Strongest Gearbox Possible

[Geoff]

As long as it's being proposed to cast an entire new gearbox, perhaps even having a separate diff box, why not jut design a box casting that will accept standard race gears?

I agree. If you want to cast a new gearbox, why not design it so it fits standard components? You could design it around standard bearings, standard shift forks, standard gears etc. Standard meaning more plentiful than parts for a 16+ year old car...

There are tons of gear options for standard racing gearboxes.
http://www.pegasusautoracing.com/group.asp?GroupID=TRANSGEAR
And parts available too
http://www.hewland.com/
Plus you can find used gearsets for these as well (or complete gearboxes to part out to fit int the newly designed case).

The differential could even be a standard so there would be choices for limited slips and possibly ring and pinions.

Or if someone has $ they could have Hewland develop a complete gearbox for the C900/99 :-)

[99Super]

though the current design is quite good at keeping the weight over the wheels....making them the good snow cars they are

Well, yeah, there is that...

[Nate]

I'm all for using new standard components. However, I'm not an engineer so that is beyond my abilities. What would we use for ring and pinion gears or an output shaft?

Investment casting,

All sorts of waxes can be used ranging from bees wax and vegetable waxes to synthetic polyethylene waxes. Most of the time an "alloy" of wax will be created from several different types of waxes to give the best properties.

The sand isn't "packed" per say, it is used as a coating. A slurry is made from different types of silicas and then is applied to the wax pattern. The slurry looks and feels like a liquid chalk. Several coats of slurry are applied and allowed to dry before the sand is introduced. Dry sand is applied to the wet slurry and it sticks to the slurry. (I don't know about you, but in preschool, we made "sand" pictures, by running glue on paper in the figure that we wanted and then sprinkled sand on it and waited for the sand to dry. We then shook the excess sand off. Then we had are picture drawn out of sand. Same idea)

So you then wait for the sand to dry them apply another coat of slurry on it and repeat the process until 10-12 layers of sand are applied. This sand builds a shell around the wax pattern, which then will hold the liquid metal later on.

After the shell is made, the waxes needs to be melted out of it. It is flash fired to break the waxes away from the side walls and then is placed in an oven to melt the rest out.

Then the shell is fired in another oven at like 1000-1500 degrees F. This cures and hardens the shell. After that, metal is poured into the shell through a hole in the top and after all of that work, you have a casting.

and yes investment castings, have the best dimensional accuracy.

[Geoff]

I'm all for using new standard components. However, I'm not an engineer so that is beyond my abilities. What would we use for ring and pinion gears or an output shaft?

Whatever Hewland or similar companies offer as standard products! Of course there would be some components that you'd have to build to make it all adapt to the standard components. (4 chain primary drive, some shafts, etc.)

[Neil Jones]

Been doing a lot of reading. Purchased a book called: How To Rebuild And Modify Your Manual Transmission by Robert Bowen. Published by Motorbooks. ISBN-13:978-0-7603-2047-1.

Feel free to agree or disagree with the following statements taken from this book:

Page 20: "There are a few trade-offs with helical gears. The biggest downside is increased thrust or axial (along the length of the shaft) forces that put outward pressure on the bearings supporting the transmission shafts. Another is increased spreading (opposing radial) forces. As helical gears turn together, the input torque creates forces that attempt to spread the two gears apart, stressing the shafts, bearings and transmission case. In addition, the gear teeth create friction by sliding against one another, which creates heat and makes helical gears slightly less efficient than spur gears. All of these effects become worse with higher gear speeds, higher torque levels, and steeper gear-tooth helix angles."

Page 21: "Many race transmissions (though not all) use spur gears for a couple of reasons, including ease of of manufacture and a reduction in the thrust loads and frictional losses that come from helical gears. Straight-cut spur gears are extremely noisy, however and can be weaker than an equivalent helical gear."

[DrewP]

I agree with what he says on page 21, and most of what he says on page 20, but the way I learned it, a properly profiled helical cut gear will not have the faces sliding across each other any more than a properly profiled spur cut gear. That's the whole point behind the involute profiling geometry I mentioned earlier. Involute profiles are used on helical and spur gears.

When done perfectly the two mating teeth are supposed to roll across each other with no sliding. Any manufacturing tolerance or deformation in the assembly which takes the gear away from the as-designed mating geometry would compromise that, and I would bet that the additional axial and thrust forces in helical gears would compound that problem.

For moderate outputs it isn't a problem, but for high torque loading in a geartrain optimized for low weight and low rotating inertia it might become more of a problem.

And like he points out, the bearing requirements for the helical gears are greater than spur gears, thus why there are either tapered rollers or deep-groove ball bearings supporting the cluster and output shafts.

I have seen that book, and read through it at the book store, and I thought it was excellent. The photographs are phenomenal, full page color close-up shots, you guys should check it out if you haven't seen it yet, though it's a little general, and his detailed rebuild is for one of the Honda B-engine transmissions I think.

Best,
Drew

[Neil Jones]

Drew,

At this point I feel that to properly support higher power and torque outputs that the Saab 900 gearbox should use spur gears and not helical gears. This might very well prevent having to cast or machine a new case.

I'm looking into the possibility of making involute spur gears with a hob on used, older, manual gear hobing equipment. I still have a long way to go in my research on whether this will make sense or not.

I still need to do what you suggested and get my hands on a used 900 gearbox so I can tear it apart and see if larger shafts, bearings, bearing housings, etc. will be feasible and what might make sense to try.

[Neil Jones]

I agree. If you want to cast a new gearbox, why not design it so it fits standard components? You could design it around standard bearings, standard shift forks, standard gears etc. Standard meaning more plentiful than parts for a 16+ year old car...

There are tons of gear options for standard racing gearboxes.
http://www.pegasusautoracing.com/group.asp?GroupID=TRANSGEAR
And parts available too
http://www.hewland.com/
Plus you can find used gearsets for these as well (or complete gearboxes to part out to fit int the newly designed case).

The differential could even be a standard so there would be choices for limited slips and possibly ring and pinions.

Agree, strongly except I'm not sure a casting is the only way or even the best way to go for a very limited production / prototype gearbox case. Maybe a modular design that can be machined or perhaps a weldment?

[Crazyswede]

You could go wild and build a manual planetary gearbox. Think of the possibilities. I say manual only because all automatic gearboxes use planetary gearsets along with hydraulics or electronic solenoids to control the shifting.

[DrewP]

Drew,

At this point I feel that to properly support higher power and torque outputs that the Saab 900 gearbox should use spur gears and not helical gears. This might very well prevent having to cast or machine a new case.

I'm looking into the possibility of making involute spur gears with a hob on used, older, manual gear hobing equipment. I still have a long way to go in my research on whether this will make sense or not.

I still need to do what you suggested and get my hands on a used 900 gearbox so I can tear it apart and see if larger shafts, bearings, bearing housings, etc. will be feasible and what might make sense to try.

We'll have to see what we can do to break the work up, especially since it sounds like you might have access to some hobbing equipment, and people who know how to use it.

A good point about this being basically a prototyping expedition, at least until some actual data is gathered to see what actually needs to be done. A quick-n-dirty assembly could probably be made pretty quickly to try out internals and to see what can be done for all the packaging, especially if using off the shelf components.

Best,
Drew

[Jon1]

What are your thoughts on the modifications that Jorgen has done on the classic 900 gearbox?

Jon
jon1..

http://www.nysoc.com/gallery/album186

[Crazyswede]

What are your thoughts on the modifications that Jorgen has done on the classic 900 gearbox?

Jon
jon1..

I would say if thats a photo of what he did I don't think its a good idea to mix the engine and gear box oil together....and that whole letting the oil go out the side of the case won't be much of a gain in the long run.

[airsweden]

Hi Jon, I was looking at the pics of the latest project car, what are you doing with that thing......... racecar?

We gonna see you iceracing at lake george this year?

Josh Teresco

[jerrit]

"Or if someone has $ they could have Hewland develop a complete gearbox for the C900/99 :-)"

Seriously, if more than a few people are interested in this then why not go for it and distribute the development cost on top of the unit price of the first batch ordered? Like a group buy. As long as the cost isn't TOO high (definitions of this vary widely, I know), I'd be interested in this route.

Hewland's site says they discourage developing for street use, but I think that's mostly a disclaimer. There's also Quaife; they make rebuild kits for the F35 that's in the viggen, as well as whole replacements for a lot of other cars, and I'm pretty sure they do custom work. Anyone know of any other companies to try?

[Geoff]

Sellholm, they made the diff that Jorgen sells http://www.sellholm.se/ Pratar du svenska?

My guess is that if it could have been done with 10 people putting in the funds, the Swedes might have done it already... But you don't know until you try.

Another idea would be to buy a standard RWD racing transaxle like this:

And build a custom primary drive or adapt a stock one. Then you'd weld a housing around it to adapt to the engine and act as a sump. However that box is $32,000. There are other cheaper ones (around $5500) but since the differential is so far forward you'd have to re-engineer the car to have the wheels closer to the front of the car, or stick the gearbox out past the steering rack and protrude into the passenger compartment. Neither of which would be bad for re-distributing weight balance in a 99/900. The cheaper ones are for less powerful cars too. At that point is it REALLY worth the effort and costs??

Jon, I think Jorgen's stuff is really good but we haven't seen too many of his boxes here in the US so maybe people don't really know how strong the gearboxes he makes can be. He makes them for all applications (and all costs).