reading an article about Xtrac transmissions a while back that mentioned there were special considerations they take into account such as tire contact patch/traction, etc. Anyone know of any reference material that has equations or information on that? Thanks.

 

[MikeHalloran]

It's not specific to automotive applications, so you won't find much about tire contact patches there, but you need a copy of Darle W. Dudley's "Gear Handbook" to get started.





Mike Halloran
Pembroke Pines, FL, USA

 

[IceStationZebra]

One could imply that they are simply trying to determine the maximum torque loads. But awhile back I read an article on Xtrac and, though the author didn't state it, made it sound like they build some torque compliance into their design. You would have to know the application pretty well so you don't get into fatigue or gear mesh/shaft deflection issues. And obviously you would need to take the complete system torsional stiffness into consideration.

ISZ

 

[dicer]

The gear set DESIGN has nothing to do with anything that affects the power transfer to the road. Why? Because said gear set could be used for a nonvehicle application. Design of a gear set will involve the loads the gears are to see, the materials to be used the heat treats, the type of machines used to make said gears etc.

After rereading your post, its more of a what off the shelf transmission or gear ratios should I use. Correct?

 

[ccfowler]

If you are seriously considering building a transmission "from scratch," then your budget must surely be large enough to hire competent, experienced gear design engineers, manufacturing facilities and staff, etc. Gearing and gear system design is highly dependent upon the "art" of engineering in addition to the sciences involved. Every aspect of gear design is interdependent on multiple factors including but NOT LIMITED to specific material selection, machining choices, heat treatment, gear case and beaing design, weight limitations, configuration constraints, loading characteristics, anticipated service life, lubrication, and operating temperatures.

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[fhalperformance]

I was curious about what I read in that article about xtrac and whether or not it was another consideration I should look into. I do not plan to use off the shelf gears as I'm not really concerned with making any sacrifices to my reliability requirements. Lots of good answers and advice here thanks all.

 

[geesamand]

I know a fellow who designs gearing for G-Force (similar to Liberty or X-trac) and basically he said he there are no strength equations in the literature that line up cleanly with racing applications. Research and development leads to an empirical understanding of the strength rating of these designs.

Basically, if you don't want to compromise reliability, and you do your homework and learn all of the gearing design parameters required, you'll still miss the mark and end up with too much metal or losses or not enough capacity. I would use the knowledge that's out there and spend my time adapting a well-tested racing box that has the torque and ratios I need.

 

[BrianGar]

I would Imagine what you read went something like ''a gearbox is designed not with the engine torque in mind, but the amount of torque that can be passed through the contact patch before it breaks away''

Meaning, if you have got a 1000hp engine, and small bicycle wheels, your never going to be able to load the transmission enough for the gears to have to cope with that 1000hp, the tyres will simply spin.

This analogy is far more common in light racecars - treat it with care.

BUT, you cannot ignore the loads that arise from engine braking on a more normal car.

If you change down at 100mph, to say second(If your box allows) then you may see very high tooth loads depending on how large a rotating mass you have the engine side of that gear.
If your car is FWD then its amplified.
You will break away a FWD with little hassle when pulling away since the weight transfers to the rear(hate that term).
This may give you a false sense of security and lead you to think you have a great box.
Try changing down to a very low gear at high speed when all the weight transfer comes onto the front wheels and tell me how you get on then.

Its a complex subject as you probably realise.

A few points I would say watch out for are,

If its a light racer, watch out for engine braking, and remember what down force does for grip, downforce not being present off the line.

Dont put top notch expensive gears into a weak casing. No point spending an age designing super gears, with brilliant heat treat if you case wont allow them to run to the spec the designer called for.

Dont assume you can always rev match when down changing. You may miss a throttle blip one time.

5mm steel dividers can be laser cut, and built into a weak casing keeping the gears and shafting together. This can range to an easy or very complex operation depending on case, and set layout.

Details of such a plate can be seen in link below. I know its more of a bandage fix for a bad design, but, it can be very hard to make an alloy case stiff enough without adding weight.

http://stweb.peelschools.org/sssweb/TTI-WEB/cars/pictures/AudiTransaxleUpgradeStepbyStep/index.html

Theres loads more, but that may help you a small bit.

Sometimes an answer can lead to a hundred more questions.

Brian,