Rear Transaxle Se7ens

[DLW]

Looking for advice on building a Sevenesque sports car with a rear transaxle and front engine but traditional esthetics... potential donor cars are Porsche 924/944, and Alfa Romeo Alfetta, Milano and GTV-6.

[s2k7]

Honda s2000, WRX, Cosworth sierra oh! 240Z with R200 diff and RX7 will do too.

[slngsht]

I think he is looking for a rear transaxle... i.e. tranny in the back. no?

[DLW]

I think he is looking for a rear transaxle... i.e. tranny in the back. no?

 

Yes, sir... you're right.

[slngsht]

how much power are you planning to pass to the wheels?

 

vette transaxles are pretty big. there are several porsche choices depending on power level that might work- from 924 to flipped 911 transaxles. Then there are several configurations of corvair/pontiac tempest variants. Those things are pretty light if I recall correctly.

 

it opens up alot of room in the footwell area, but you have to consider that hip area is in short supply in a seven too, and that's where the transaxle will live because you will sit pretty far back in the car. For most of us, the tunnel is barely big enough for the driveshaft near the back.

[DLW]

About 200 horsepower, naturally aspirated. The Corvair transaxles would be ideal, since parts are easier to come by than for Alfa transaxles and are fairly inexpensive, and there are junkyards with lots of Corvairs near where I live. They are also fairly compact IIRC, and with my dimensions (6' and 250 pounds), I need all the hip space I can get.

 

If this makes any difference in the construction and dimensions, in the event that I build such a car, I plan to pattern the spaceframe after McSorley's 4-4-2 for size reasons.

[supersportsp]

I guess my first question would be why?

 

One of the primary reasons a transaxle is utilized is to help with weight distribution and a car where much of the weight would otherwise be up front.

 

In a Se7en, your butt is weight distribution as you sit right in front of the rear axle. So, you are adding more weight to the rear of the car which, judging by your weight is not needed.

 

I would think packaging would also be a challenge as you would need to totally re-engineer any current Se7en based frame in order to do this or to start from scratch.

[WestTexasS2K]

Elite makes a nice compact sequential transaxle that would fit the bill. http://www.eliteracingtransmissions.com/category.php?c=2

[JohnCh]

The Westfield FW400 used a Hewland transaxle, so it has been done before.

 

I agree about the packaging issues, but I wouldn't mind a heavier rear balance in my car. Good for traction, good for braking.

 

-John

[supersportsp]

The Westfield FW400 used a Hewland transaxle, so it has been done before.

-John

 

Right John is. I just dug up my FW400 brochure I got when they were new. It uses a Hewland LD200 5-speed unit. I even have the gear ratios if anyone is interested.

 

http://www.jandjracingltd.com/images/Hewland/LD200PartsLocator/LD200Pic.jpg

[slngsht]

wonder how much one of those costs... he was looking for something reasonably affordable.

[escondidoron]

Here's a few thoughts to consider:

 

1) The hip room issue can be resolved with a gearbox that is behind the differential and not in front of it. The Porsche gearbox is such a design. 1973 and later 914s have the shift linkage interface facing towards the front of the car too.

 

2) You also could look at the Audi transaxles. Unlike most front wheel drive vehicles, which have transverse engine/gearbox assemblies, many of the Audis have north / south engine configurations. They also have some that are 6-speed units.

 

3) Make sure that you take into account the direction of rotation of the engine relative to the differential assy in the transaxle that you use.

 

4) Corvair gearboxes were based on standard Chevy units for cost reasons. Corvair engines were reverse rotation as a result.

[slngsht]

Here's a few thoughts to consider:

 

1) The hip room issue can be resolved with a gearbox that is behind the differential and not in front of it. The Porsche gearbox is such a design. 1973 and later 914s have the shift linkage interface facing towards the front of the car too.

 

2) You also could look at the Audi transaxles. Unlike most front wheel drive vehicles, which have transverse engine/gearbox assemblies, many of the Audis have north / south engine configurations. They also have some that are 6-speed units.

 

3) Make sure that you take into account the direction of rotation of the engine relative to the differential assy in the transaxle that you use.

 

4) Corvair gearboxes were based on standard Chevy units for cost reasons. Corvair engines were reverse rotation as a result.

 

I'm not sure having a rear mounted transaxle is an improvement. Yes, it does transfer more weight to the rear, but I think it increases the rotational inertia of the car (can't think of the term... I mean resistance to change in yaw). You want the majority of the mass as close to the center of mass as possible.

pushing the engine back, then tranny, then axle accomplishes that - or a longitudinal mid engine design.

Even with the big lump in my car, I still have rear weight bias.

 

There is alot of material out there on the corvair design... and 2 kits to change the corvair to a mid engine configuration with normal rotation. Also look at the tempest configuration... same basic components reconfigured for a front engined car with correct engine rotation.

[DLW]

Good advice... the Porsche and Audi transaxles are definitely worth looking at, and a Pontiac Tempest-derived drivetrain would be interesting, especially with the slant-four. Does anyone know anything about Renault and Saab 900 transaxles? Those are pretty common around junkyards, but I would be concerned about parts availability and durability.

 

The ideas behind the rear transaxle were mainly about increasing traction to the rear and the fact that few people have a Se7en with a rear transaxle.

[scannon]

What are you looking for in a front/rear weight ratio?

 

My Caterham has 53% on the rear wheels before I get into the car. I'm far from an expert but I don't think I would like it any heavier on the rear, especially if the weight were added at the extreme rear of the car.

 

Other than if you wanted to make a drag racer out of your car I can see no advantage to increasing the rearward weight bias.

[supersportsp]

 

Other than if you wanted to make a drag racer out of your car I can see no advantage to increasing the rearward weight bias.

 

This is what I was thinking as well. I don't have the numbers in front of me, but my car was also biased towards the rear without me in it. Add in my 200 lbs and I am adding over 14% of the cars weight to right in front of the rear axle.

 

The other thing to consider is high speed stability. I am not familiar with any standard Se7en design that doesn't have lift on the front of the car. Moving more weight to the rear is only going to make the car even lighter on the front end at higher speeds. Not exactly a desirable trait on these cars.

[JohnCh]

My car came in at 50.0/50.0 F/R with half a tank of fuel and no driver (no spare on the rear either). With 190lb of ballast in the driver's seat it came out at 46.9/53.1.

 

-John

[escondidoron]

slngsht said:

(can't think of the term... I mean resistance to change in yaw)

 

The correct name for the physical property being discussed here is polar moment of inertia and you described it correctly. It is the resistance to change in direction. High polar moment means a high resistance to directional change. If an object is going straingt it wants to continue going straight. If it is turning it wants to continue turning.

 

On F/R weight distribution, the idea that 50/50 is ideal is possibly not correct. I am basing this assumption on a quick survey of high performance road & racing cars. At the highest level, most purpose built racing cars and high end street cars are mid-engined (to reduce 1, weight -- i.e. minimal driveline mass, and 2, to minimize polar moment. Most of these cars are closer to a F/R weight bias of 40/60. A case in point are Lotus Esprits. A pretty decent handling car. My '84s weigh in at about 2650 lbs with about 1075lbs on the front wheels and about 1550lbs on the rear. That is 41/59 F/R. I suspect that these figures are pretty typical for weight bias for mid-engined racing cars. My '77 S1 Esprit is about 150 lbs lighter but the F/R weight bias is almost identical.

 

As for how polar moment would be affected by placing high mass components at opposite ends of the vehicle, you could do the math, or you could just consider a quick example: Imagine holding up a 16lb bowling ball, with your hand raised up above your arm, and rotating it clockwise and then quickly reversing direction to counterclockwise.

 

Now imagine performing the same activity, but with the same 16lbs in the form of a barbell with two 8lbs weights, each 8lb mass at opposite ends of a weightless pipe.

 

In this imaginary experiment it is easier to reverse rotational direction with the bowling ball because of its lower polar moment of inertia.

 

A number of valid points have been raised in different responses in this thread. However, out in the real world, while polar moment is a real design concern, factors such as suspension design, steering geometry, tire compound, physical packaging constraints and weight distribution all play significant parts in how the car will respond to a driver's inputs (i.e., how it will handle).

 

With careful selection of components and thoughtful design it is possible that the polar moment of a transaxle Seven type vehicle would not change significantly from a conventional in-line transmission equipped car of otherwise identical design.

 

For instance, if you built a "conventional" layout Seven car using live axle rear suspension with a cast iron differential and a cast iron transmission housing, or an IRS system, again with a cast iron differential and compared either of those designs polar moment and overall mass to that of a car designed with an aluminum transaxle you might find very little difference in polar moment but a large difference in overall weight and / or weight distribution.

 

At the end of the day, how you deal with all of these other issues (suspension design, steering geometry, tire compound and physical packaging constraints) will likely have a greater influence on handling than pure polar moment of inertia and (within reasonable limits) weight distribution.

 

Just my $0.02.

[slngsht]

yes. polar moment is it. losing my noggin. let me go find that BSME degree that's in the basement somewhere :rofl:

[s2k7]

"Just my $0.02."

 

Thanks professor, I appreciate the case study. Very informative and technical.

Worth more than $0.02 cents.