How much of the vehicle's unibody structure and suspension are you willing to change?

Anything can be made to fit anything else with enough cutting, fabrication, and welding. This certainly is not going to be a bolt-on job. You will have to change everything.

Is it a waste of time? Most hobbies are...

It will be better to sell this vehicle and buy a different one that already has your preferred powertrain layout as factory equipment. Audi (real one, not VW platform with Audi badge) and Subaru come to mind.

when it comes to unibody im not worried because a lot of ppl already build rwd swaps you can even buy kit for that
there is enough space to run shaft i just need to cut some metal on the rear so i can mount differential the only issue would be the exhaust but i plan to run it trough fender or bonnet

when it comes to mounting the engine that pretty much is solved too

i came here to discuss using fwd gearbox as central differential with axles coming out as shafts that attach to front and rear differential

there are awd builds of the same chassis using as i mentioned awd gearboxes that came in SUVs but these gearboxes dont use central differential what will cause a lot of stress on gearbox when turning
i wouldnt worry about that if this build was meant for offroad but its meant for tarmac

i know race cars need servicing often but i just to not want an obvious failure point

Two problems you're going to have:

-packaging an additional differential in the front. Differentials are not small, and especially the type of differential you would need to use here (not an oem design that's integrated into the front powertrain already). Don't forget that you don't have to just package the front diff itself; you also have to package enough structure for the diff, rotated engine, and original transmission to live in very close proximity and be mounted strongly enough to handle all of their various loads. If this is a Macpherson strut car in the front (which I'm assuming it is... ) you can't just cut out whatever you want. Inner fenders are a structural component in a macpherson car.

-You'll need to sort out the existing differential in the current transmission. You're effectively turning the existing front diff into an AWD center diff; what works in the front is not going to work well in the center. What options you have will depend heavily on what specific transmission you're dealing with.

I think you'd be much better off adapting an existing AWD transmission system designed for front transverse mounting (ie Mitsubishi) to your existing engine. It'll be a lot less work and you'll be using components which are designed to be mounted the way you want to mount them.

It perhaps might help to know which vehicle the original poster is talking about here.

Another nuisance is that the final drive ratio of the transverse transmission is built into the transmission itself. Can you get a hypoid-gear-and-differential assembly with a 1:1 (or close) final drive ratio? It has to be the same ratio front and rear, also.

I have run across longitudinal-rear-drive conversions of front-drive bodyshells before. Invariably they are using a rear-drive transmission meant for longitudinal installation and without its own final drive ratio contained inside it, so that the rear differential can have a final drive ratio that is in the conventional range (3:1 - 4:1 or thereabouts).

There are transmissions for transverse installation which include a "center differential". There is a version of the ZF 9HP automatic transmission used in the Jeep Cherokee that has such a thing. There may be others.

Various random thoughts, most of which arise from not knowing exactly which vehicle the original poster is proposing to convert.

Most front-drive (transverse) transaxles have the diff as close as possible to the center of the vehicle. Turned longitudinally, that puts the diff pretty far forward. Where's your front-axle centerline relative to what would normally be the right-hand differential output flange?

Pay attention to where the steering rack is, and whether it's going to interfere with your plans.

Ouch! - those driveshaft angles.

je suis charlie

The absence of a center diff is going to make the handling rather variable.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Wait a couple of years and just use two electric drive assemblies. No need for a front-to-rear drive train connection.

Quote (BrianPetersen)

It perhaps might help to know which vehicle the original poster is talking about here.

Civic ej8 1996 Coupe

Quote (BrianPetersen)

Most front-drive (transverse) transaxles have the diff as close as possible to the center of the vehicle. Turned longitudinally, that puts the diff pretty far forward. Where's your front-axle centerline relative to what would normally be the right-hand differential output flange?

model I posted is very close to reality
the right axle is much shorter than left one
what means that center diff ends up just before firewall

Quote (BrianPetersen)

Pay attention to where the steering rack is, and whether it's going to interfere with your plans.

Right, I will adjust the design

Quote (gruntguru)

Ouch! - those driveshaft angles.

I know...

I am not 100% sure, but I think that generation is the last one before Honda started using a full-perimeter subframe around the whole drivetrain. It has what looks like a "subframe" for the lower control arms and the steering rack. Minor digging reveals that "subframe removal" does not entail engine or transmission removal, which suggests that it's a steering-rack and lower-control-arm subframe but is not involved in engine or transmission mounts.

I suspect that your proposed driveshaft to the rear is going to be going right through that area.

What differential (with the aforementioned approximately 1:1 final drive ratio) are you proposing to use?

And how are you going to cope with the random front/rear torque split?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Well, with the open diff in that transmission, and with the same final-drive axle ratio in both the front and rear (which had better be the case to avoid burning up the diff in the transmission), it's going to be locked at 50% front and 50% rear torque split, whether that's what's needed or not. It will act as any other open diff does. If the rear wheels are on ice, the front wheels won't deliver any torque. (And vice-versa.)

There's reasons that normal rear drive units on transverse-powertrain vehicles use a viscous clutch or a controlled clutch of some sort rather than an open center diff. Those are the ones that have variable front-rear torque split. (Normally 100% front 0% rear unless there is slippage of the front tires, and some systems can achieve almost 100% rear torque bias by locking that center clutch so that the rear powertrain is driven no matter if the front wheels are completely slipping or off the ground)

So it has a center diff?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

From what I can tell - He's wanting to use the normal differential that's built into the (normally front-drive, transverse) transmission as the center diff, and then use the now-forward-pointing halfshaft output to drive the front diff and the now-rearward-pointing halfshaft output to drive the prop-shaft to the rear and the rear diff. So the original-equipment final-drive diff that's built into the transmission will now be the center diff.

Quote (GregLocock)

And how are you going to cope with the random front/rear torque split?

There are "fake LSD" that are made for these gearboxes

its made out of two plates that have springs pushing them againts spider gears - preload
locking pin that goes trough differential is also being pushed against cut out in the plate that is also puting more force on spider gear locking it more when loaded - locking angle
so by changing the angle of the cut for one of the plates I can actually change torque distribution
I could even change torque distribution for breaking
essentially it is horribly simplified LSD

its not a good design by any means for FWD car that often runs axles at different speeds but in central diff where we want pretty consistent speeds - I think it will hold up
I also redesigned it a bit so there will be a small clutch disc between the plates and spider gears
it also includes changing a metal washer under the spider gear for something with a bit more grip - clutch pack
I talked with some offroad guys and a heard that a lot of them used these "fake LSD's" in their AWD hondas
of course they used it in front and rear diff because honda design is missing center differential but I got a proof that it is kinda working

because the front differential will be almost directly connected to center diff the torque will be split a bit more to the front - I can cope with that

Quote (BrianPetersen)

From what I can tell - He's wanting to use the normal differential that's built into the (normally front-drive, transverse) transmission as the center diff, and then use the now-forward-pointing halfshaft output to drive the front diff and the now-rearward-pointing halfshaft output to drive the prop-shaft to the rear and the rear diff. So the original-equipment final-drive diff that's built into the transmission will now be the center diff.

yes, exactly what I'm talking about

Quote (BrianPetersen)

What differential (with the aforementioned approximately 1:1 final drive ratio) are you proposing to use?

I will probably go for a shorter gearbox, and find diff that will fit
I know ppl have used gearboxes from 1.3 1.4 and 1.5 D series engines
gears are much shorter in some of them
i totally forgot about it and didnt though of it until couple days ago so I did not yet prepared the exact parts I would fit

Quote (BrianPetersen)

There's reasons that normal rear drive units on transverse-powertrain vehicles use a viscous clutch or a controlled clutch of some sort rather than an open center diff.

in case of honda trasverse awd gearbox fron wheels are powered all the time and controlled clutch only engages rear differential so by this design most we can get is probably 45% rear torque

You can get 100% rear torque split if the front wheels are on ice and the clutch to the rear is up to the task.

...which most of them aren't. Look at how skinny the rear halfshafts are and how small the rear diff is on a lot of them. It's just enough for the manufacturer to claim it has all wheel drive.

... but this is not a limitation of the layout, only of the components selected.

Quote (Rokita)

I will probably go for a shorter gearbox, and find diff that will fit
I know ppl have used gearboxes from 1.3 1.4 and 1.5 D series engines
gears are much shorter in some of them
i totally forgot about it and didnt though of it until couple days ago so I did not yet prepared the exact parts I would fit

He's talking about your front and rear differentials - not the trans.

The final reduction in the FWD gear box is built into the gearbox, meaning in your setup what is normally the final reduction is BEFORE the split to front and rear; if you use a relatively standard reduction ratio in your front and rear differentials, your total gearing is going to be EXTREMELY short. For the gearing to be remotely sensible, if you use the same trans, you need the reduction at your front and rear diffs to be as close to 1:1 as possible.

Quote:

He's talking about your front and rear differentials - not the trans.

transmission gearing actually affects the speed

rn my final ratio is 3.8 i can find 3.2 gearbox and fit longest gears aviable in it
then even with keeping 3.2 final ratio on differentials it wont shorten the gearing that much
engine will also rev much higher than its stock
so gearing will be fine probably
i didnt calculate that exactly

I think you had better calculate your overall drive ratios (and figure out what RPM it's going at what road speed in each gear) based upon available components.

I think you are in for a rude awakening unless you find appropriate 90-degree final drive and diff assemblies with extremely tall (ideally 1:1) final drive ratios.

Quote (Rokita)

engine will also rev much higher than its stock
so gearing will be fine probably
i didnt calculate that exactly

Unless this is a tractor pulling rig where you need 10,000,000 lb ft at the wheels, or 'much higher than stock' means 20,000 RPM, you have a lot of important work to do.

The gearing will not be fine unless you calculate what your true ratios are and plan accordingly. I can assure you, as setup with back-to-back 3.2:1 reductions is not going to work.

Spot check. It appears that the rear axle ratio in a Honda CRV (don't know/care which generation) is 2.529:1. This is not the same as the final drive ratio in the transverse front-drive transmission - because the drive to the rear is taken from another set of gears driven from the front differential housing after the front transmission's final drive, with (presumably) a 2.529:1 speed-up ratio built into the front transmission to correspond to the 2.529:1 reduction ratio in the rear axls, so that (nominally) all four wheels are driven at the same speed (not counting the intentional slippage in the clutch that is actually used to supply torque through the rear drivetrain). This set-up is very typical of AWD systems that are based on transverse front-drive powertrains.

Better choice ... Seems that there is a VW Haldex rear differential assembly with a 27/17 tooth count, 1.588:1 reduction ratio. That is starting to get within the realm of plausibility when combined with taller gearing inside the transmission. It's still going to end up way shorter than stock but it's within reach of something that "could work".

Has the use and speeds been specified? Tire sizes? Off-road guys do what is being proposed, I have a neighbor who found that 4.xx differential ratios still didn't give enough reduction.

Quote (BrianPetersen)

I think you are in for a rude awakening unless you find appropriate 90-degree final drive and diff assemblies with extremely tall (ideally 1:1) final drive ratios.

For reference - by my math in whatever transmission gear is 1:1 (usually 4th in a 5 speed trans) with a 23.5 inch tall tire, back-to-back 3.2:1 ratio gives a wheel speed of 108 mph at an engine speed of 10,000 RPM. 10,000 RPM in 1st gear using a sensible 1st gear ratio is around 22 mph.

Calculate your ratios OP before you go too far down this rabbit hole.

Quote (LionelHutz)

Has the use and speeds been specified? Tire sizes? Off-road guys do what is being proposed, I have a neighbor who found that 4.xx differential ratios still didn't give enough reduction.

per OP this is a Honda Civic, not a unimog.

I calculated everything and it looks like that

speed at 6500rpm on 1st gear is 46km/h
with longer ratios, crv differential + bigger wheels it would drop to 30km/h

6500rpm on 5th gear is ~200km/h
after modifications it would be 120km/h

since i want to increase redline of my engine from 7800 to 11000 it would be
5th 206km/h at 11000rpm

not ideal but i would for sure have A LOT of torque

i personally had no luck in finding shorter final drive differentials
haldex is a bit too big to fit in my design
but with 1.588 final drive i would actually go faster than fwd setup
5th 213km/h at 6500rpm

after talk with u
If I decide to go AWD I will keep it transverse by going the shorter way of swaping CRV power train

but talk is yet not over

now I need help in understanding the AWD powertrain of CRV

what will exactly happen when I try to use the car with AWD turned on? does the lack of center differential cause damage ? how can I work around it

the thing is: is it safe to drive fast with awd locked on and take corners?
is there a chance that when turning difference in speed of axles may cause over/under steer

from my basic under standing: the avg radius that rear wheels follow when turning is shorter than front wheels
is that right?

so then I guess the rear wheels will try to make a burnout when taking a corner what will for sure cause problems in grip
Am I right?

what should I do to make it work
using the CRV AWD module is not possible because its integrated into the ECU
and im either going to keep my chipped ECU or go for SPEEDUINO
which neither one of them can control AWD

You need to find out what's inside the magic clutch that engages and disengages (more properly, modulates) the rear drivetrain, and probably develop your own control strategy for it.

Odds are, there's a pulse-width-modulated (PWM) control solenoid in there somewhere.

Or ... You transplant the entire vehicle electronics and powertrain module and wiring harness etc of the CRV into the Civic such that it has no knowledge that it's inside a bodyshell of a different shape.

OK so i digged up a bit and there is a solution welding up the clutches in rear diff then installing a drive shaft from land rover with viscous clutch

there are no numbers but its assumed it splits torque 70% to front

So you want to have the rear wheels locked to each other even though they're not driving most of the time? That's sure to have some handling characteristics that are ... interesting.

What, exactly, do you propose to be doing with this vehicle?

Going fast and turning corners????

Why not take the FWD drivetrain out of the front and install it in the rear? Adding weight to the front of a vehicle rarely helps it turn corners.

If the objective is to go fast and turn corners in something that has a Honda badge on it, the appropriate course of action is to sell the existing car (as-is) and buy a current-generation Civic Type R ... accepting that even this does not have all-wheel-drive.

If the objective is to get up a steep driveway in winter without having to shovel snow first and with something that has a Honda badge on it, the appropriate course of action is to sell the existing car (as-is) and buy a CRV or whatever the all-wheel-drive version of the Fit/Jazz is, which already has all-wheel-drive built in from the factory and comes complete with warranty.

If you can forgo the Honda badge, and you want to go fast and turn corners in something that has all-wheel-drive, buy a Subaru WRX. Or a VW Golf R.

Quote (BrianPetersen)

If the objective is to get up a steep driveway in winter without having to shovel snow first and with something that has a Honda badge on it . . .

. . . reverse it up.

je suis charlie

True but the point is that I don't understand the point of doing a potentially failure-prone home-grown butcher job that is (to be brutally honest) unlikely to be successful, when you can buy something that does what you want it to do off the shelf with a warranty.

Original poster needs to explain this to me.

While my opinion is that OP is in way over his head here.. you guys don't sound like much fun. Never had a project car? Never tried to build something just to see if you could do it?

Grins, yes, here's one of mine.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Quote:

per OP this is a Honda Civic, not a unimog.

You've never seen a 4x4 Civic before?

I "get" project vehicles, I have one, a 1990 Yamaha FZR400 that I completely rebuilt and refreshed from ground up more than 8 years ago and spent more than the bike was worth doing so. But, I still have it, and I'm still riding it, and I've put 25,000 km on it since, and it runs well, and I get people asking about it whenever I am out and about with it. Works for me.

I am encouraging the original poster - who seems to have disappeared! - to set some objectives, targets, design criteria. Then evaluate feasibility and alternatives, and make sure to understand most of the hurdles that will need to be overcome ... with the understanding that you'll never predict all of them, so expect the unexpected. It is an engineering project.

Quote (LionelHutz)

You've never seen a 4x4 Civic before?

Remember the "Wagovan", a.k.a. Honda Civic Shuttle?

They actually had two generations corresponding to Civic generations 3 and 4. Both were primarily front-drive with (I think) a viscous-coupling drive to the rear. First generation used a live rear axle. Second one used IRS with the rear diff mounted to the rear subframe. I think the mechanical layout of the
second-generation one (Civic generation 4), is what eventually morphed into the first-generation CR-V.

Until very recently, the mainstream models of the Civic have never been designed to accommodate all-wheel-drive. I think the rear subframe of the new one will allow it, although they still haven't done it.

Expecting to make the engine turn 11,000rpm makes me suspect it'll never get built, at least no how the OP is questioning here.

Honda K-series builds are hitting the 11k RPM mark without all that much drama. Ain't cheap though.