Tube sizing, Shaft dimensions, lobe dimensions, etc 2

[taintedcheese]

So I'm new here and wanted to know how you go about knowing/calculating the dimensions of various parts of the automobile. I've gone through a lot of textbooks on the subject but they don't really go into detail of adequate size or dimensions of camshafts, crankshafts, etc. I haven't seen anything that can tell me about fluid flow either (tubing diameters of power steering, and brakes in particular). Spring stiffnesses for shocks, lengths of connecting rods, spacings of engine chambers, etc, etc. Gear design for differentials, transmissions, is also another topic.

Another thing that I'm curious about is the exact fluid mechanics of valve body designs on automatic transmissions. How the hell are they able to actuate gears with pressures and flow?

I don't know if anyone can tell me about any of this from their experiences or if there's some literature out there for any of this (I've tried looking for it but with no success). I've looked at the SAE site for standards but it'll cost me more than I can afford.

Until I know or have access to most of these things I mentioned I'm just aimlessly CADing crap in Solidworks, simulating stuff in ANSYS and that gets old when you have no range of measurement or any sensible practicality behind any of it.

 

[GregLocock]

Trouble is you've covered more in one post than I've learned in 40 years. I've never sized a brake hose, worked on a cylinder head design, and the internal designs of auto trans are a special black art (I know enough to be dangerous there). You are right, SAE standards don't really help that much, papers are the useful product. The exception is the SAE spring design manual, that is actually pretty good.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[CapriRacer]

As Greg said, you've covered a whole range of components - all of which start with the basics of engineering, then branch off into specifics as needed. Much of what you seek is covered by proprietary information - info that component manufacturers will not share because their competitors might find out things of value.

Your best bet is to read up books on the specific compinent you are interested in - and there are many.

 

[dgallup]

Not many designs start from a clean sheet of paper. Either you've made several (million) widgets before or you buy the best widgets on the market and test them, tear them down and analyze the whee out of them then try to make a better widget. Very few engineers start at the top level vehicle design. Most of us work on some little niche widget on the vehicle toiling away in obscurity for the betterment of the entire beehive.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[CWB1]

To learn anything well takes time and effort. If you want to succeed in this industry expect 3-5 years minimum of intensive study on the job in a given niche before anyone considers you competent, much less an expert. If you divide a given vehicle into powertrain, chassis, and body systems, then figure there's a couple dozen niches within each its easy to see that one could spend multiple lifetimes in this industry learning.

One recommendation if you're considering this as a career - learn a proper engineering modeler (Solidworks doesn't count), preferably CATIA, Creo, and/or NX.

 

[romke]

I doubt whether carmanufacturers do design all the components you mentioned. Most of them rely quite a bit on the components industry and only supply sufficient basic information to them to let them come up with suitable suggestions. Of course those suggestions are then extensively tested and based on the findings modified as needed, but in the majority of cases detailing is left to the component industry, as is the actual manufacturing.

 

[bentwings2]

I worked in the automotive field for a number of years. Some R&D and some system design. One person doesn’t do a car. One person may design door handles...period. There is a lot to a door handle. The art team give you what they want, you inturn give then what you can design mechanically, then manufacturing gives what they can make, then the all important “bean counters” get in the act and say the whole thing costs too much. Now everybody has to rethink the project. Save a nickel in the auto industry and you are a hero. I worked a lot in assemble machine design. We were a relatively small player but as many as 30 of us worked on just a basic body assembly line. Just managing the CAD system was several people who just did that. A mistake in thinking how the Cad models were to be restrained cost us a month in design time.

So as noted, finish school maybe get an advanced degree. Heavy in math and machine design. I’d highly recommend a year or more in the machine shop learning basic machine tools.

I once worked with a very well educated lady. We were discussing an R&D project we were teamed on. She said she could handle the math and chemistry if I could do the design mechanicals as she felt weak there. And we did. Our project ran as required right out of the box. And no we never dated or went out for drinks.LOL

 

[GregLocock]

Hmm, I think you are being overspecific, although to be fair I haven't worked for a monolith manufacturer.

That is, the door handle engineer would actually be working on at least door closing mechanisms in all their glory rather than just the bit that sticks out of the car. At one point on the Mercury Capri there were exactly two chassis design engineers. My mate Brent copped the tricky bits, I was just wheels, tires, suspension, steering gear and steering column. A bit leaner than we'd do it today!

You wouldn't believe how many problems this wheel caused

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?