Board logo

Subaru Powered Midy/Thoughts?
rontyler - 20/11/07 at 10:25 PM

Greetings,

Been a while since I last posted... 'stuff' gets in the way sometimes.

Anyhow, been seriously considering getting back to the scratchbuilt midy project, from a new perspective. That is to say, a different powertrain. Yup, you guessed it, Subaru in a 2WD config... most likely the EJ22.

Before I get all crossed up and make the wrong choice, I'd like to hear opinions on said powertrain. Pro's and con's if you will.

I expected width to be an issue, but my preliminary drawings suggest its a realistic hurdle. So lets ignore that for now.

Any thoughts?


tegwin - 20/11/07 at 10:27 PM

Its interesting..

I was thinking just this the other day....But with the latest turbo jobby...

I guess you might be able to use a porsche gearbox if you turn it upsideown...

The engines are not actually 'that' wide...still a problem


rontyler - 20/11/07 at 10:44 PM

quote:
Originally posted by tegwin
I was thinking just this the other day....But with the latest turbo jobby...


The second 'stage' would probably see a pair of turbo's. Room needs to be set aside for them, but I'm leaning towards keeping it simple, initially.

quote:
Originally posted by tegwin
I guess you might be able to use a porsche gearbox if you turn it upsideown...


Any reason not to use the stock gearbox?


meany - 20/11/07 at 10:57 PM

you could use a 2dw box from the old front wheel drive models, not sure how strong the internals are on those.

you could however, make it 2 wd, basically bay welding up the centre diff.

see below.( itsays for rwd but i expect it will still work...maybe)
http://www.clubwrx.net/forums/showthread.php?t=27532&pp=20

The impreza Track is not as wide as you think, not as wide as a sierra.

i have a set up lying in the way in my garage.


meany - 20/11/07 at 10:59 PM

plus..if i recall, someone is actually doing a middy Subaru project.


caber - 20/11/07 at 11:07 PM

I have 3 friends, weel 2 and a brother with scoobies at tyhe moment, I am just waiting until one of them get written off so I can buy the salvage as a donor. I was thinking similar put the engine and box at the back possibly even use the suspension or at least the uprights.

I do like the idea of something like a 1960s le mans or can am car based around a space frame and some plastic panels!

It should be possible to come up with some reasonable geometry for the suspension though I can't get through the maths!
Caber


rontyler - 20/11/07 at 11:15 PM

quote:
Originally posted by meany

you could however, make it 2 wd, basically bay welding up the centre diff.
see below.( itsays for rwd but i expect it will still work...maybe)
http://www.clubwrx.net/forums/showthread.php?t=27532&pp=20


Thanks for the link.

My wife's daily driver is a '95 Legacy... it doesn't seem as though it has a center diff. as it chatters a bit in tight turns.
I haven't researched that aspect... yet.


quote:
Originally posted by meany
The impreza Track is not as wide as you think, not as wide as a sierra.



Fairly wide wheels will address this to some degree. Aside from that, I'm not entirely opposed to a little narrower track in the rear.

quote:
Originally posted by meany
plus..if i recall, someone is actually doing a middy Subaru project.


I did a little searching before hand and the closest thing I found was the "Blast". If you can point me to another, I'd be grateful.

Thanks!


meany - 20/11/07 at 11:19 PM

if anyone wants any measurements, just ask. Im in my 5th year of scoob ownership.

the pics below are from a 96WRX.
and a legacy gearbox...same length.
its one long MoFo..lol.

http://s8.photobucket.com/albums/a33/meanandgreen/scooby%20bits%20for%20sale/


kb58 - 20/11/07 at 11:24 PM

Here in California, Subaru STi drivetrains are very expensive, around US$10k, which is nuts.


rontyler - 21/11/07 at 12:10 AM

quote:
Originally posted by kb58
Here in California, Subaru STi drivetrains are very expensive, around US$10k, which is nuts.


Kurt,

Thats what steers me to the EJ22. Around here, Pick N Pull's sell complete, running engines for $160 and tranny's for $100. They're only guaranteed as cores, but out of the last 7 I looked at, I'd bet 2 of them were perfectly useable... at least for kicking off a project such as this.

P.S. I looked into the center diff 'issue' and apparently the '95 does, in fact, have one... and it so happens that they occasionally 'lock-up'. Must be the case with my wife's car <shrug>.

[Edited on 21/11/07 by rontyler]

[Edited on 21/11/07 by rontyler]


gator - 21/11/07 at 03:34 AM

If I can get this thing to work, here are a couple of photos of 4WD and 2WD boxes, the 2Wd is a bit shorter (and lighter) but the rear mounts look to be in the same place. One has cable clutch (2wd), the other hydraulic. Regards, Alan. Rescued attachment 100_1334.jpg
Rescued attachment 100_1334.jpg


gator - 21/11/07 at 03:38 AM

And another Rescued attachment 100_1332.jpg
Rescued attachment 100_1332.jpg


gator - 21/11/07 at 03:42 AM

And one more! Rescued attachment 100_1324.jpg
Rescued attachment 100_1324.jpg


CGILL - 21/11/07 at 05:35 AM

http://www.constructorscarclub.org.nz/profiles/profile-brucetur.html

http://www.saker.co.nz


rpmagazine - 21/11/07 at 09:01 AM

early 2wd and 4wd boxes are weak. Late model are good.


RazMan - 21/11/07 at 09:24 AM

After seeing the Murtaya at the various shows this year I was really impressed with the Scooby engine - really compact and with a low CofG. Power potential is good too - 400bhp+ is achievable without too much investment.
The Saker used the Legacy engine and box didn't it?


Delinquent - 21/11/07 at 09:41 AM

we were discussing using a boxer type engine on Pistonheads a couple of weeks back - after I started thinking about using the entire drivetrain from a Boxster. By my reckoning the advantages in the lowered C of G and compactness of the engine make it a very worthwhile avenue.

a 911 turbo engine has been used in an Ultima, so width doesn't appear to be an issue - although triangulation of the chassis would appear more difficult to start with, when you consider how shallow the whole engine can be you can carry out substantial triangulation above it, just make it bolt in so the engine has an easy exit path.


rontyler - 21/11/07 at 09:47 PM

Gator,
Thanks for the pics. What are the tranny designations for those, or what cars did they come from?

quote:
Originally posted by rpmagazine
early 2wd and 4wd boxes are weak. Late model are good.


Define late/early? Is the weakness of the early trannys something thats addressable?

quote:
Originally posted by CGILL
http://www.constructorscarclub.org.nz/profiles/profile-brucetur.html

http://www.saker.co.nz




Thanks!

quote:
Originally posted by Delinquent
a 911 turbo engine has been used in an Ultima, so width doesn't appear to be an issue - although triangulation of the chassis would appear more difficult to start with, when you consider how shallow the whole engine can be you can carry out substantial triangulation above it, just make it bolt in so the engine has an easy exit path.


Precisely! My early drawings lead me to believe its not unreasonable...








It puts the whole powertrain below the tops of the tires.

That drawing is +/- an inch so its worthless for anything other than pondering. Clearly, it would need a few tweeks to work. And, like you say, the top would need to be removable.


gator - 22/11/07 at 03:39 AM

Hi there, the 4WD engine/box I have is from an import half cut Liberty EJ20T (single turbo) ref TY752VB1AA. The 2WD box ref TM752RH1AA was sourced separately. I assume it to be a Liberty prior to MY98 as Scoobies seem to be totally 4WD from about then. Regards, Alan.


CGILL - 22/11/07 at 06:52 AM

Re COG of 'baru engine... don't be fooled, the crank location is much higher in the block than an I4, the centerline of the crank is also the centerline of the main shaft of the box, the diff output is much lower than this, so for a clubman type car, either the engine needs to be located high, or the axles run at large angles, most of the weight advantage is from an all aluminium block and heads, although the EJ is a hell of a lot heaver than the early EA series engine. Honda B16 vtec engine and box are reasonably lighter
Saker use FWD subaru box, some use 4wd with the guts chopped out and the tail capped, they are stronger. The hole in the center of the chassis for the wide engine behind the wide driver cabin would need bolt in stress members I would say that can be removed for engine removal, the sound would be awesome


kb58 - 22/11/07 at 04:42 PM

I don't know about the sound... The U.S. Subaru STi sounds just like a six-cylinder that has two spark plug wires pulled off. At least it sounds that way to me.


kikiturbo - 22/11/07 at 06:56 PM

it depends on the exhaust manifold... a normal impreza wrx or STI usually runs a asymetric unequal length exhaust manifold which is not that good for power but has a distinct burbly sound... a more powerfull japanese spec C model (used asa a basis for group N racing) has a symetric equal length manifold that sounds really normal by comparison..


rontyler - 28/11/07 at 02:19 AM

quote:
Originally posted by CGILL
Re COG of 'baru engine... don't be fooled, the crank location is much higher in the block than an I4, the centerline of the crank is also the centerline of the main shaft of the box, the diff output is much lower than this, so for a clubman type car, either the engine needs to be located high, or the axles run at large angles,



Agreed. However, its 'profile' is still quite low... top of the is case several inches below tops of tires. Not very empirical I digress, but anecdotal perhaps?. Most of the transverse fours and sixes can only dream such a claim ;-). Maybe a better way to state my view is 'packaging'. At the rudimentary level I'm at, it seems... um... better <cring>.

This example is highly generic, but it illustrates my point...

Subaru package (axles would intersect about 1 1/2" lower at tranny)...



SHO V6 with red line indicating tops of same diameter tires...



quote:
Originally posted by CGILL
most of the weight advantage is from an all aluminium block and heads, although the EJ is a hell of a lot heaver than the early EA series engine. Honda B16 vtec engine and box are reasonably lighter



The EJ's tranny looks like a behemoth. Can't even venture a guess at this point, so I'll concede on that note. But... the engine itself is pretty paltry in my experience, tripping the scales under 250lbs complete....

http://forums.hybridz.org/showpost.php?p=802813&postcount=3


Anyone have weight data on the EJ tranny, AWD and/or 2WD?


rpmagazine - 28/11/07 at 12:24 PM

The 6G75 3.8lt I am using weighs 275kg complete with transaxle, so it is no lightweight. The height of engine relative to driveshaft is partly set by approach/GC issues in FWD cars. However the transverse engines have an important asset in that the engine and transaxle can be rotated to lower the engine or the bell housing can be modified to achieve the same outcome. For me with the heavy V6 this means that the crankshaft can (with a lot of work) sit 40-50mm below that I could achieve with the Subie. However I have no idea of relative CoG heights overall.


kreb - 28/11/07 at 03:03 PM

Here's a rather cool Subbie-powered middie:

http://www.blastautomotive.com/


rontyler - 29/11/07 at 03:14 AM

quote:
Originally posted by rpmagazine
However the transverse engines have an important asset in that the engine and transaxle can be rotated to lower the engine or the bell housing can be modified to achieve the same outcome.



Great point. As long as the lubrication is attended to... its an interesting thought... hmmmmmm....


kb58 - 29/11/07 at 04:18 AM

quote:
Originally posted by kreb
Here's a rather cool Subbie-powered middie:

http://www.blastautomotive.com/

Kind of a half-baked site. When will people learn about business, "Best foot forward!"


kreb - 29/11/07 at 06:29 AM

I think that he decided not to market it. He's a retired guy with money and probably didn't want the liability. I like the design though. Not particularly graceful, but well detailed, proportioned and fun looking. Almost buggyish..


rpmagazine - 29/11/07 at 08:34 AM

quote:
Originally posted by rontyler
quote:
Originally posted by rpmagazine
However the transverse engines have an important asset in that the engine and transaxle can be rotated to lower the engine or the bell housing can be modified to achieve the same outcome.



Great point. As long as the lubrication is attended to... its an interesting thought... hmmmmmm....



Not really original though as Lancia did oh so many years ago.


rontyler - 29/11/07 at 02:47 PM

quote:
Originally posted by rpmagazine
Not really original though as Lancia did oh so many years ago.


Now that you mention it, I remember seeing old circle tracker's with Offy's layed over nearly flat. To my knowledge, the only way to go to that extreme is dry sump... but cool nonetheless.


Delinquent - 5/12/07 at 11:19 AM

can anyone just confirm if there were any/ which subaru were 6 cyl / 2wd?


hughjinjin - 5/12/07 at 01:39 PM

hi, the alfasud i'm modifying has the same issues as the subaru drivetrain ie heavy engine for the capacity relative to an equivalent inline 4, high crank position relative to the bottom of the block and a transaxle with the output shafts below the mainshaft/ crankshaft line.
My latest plan is to fit a hewland mk 9 transaxle (basically an inverted beetle gearbox with a straight cut dog engagement gearset). This will allow me to mount the engine about 75mm lower than the alfa box.


rontyler - 5/12/07 at 11:23 PM

quote:
Originally posted by hughjinjin
hi, the alfasud i'm modifying has the same issues as the subaru drivetrain ie heavy engine for the capacity relative to an equivalent inline 4


Do we really know this or is it based on supposition? The EJ22 is under 250lbs and makes 135 hp in ‘grocery getter’ tune. A little headwork, cams, intake and exhaust I’m betting would put it over 175 (only 80hp per liter). Are there any commonly availabe inliners, with roughly 175 HP (with common mods), that weigh under 250lbs complete? I don’t mean to sound cynical, its just that no emperical data has been presented. The puprose of this thread is to generate that kind of info before I make the mistake of choosing the Suby, so my interests are genuine.


kreb - 6/12/07 at 12:33 AM

Ron - you posted very nearly what I was going to. The Subbie 4 cyclinder engines are in the mid 200lb range and the stock output of the 2.5 liter is about 175 with abundant torque. So those engines are no slouches. The 6 cylinder engines are nice too - 230 to 270 HP but packa good deal more poundage (275 or so?)


Delinquent - 6/12/07 at 09:59 AM

quote:
Originally posted by kreb
Ron - you posted very nearly what I was going to. The Subbie 4 cyclinder engines are in the mid 200lb range and the stock output of the 2.5 liter is about 175 with abundant torque. So those engines are no slouches. The 6 cylinder engines are nice too - 230 to 270 HP but packa good deal more poundage (275 or so?)


Thats really quite light isn't it? The Audi V6 I was originally contemplating comes in at around 350lbs for not dis-similar output.


kb58 - 6/12/07 at 02:33 PM

Has anyone actually weighed these engines, or is it hearsay? Also, is that of the bare engine or complete? Without knowing what's included it isn't very informative.

It's like the V8 argument that they hardly weigh any more than a four cylinder, yet it's of a shortblock! Until someone posts a picture of a *complete* Subaru engine hanging from a digital scale, I'm skeptical. And we also must include transmission weight!

Engine weights seem to be about as reliable as people's golf score...


[Edited on 12/6/07 by kb58]


kreb - 6/12/07 at 02:58 PM

Kurt - your point is, of course valid. there are a variety of published weights for practically any engine. I'm still looking for that 400 pound small block Ford! (Bare long block with aftermarket aluminum heads - maybie.) That said, I have seen mid 200s listed for Subaru H4s enough times to feel pretty good about it.

Most of your DOHC vee engines are fairly large and porky. All that valvetrain adds up!


rontyler - 6/12/07 at 04:13 PM

quote:
Originally posted by kb58
Has anyone actually weighed these engines, or is it hearsay? Also, is that of the bare engine or complete? Without knowing what's included it isn't very informative.

It's like the V8 argument that they hardly weigh any more than a four cylinder, yet it's of a shortblock! Until someone posts a picture of a *complete* Subaru engine hanging from a digital scale, I'm skeptical. And we also must include transmission weight!

Engine weights seem to be about as reliable as people's golf score...


[Edited on 12/6/07 by kb58]



Well Kurt, put your bib on so I can spoon feed you :-D

About 14 posts back, I gave you a link to the very thing you're asking. Here it is again for your viewing pleasure...

http://forums.hybridz.org/showthread.php?t=125246

Scroll to the bottom.

You'll see an EJ22 that weighs 214lbs. It is wet weight, missing the damper, flywheel, alt., and exhaust manifolds.

Allow me to edify this... This was done with *my* scale. A scale thats commercial grade, used prolifically by the largest battery manufacturer in the world. Those weights were taken by a close friend of mine, of 15 years. A gentleman that I know better than my own brothers. He's as honest as the day is long. He's certificated A&P, a pilot, and known internationally for his engine building success. He takes technical tasks seriously. If you'll read the first post in that thread, you'll see that his views mirror your own. If you read the second post, you'll see that the weights were taken with a hanging, digital scale, per your request.

I will personally guarantee the repeatability of that number.

Now, the question begs, can anybody show me something similar with an inliner?

[Edited on 6/12/07 by rontyler]


kb58 - 6/12/07 at 09:28 PM

Here's a data point: 1995 Honda H22A1, including complete engine with starter, alternator, flywheel, pullies, intake, header, tranny and axles. 475lbs. 190whp with only intake and exhaust.

If everyone posted with this detail we wouldn't have to guess engine and drivetrain weights.

So what's the entire Subaru drivetrain weight - all of it.

[Edited on 12/6/07 by kb58]


rontyler - 6/12/07 at 10:18 PM

Kurt, I was really hoping you wouldn't take my post 'personally'. That wasn't the intent. It was more in 'jest'. Peace?

If you read through this thread, top to bottom, I think you'll see that I've done my best to keep it factual and empirical, weeding out the hearsay as much as practical.

quote:
Originally posted by kb58
Here's a data point: 1995 Honda H22A1, including complete engine with starter, alternator, flywheel, pullies, intake, header, tranny and axles. 475lbs. 190whp with only intake and exhaust.


Great, thank you!

quote:
Originally posted by kb58
So what's the entire Subaru drivetrain weight - all of it.



That's the scary part isn't it? As I mentioned earlier, the Suby tranny is not looking very svelt. Sorry, I don't have one in my possession to weigh. But I suspect its rather dissapointing. On the other hand, It would have to weigh over 200lbs to have a deficit on the Honda you quoted. And, well, it could actually be that heavy. If a guy was willing to spend some money, there are other transaxles that can be made to work, both lighter and with a better input shaft to axle split. However, I very much would like to keep this 'locost'. In the end, the tranny certainly is the most offensive part of this drivetrain.


THAWA - 7/12/07 at 12:58 AM

There's no way the a FWD 5mt weighs more than 200 lbs. I've moved many AWD 5mts alone, and many EJ engines, and an EA engine alone. Just the short block of the EJ feels heavier. Yes I know this isn't real data, and the two are different shapes with different CoG's, but I struggle at times with a shortblock, depending on what's still attached, whereas a tranny isn't THAT bad. I think someone somewhere posted the weights, cant find it though.

FWIW a 4eat feels to weigh about the same as a block with one head and intake manifold still on. and 4eat's are much heavier than 5mts.

[Edited on 12/7/2007 by THAWA]


rontyler - 7/12/07 at 01:13 AM

quote:
Originally posted by THAWA
There's no way the a FWD 5mt weighs more than 200 lbs. I've moved many AWD 5mts alone, and many EJ engines, and an EA engine alone. Just the short block of the EJ feels heavier. Yes I know this isn't real data, and the two are different shapes with different CoG's,


That's encouraging. If anybody can produce even rough tranny weights, Id be greatful.

I could use a little help with the designations... 5mts and 4eat? 5 speed manual and 4 speed auto I'm guessing?


THAWA - 7/12/07 at 06:58 AM

You got it.


Delinquent - 7/12/07 at 11:18 AM

some actual sizes of various engines would also be very helpful to build a decent database for those looking for something outside the ordinary - I have U2U'd Meany after his post early on in this thread but nothing back yet.


gator - 10/12/07 at 03:44 AM

Hi there, I must keep the controversy raging. Using half of my incredibly expensive corner weight set, I would suggest that a 2WD scooby box (less clutch fork etc) weighs in the order of 43kg. The 4WD box (with clutch fork) 57.5kg. Regards, Alan Rescued attachment 100_1582.jpg
Rescued attachment 100_1582.jpg


gator - 10/12/07 at 03:46 AM

The other view. Regards, Alan Rescued attachment 100_1586.jpg
Rescued attachment 100_1586.jpg


rontyler - 10/12/07 at 11:05 PM

quote:
Originally posted by gator
Using half of my incredibly expensive corner weight set, I would suggest that a 2WD scooby box (less clutch fork etc) weighs in the order of 43kg. The 4WD box (with clutch fork) 57.5kg. Regards, Alan


My HERO! I owe you a beer or three :-) Thanks!

Loosely speaking, that puts me 100lbs under Kurts example, although with a bit of a power deficit. I can live with that.

So the next issue is CG. It within a range that I'm willing to accept, but wouldn't mind dropping it if I could. I was kicking around the idea of flipping the trans over. 'Course, it would spin the drive wheels backwards. There probably isn't a chance in hell I could expect to flip the ring gear. Probably have better luck with a lottery ticket. Outside of that 'little hurdle', anybody know of the consequences of running a transaxle upside-down?

The other thought that occurred to me is running the engine backward... but thats a can of worms I'd rather avoid.

Thanks again.


RazMan - 10/12/07 at 11:51 PM

I think the Porker boxes are often flipped over without too much drama - at least the Ultima boys seem to get away with it.


Delinquent - 11/12/07 at 04:02 PM

quote:
Originally posted by rontyler
Outside of that 'little hurdle', anybody know of the consequences of running a transaxle upside-down?


I'd have thought the most obvious (other than having lots of reverse gears) would be lubrication - how is that handled in the scoobie box and would flipping it result in oil not getting where it's supposed to?


cymtriks - 15/12/07 at 12:09 PM

Some thoughts on mounting the engines and CoG height.

First with a flat engine:
allow 4 inches under the car
allow 6 inches for the exhaust headers
asume the cylinder heads are 8 inches wide (so half this, 4, to the engine CL)
This gives 4+6+4=14 from ground to the crank CL
assume the input to output shaft distance for the gearbox is 3 inches and the Gbx is not inverted
the output shafts are therefore at 11 inches
for a normal tyre size for a non turbo 2.5 engine, say 225/45-15 at the rear and 195/50-15 at the front, the distance from the ground to the wheel centres will be about 11.5 inches
So the axles will slope down by about half an inch from the hubs to the Gbx.

That all seems to work well. There are no extreme ground clearance or drive shaft angles to be concerned about.

The sump on the engine looks very deep so it may need cutting in the above example but a six inch deep sump isn't that extreme, unless someone is about to put me right on this point!

scaling from a picture I found on Google it looks as if the top of the intake is about 14 inches above the crank which puts it at 28 inches

now for a straight four and inverted box
the lowest point on the whole assembly is now the bell housing so...
allow 4 inches to the bell housing
assume a further 7 inches to the crank
the crank CL is therefore at 11 inches above the ground.
for a wheel radius of 11.5 inches and an input output Gbx distance of 3 inches we have a drive shaft slope of 2.5 inches.

This seems a bit much given that permanent misalignment of over 7 degrees (a 1:8 slope) is considered a no no for CV joints. Note that I used the word permanent there, I know that temporary much larger angles are no problem.

Assume that the top of the engine is 17 inches above the crank and this is at 28 inches.

This is exactly the same overall height as a flat four BUT the straight four has big cast bits such as the block and the head above the crank so the CoG is higher.



Conclusion...

Use a flat engine with the Gbx the normal way up, the CoG is lower, the drive shaft angles are better, the overall height is the same and you don't need to mess about with extra bits or mods to invert the box.

What about the Ultima set up?
lets assume 4 inches under the gearbox and a further 7.5 (assuming a slightly bigger housing than above) to the crank.
The crank is at 11.5.
The gearbox is inverted so the output shafts are at about 15 inches (again assuming a slightly bigger box than the above calcs)
The radius of the wheels is going to be about 13 inches (The ultima site gives two sizes, either side of this)
so the drive shafts slope 2 inches which puts them close to the maximum permanent slope of 7 degrees.

The height above the crank to the top of the block is about 15 inches so the overall height, excluding the intake, is going to be 26.5 inches.

Conclusion...

Inverting the box works when you have bigger tyres/wheels to keep ground clearance and drive shaft angles acceptable. The overall height isn't far off a smaller straight four with an inverted box or flat engine with a box the normal way up.

[Edited on 15/12/07 by cymtriks]


rpmagazine - 16/12/07 at 12:21 AM

it is a reasonable work through but fit to purpose becomes an issue. If you want to seriously use the vehicle on a track one of the first issues is wheel/tyre availability. 15" wheels immediately limits your choice of brands and compounds in my market (note US/UK may be different).
You are also assuming no modification to components. With more modification the advantages are lessened or changed...but the Subie is limited in this respect.
Having said that I think they are a very good thing, it's just that the picture is always more convoluted that we realise.


kb58 - 16/12/07 at 06:42 PM

quote:
This is exactly the same overall height as a flat four BUT the straight four has big cast bits such as the block and the head above the crank so the CoG is higher.


But then there's the accessaries. There's nowhere for the heavy alternator, starter, intake, and maybe turbo to go, except up above the engine. I agree that the CG will still lower than a I-4, but I suspect the advantage is less than expected; the CG certainly isn't at crank centerline. The flat-4's advantage is even smaller when compared to an I4 with an aluminum block and head.

[Edited on 12/16/07 by kb58]


gator - 17/12/07 at 12:35 AM

A mesage for Del, I have accidentally deleted your Email address when clearing spam (Oh Bugger).
The dimensions you were after: O/all length 760mm; distance from leading edge of bellhousing to centerline of driveshaft approx 165mm. Hope this helps, sorry for the stuffup.
Regards, Alan.


cymtriks - 17/12/07 at 11:20 AM

quote:
Originally posted by rpmagazine
it is a reasonable work through but fit to purpose becomes an issue. If you want to seriously use the vehicle on a track one of the first issues is wheel/tyre availability. 15" wheels immediately limits your choice of brands and compounds in my market (note US/UK may be different).
You are also assuming no modification to components. With more modification the advantages are lessened or changed...but the Subie is limited in this respect.
Having said that I think they are a very good thing, it's just that the picture is always more convoluted that we realise.


Assuming you you mean the traditional racing size of 13 inch wheel then the numbers won't change much.

For racing the rolling radius will be slightly less (but not much as side walls of the tyres tend to be taller) but the ground clearance will also be less as it's a race car.

The result will be that my numbers are still about right, only with a small change in drive shaft angle and overall height due to everything being a bit lower.


Delinquent - 17/12/07 at 11:35 AM

quote:
Originally posted by gator
A mesage for Del, I have accidentally deleted your Email address when clearing spam (Oh Bugger).
The dimensions you were after: O/all length 760mm; distance from leading edge of bellhousing to centerline of driveshaft approx 165mm. Hope this helps, sorry for the stuffup.
Regards, Alan.


Many thanks Alan - when you say overall length, does that include the gearchange link (which I believe still pokes out the back of the 2WD box?)

Would be rather handy if it did, as it is a perfect fit within the bodyshell design at that size - would only require me to move the fwd mounts for the wishbones back - which is a point to note for anyone else considering this unit, the distance from the driveshaft centres to the back edge of the engine mean you're either going to have very short wishbones or the leading leg of the wishbone is going to be nearly perpendicular to the centreline.


cymtriks - 17/12/07 at 11:36 AM

quote:
Originally posted by kb58
quote:
This is exactly the same overall height as a flat four BUT the straight four has big cast bits such as the block and the head above the crank so the CoG is higher.


But then there's the accessaries. There's nowhere for the heavy alternator, starter, intake, and maybe turbo to go, except up above the engine. I agree that the CG will still lower than a I-4, but I suspect the advantage is less than expected; the CG certainly isn't at crank centerline. The flat-4's advantage is even smaller when compared to an I4 with an aluminum block and head.


For a flat engine I'd expect the intake to be balanced by the exhaust hedders and sump casing which leaves only the starter and alternator.

I have no idea where the starter is on a flat engine, I'm just guessing that it'll have to be on top like the alternator is!

I'd guess that the CoG is about 2 iches above the crank.

For a straight four the cylinders, head, intake and exhaust are all above the crank. The cylinders by about 4 inches, the head, exhaust and intake by about 11. The crank is probably the heaviest bit of the asembly by a long way.

The starter and alternator are often above the crank as well but even if they're not I can't see the CoG being less than half way up the block.


So the advantage of a flat engine is all of 2 inches in CoG!

There is still the advantage of overall height without having to mod the gearbox but, as rpmmagazine points out, there's nothing outside fiddling with the turbo boost in the way of tuning for Subaru engines and personally I'd steer clear of that in a fibreglass car (turbo heat and fibreglass can be bad neighbours). On the other hand there's a huge ammount of normal aspiration tuning stuff for Duratec straight fours.


kb58 - 17/12/07 at 02:33 PM

quote:
Originally posted by cymtriks I'd guess that the CoG is about 2 iches above the crank.


I would have thought that the in-lin CG would be lower, due to the heavy steel crank moving the CG to below half way up the block. I forgot about the CG effect of the intake assembly... I'm assuming an aluminum block and head, too. If both are steel, it's a lost argument for the inline.

While I'm not up on modified Subaru engines, I agree that aftermarket "go-fast" parts are much more widely available for inline engines. Just look in any car magazine and it's hard to find ads for Subie drivetrain parts.

All this stuff aside, the homebuilt car that a Subie will fit must have a really wide engine bay, and accomodate the long transaxle tail shaft. It would be a very different Locost, and not so low cost either. I recently saw a complete drivetrain from an STi... $10,000. That's crazy. Oh sure, there are Subi engines much cheaper, but they're the old ones that don't make much power.

[Edited on 12/17/07 by kb58]


cymtriks - 17/12/07 at 05:27 PM

CoG heights assuming the lowest mounting for each type-

Flat
4 inches for ground clearance, 6 for the exhaust and 4 for half the cylinder head gives a crank centre at 14 inches. The weight of the intake above the crank is balanced by the weight of the exhaust and sump below it (though the sump will need cutting as it looks deeper than the exhaust as standard). That leaves the effect of the remaining ancilliaries which raise the CoG to, as a guesstimate, 15 to 16 inches. The overall height is about 28 inches.

Straight
4 inches for ground clearance, 7.5 for the bell housing (the lowest point on the assembly, not the sump) gives a crank centre at 11.5 inches. The sump is just a hollow box and the cylinders, head, intake and exhaust are all higher. While the crank is still the most massive bit the combined weight of the intake, head and exhaust can't be far off and it is about 12 inches above the crank. So a guesstimate of the CoG would be about 15 to 16 inches. The overall height is about 28 inches.

So assuming you are not restricted by anything other than the engine there's no gain either way. The CoG is in about the same place and the overall height is about the same.

BUT

In order to mount a flat engine at the height given above you can use the transaxle as it came and run low drive shaft angles. To do this with the straight engine requires that you invert the box (cost and time to modify) and run higher driveshaft angles (possible long term reliability issues).


Question-
Can the cheaper (not porsche G50 or Hewland!) boxes be inverted at reasonable cost?
Audi?
Renault?
Subaru?
For the home builder this might be the deciding factor.


kb58 - 17/12/07 at 08:36 PM

I recently discovered that Volvo has an in-line transaxle, too.


Ringius - 17/12/07 at 10:54 PM

Hi Kimini!

Do you mind telling what volvo has the in-line transaxle? Since I live in Sweden, this might be an attractive alternative to the Audi box for me (availability of wrecked Volvos is pretty big here).

BR,
Ringius


gator - 18/12/07 at 12:14 AM

quote:
Originally posted by Delinquent
quote:
Originally posted by gator
A mesage for Del, I have accidentally deleted your Email address when clearing spam (Oh Bugger).
The dimensions you were after: O/all length 760mm; distance from leading edge of bellhousing to centerline of driveshaft approx 165mm. Hope this helps, sorry for the stuffup.
Regards, Alan.


Many thanks Alan - when you say overall length, does that include the gearchange link (which I believe still pokes out the back of the 2WD box?)

Would be rather handy if it did, as it is a perfect fit within the bodyshell design at that size - would only require me to move the fwd mounts for the wishbones back - which is a point to note for anyone else considering this unit, the distance from the driveshaft centres to the back edge of the engine mean you're either going to have very short wishbones or the leading leg of the wishbone is going to be nearly perpendicular to the centreline.


Hi, the 760mm includes the basic gearshift mechanism as shown, it's the most rearward part. I haven't given any thought to the arrangement of linkage(s) forward to the cockpit as yet.
Regards, Alan. Rescued attachment 100_1610.jpg
Rescued attachment 100_1610.jpg


rpmagazine - 18/12/07 at 09:19 AM

quote:
Originally posted by cymtriks
CoG heights assuming the lowest mounting for each type-

Flat
4 inches for ground clearance, 6 for the exhaust and 4 for half the cylinder head gives a crank centre at 14 inches. The weight of the intake above the crank is balanced by the weight of the exhaust and sump below it (though the sump will need cutting as it looks deeper than the exhaust as standard). That leaves the effect of the remaining ancilliaries which raise the CoG to, as a guesstimate, 15 to 16 inches. The overall height is about 28 inches.

Straight
4 inches for ground clearance, 7.5 for the bell housing (the lowest point on the assembly, not the sump) gives a crank centre at 11.5 inches. The sump is just a hollow box and the cylinders, head, intake and exhaust are all higher. While the crank is still the most massive bit the combined weight of the intake, head and exhaust can't be far off and it is about 12 inches above the crank. So a guesstimate of the CoG would be about 15 to 16 inches. The overall height is about 28 inches.

So assuming you are not restricted by anything other than the engine there's no gain either way. The CoG is in about the same place and the overall height is about the same.

BUT

In order to mount a flat engine at the height given above you can use the transaxle as it came and run low drive shaft angles. To do this with the straight engine requires that you invert the box (cost and time to modify) and run higher driveshaft angles (possible long term reliability issues).


Question-
Can the cheaper (not porsche G50 or Hewland!) boxes be inverted at reasonable cost?
Audi?
Renault?
Subaru?
For the home builder this might be the deciding factor.


When we are engineering from given components we are in some respects reverse engineering. Our priorities for intended purpose and ability to produce our final item will make many decisions for us.
For me the wheel diameter was set by the convergence of two facts: in my part of the world there is more competition tyre choice in 17" in the desired sizes and going larger resulted in a heavier wheel/tyre combination (and more weight further from the centre increasing gyscopic effects, power loss and cost as tyres were much more expensive, plus there was a loss of desirable compliance in tyre sidewall, but I digress.)
So wheel diameter is set by what is important to you. This then sets the outer driveshaft height. This influences the inner driveshaft height as you cannot have more than a given static deflection dependent on driveshaft joint type. The often quoted 7deg is a misnomer for normal ride height as you need a couple of degree of leeway so 5deg is safer, remembering that the angle will increase/decrease dependent on level of suspension travel.
So wheel center actually 'sets' inner drive shaft flange height. The ability to lower mass below this point then becomes the issue. With an inline engine this is difficult as the input shaft is fixed and competition transaxles are expensive. Flipping the transaxle will work, but not all are suitable. However there is another element: box ratios and suitability for purpose and cost of replacement...a Pfitzner gear cluster for a Subie box will cost $4000AUD plus build. So this may influence choice.
But getting back to CoG even if you flip the box, that is about all you can do.
Transvarse box has all the same driveshaft issues, however you can cut and rotate the box on the bell housing and thus lower the engine relative to the driveshaft or you can rotate the whole assembly assuming the oil drainage in the engine and lubrication in the box will allow it. You can maximise this effect by reducing the FW diameter and cutting the bell housing down to suit and then rotating the box/bell housing some more...you cannot do this with the inline if the driveshafts angles do not allow it.
Now having said all of that the inline allows a greater level of lowering in most instance due to the fact that the driveshafts are longer
Let me finally add that I think the Subie gear is very good.

[Edited on 18/12/07 by rpmagazine]


Delinquent - 18/12/07 at 11:51 AM

quote:
Originally posted by gator

Hi, the 760mm includes the basic gearshift mechanism as shown, it's the most rearward part. I haven't given any thought to the arrangement of linkage(s) forward to the cockpit as yet.
Regards, Alan.


lovely job - that would fit nicely, cheers


Benonymous - 19/12/07 at 04:33 AM

Am I missing something here? If one was to use a Subaru engine/gearbox, in a mid rear engined car, then the gearbox is already twirling the wheels in the right direction. No need for flipping gearboxes because the Subaru motor is already ahead of the gearbox in the engine compartment. The reason for flipping Porsche gearboxes is to get the mid rear engine to spin the wheels in the right direction due to the fact that the venerable 911 is rear engined. Also, using a 2WD box would definitely be the best solution as it would lack the components for 4WD output. have you ever noticed how similar Subaru parts are to VW beetle pats ?


gator - 19/12/07 at 05:03 AM

I'm with you! I know someone who has the setup in a Porsche spyder replica. Seems OK to me despite the concerns most express on gearbox weaknesses.
Regards, Alan


Delinquent - 19/12/07 at 12:16 PM

quote:
Originally posted by Benonymous
Am I missing something here? If one was to use a Subaru engine/gearbox, in a mid rear engined car, then the gearbox is already twirling the wheels in the right direction. No need for flipping gearboxes because the Subaru motor is already ahead of the gearbox in the engine compartment. The reason for flipping Porsche gearboxes is to get the mid rear engine to spin the wheels in the right direction due to the fact that the venerable 911 is rear engined. Also, using a 2WD box would definitely be the best solution as it would lack the components for 4WD output. have you ever noticed how similar Subaru parts are to VW beetle pats ?


People flip the box to try and get the engine seated as low as possible in the car (driveshafts sit below centreline in "normal" position) - as you say with a Scoobie unit this poses a problem as you'd then be spinning the wheels backwards, so would need additional work on the diff to correct it.

The VW box IMO is no good for this type of app - I used to be heavily into modded VW's and you could pretty much guarantee an upgrade in the engine would result in ripped up boxes. Even the Hewland Mk9 (based on VW shell) is only rated to 180 lb ft


kb58 - 19/12/07 at 02:23 PM

Volvo has some FWD transaxles, oriented for north/south engines.


rpmagazine - 24/12/07 at 01:47 AM

http://www.blastautomotive.com/photos.html


THAWA - 30/12/07 at 07:03 PM

quote:
Originally posted by kb58
quote:
Originally posted by cymtriks I'd guess that the CoG is about 2 iches above the crank.


I would have thought that the in-lin CG would be lower, due to the heavy steel crank moving the CG to below half way up the block. I forgot about the CG effect of the intake assembly... I'm assuming an aluminum block and head, too. If both are steel, it's a lost argument for the inline.

While I'm not up on modified Subaru engines, I agree that aftermarket "go-fast" parts are much more widely available for inline engines. Just look in any car magazine and it's hard to find ads for Subie drivetrain parts.

All this stuff aside, the homebuilt car that a Subie will fit must have a really wide engine bay, and accomodate the long transaxle tail shaft. It would be a very different Locost, and not so low cost either. I recently saw a complete drivetrain from an STi... $10,000. That's crazy. Oh sure, there are Subi engines much cheaper, but they're the old ones that don't make much power.

[Edited on 12/17/07 by kb58]


10 grand for a US STi swap is ridiculous. Especially when you can get a v8 swap for like 8.

You're right there isn't a "tuning catalog" like there are for hondas, ford, nissan , and whatever. Which I think is good. Helps keep the silly poo to a minimum.

You can easily make tons of power from any subaru engine, all you need is a good turbo, injectors, and engine management. Sure in stock form the EJ257 is killer, but the same amount of power can be made with an EJ22T, EJ20G, EJ205, or EJ20K. Actually the EJ20G and EJ20K pretty much the same power, 280ps vs 300 hp, and cost significantly less.

Depending on the dyno a stock STi will put down 230-250 WHP, and 230-250 FT/LBS. There are multiple people dynoing EJ22T's at 200-250 WHP, and 230-300 FT/LBS, with nothing more than a larger turbo, bigger injectors, an intercooler, and sometimes management. The EJ22T stock is 160hp, 181 ft/lbs at the flywheel. People with EJ20G's are putting down good numbers with less changes to the engine, simply because it's made to do so.

Of course if you take an STi engine, and give it the same treatment, you will make more power, but it's not like it's unattainable with an older engine.

Oh BTW,
EJ22T = 91-94 US Legacy Turbo
EJ20G = 89-92 non-US Legacy/Liberty Turbo, and 92-97 Impreza WRX and STi
EJ20K = 96-97 Impreza WRX and STi sedan only
EJ205 = 99-present non-us Impreza WRX, and 02-05 US Impreza WRX
EJ257 = 04-present US Impreza WRX STi

The EJ22T is a sad engine in stock form. It has the smallest turbo since the Loyale, no intercooler, tiny SOHC heads, and an old 8-bit ECM.


Gakes - 13/1/08 at 06:59 PM

I know this is an old thread but, the advantages of having the flat 4 suby is that almost all mechanical forces are balanced, including gearbox, and shafts. most i4's have a long and short driveshaft that creates torque steer in front wheel drives and tail happy middy's


TheGecko - 13/1/08 at 11:47 PM

quote:
Originally posted by Gakes
I know this is an old thread but, the advantages of having the flat 4 suby is that almost all mechanical forces are balanced, including gearbox, and shafts. most i4's have a long and short driveshaft that creates torque steer in front wheel drives and tail happy middy's


Not strictly true nowadays. Most modern transverse drivetrains with any significant sort of power output have a fixed shaft with the inner CV on the end of it so that the moving driveshafts are equal length.

I have two different drivetrains at home (long story ) and both of them are configured that way. And they're a lot more compact in plan than a boxer layout.

Dominic


kb58 - 14/1/08 at 02:19 PM

Can someone please tell me the approximate size of the drivetrain:
1. How far forward of axle centerline is the front pulley of the engine?
2. How tall is the engine above the crankshaft, including all accessaries?
3. How wide is the engine, side-to-side, including any manifolds.
4. How far behind axle centerline does the transmission tail shaft extend?

Thanks!


RazMan - 14/1/08 at 04:16 PM

I can't take credit for these so apologies to the person who originally uploaded them Rescued attachment imprezaenginesize.jpg
Rescued attachment imprezaenginesize.jpg


RazMan - 14/1/08 at 04:17 PM

Any help? Rescued attachment imprezaenginedimensions.jpg
Rescued attachment imprezaenginedimensions.jpg


kb58 - 14/1/08 at 06:24 PM

Very cool, thank you both!


Delinquent - 16/1/08 at 09:35 AM

quote:
Originally posted by kb58
Very cool, thank you both!


Not sure if it's already been mentioned in this thread, but you can purchase adapters from Saker to convert the 4wd scooby boxes to 2wd by removing the rear end of the box - considerably shortens it but keeps the strength advantage the 4WD box has over the 2WD.


THAWA - 19/1/08 at 05:02 AM

quote:
Originally posted by Gakes
I know this is an old thread but, the advantages of having the flat 4 suby is that almost all mechanical forces are balanced, including gearbox, and shafts. most i4's have a long and short driveshaft that creates torque steer in front wheel drives and tail happy middy's


That's really not true. A boxer 4 still has imbalances, and any Subaru will torque steer given either enough torque, lack of grip, or shitty tires. Of course both of these cases will still be less than a 2wd inline 4, but they're still apparent in a boxer4, 2wd or awd.


Gakes - 21/1/08 at 08:54 PM

quote:

Thawa

That's really not true. A boxer 4 still has imbalances, and any Subaru will torque steer given either enough torque, lack of grip, or shitty tires. Of course both of these cases will still be less than a 2wd inline 4, but they're still apparent in a boxer4, 2wd or awd.



You are correct. I'm being biased of the boxers. There is much less torque steer on these powerplants, it is very noticeable with good tyres. But it still does a little


Kaspa - 6/3/08 at 11:56 PM

guys your horribly underestimating the subbies output , down here in the antipodes the EJ20t puts out 220 hp and with few minor mods 250-300 is easily acheivable, for an instalation set up and specs Google Saker as they use the subby power train , and are designed and built by a good freind of mine from just around the cnr,
plus i have just bought a chassis all set up with a EJ20t in it, some pics of said beast in my last post
cheers Kaspa


Doug68 - 7/3/08 at 07:14 AM

Here's what you can get if you go really silly with a Suby motor. Of course I'd expect the owner to be in debt for the rest of their life.


Kaspa - 8/3/08 at 01:52 AM

Doug, actualy not as expensive as i would have thought, i spoke to a subby engine specialist yesterday and his advice was a late model 2-5ltr engine,with piston and rod upgrade , cams and turbo,= 450hp reliable engine for around $5-6 g nz.
he also added for ultra reliability get the new 6 speed box, much stronger diff, & pinion carriers [the weakness on earlier boxs, they tend to spread apart with big Hp]
cheers Kaspa