Which Forester engine codes/

[guzzla]

I know that the first series SG Forester in Australia had the 112kW engine and this was updated to 121kW in the SG facelift of mid 2005 and then updated again to 126kW in the early 2008 release of the SH Forester.
What are the different engine codes for these engines eg EJ 253 and what was actually done to them in the updates to change the power and torque?

I notice that in various "Wheels" magazine tests that with each update they gave better performance figures, which you would expect.

The other thing I have always wondered is why the apparently identical engine in the Outback had 115kW v the Forester's 112kW, and the engine in the 3rd series Outback had 127kW v the Forester's 126kW, yet the 121kW engine was common to both?

I'm certain some of you knowledgeable folks on here like Ratbag must have some of the answers that I can't find.

[EDIT] This subject has also been extensively ventilated in the thread here:
https://offroadsubarus.com/showthread.php?t=5020
[end edit]

 

[Ratbag]

You rang, Sire? :rotfl:

Basically two different engines.

The SG Series I used the EJ-251 2.5L donk. This came in two varieties - SOHC and DOHC (a different engine code). AFAIK, we never got the N/A DOHC variant here. It actually had less power and torque than the SOHC over most of the rev range. IOW, not a good idea. Great with a turbo, but not N/A. Driving the extra camshaft per bank used more power than it produced in this engine, AFAICS.

The Series II used the EJ-253.
The main differences were the heads and the use of a MAF sensor instead of a MAP sensor. The former tells the ECU the rate of flow of the inlet air, the latter tells it the mass of the inlet air Assuming that I got it the right way around ... :iconwink. This has fairly dramatic ramifications when the engine is cold, hot, or at high elevations, as it allows the ECU to deliver fuel with greater precision for the circumstances and the flow x density of the inlet air, rather than just the flow rate.

WRT to the heads, the cam and valve design are completely different; and the SII engine uses tumbler valves on the inlet manifold. The SII donk utilises variable valve lift technology (two different profile cam lobes for one of the two inlet valves dependent on the rpm/load etc). The SI does not. This extends the flat part of the torque curve dramatically at both ends. AFAIK, the SI delivers about 80% of its maximum torque between 1800-4800 rpm. The SII delivers 80% of its maximum torque from about 1200 to 6300 rpm, and 90%+ from 1800 to 6300 rpm.

The EJ-253 in SWMBO's 2009 build SH has a longer inlet manifold than the one in my SG SII. Little differences like this can make a difference to maximum torque, low & high rev torque - i.e. the flatness of the torque curve. This could account for the minor differences between various models that are all using the 'same' engine. So can injector size, ECU programming, etc, etc.

It is the flatness of the torque curve (and its adequacy for the size, weight and gearing of the car) that determines almost all performance characteristics. Acceleration is (very crudely) a function of torque, vehicle mass and gearing. Equally crudely, maximum speed is a function of power, frontal area and gearing.

Therefore, the all-important figure is the torque curve, and how flat it is. This assumes that the gearing is a good fit with the vehicle. It is with my SG; it was lousy in my '93 Impreza ...
The shape and start/finish of the torque curve determine how tractable a vehicle will be in any given gear and situation.

Hope this helps a bit.

 

[guzzla]

You rang, Sire? :rotfl:
Hope this helps a bit.

Thanks Ratbag, you're a font of knowledge.
So this accounts for the reason my brother's SG manual S1 never feels as powerful as the SG manual S2 that I drove. In fact the whole car is actually better including the seats and trim (and looks). What I like about the N/A engine compared with my turbo is the evenness in delivery without the turbo lag at low revs and the almost frightening rush and spike of power that can cause jerkiness at times. It just doesn't require the skill of operation that my turbo sometimes does.

You're certainly right about the flat, wide spread torque curve. It never ceases to amaze me the incredible drivability of that little 1.4 turbo in our new Golf with its 200Nm of peak torque from 1400rpm to 4100rpm, even with its tall gearing.

Imagine what the 1.8 turbo that VW, Audi & Skoda share that has 250Nm (and some models 300Nm) of peak torque from 1250rpm to 5100rpm must be like! Basically in these engines the peak torque lasts all the way through to where the peak power starts & is spread over about a 1000rpm rev range. Engine design and management has sure come a long way in a few years.

 

[Rally]

I drove an SH XT, and found the power delivery to be very linear. While the XT did not feel very quick, the linear power delivery would certainly make it better for off road use. As Ratbag has on many occassions pointed out, torque is king and this is where Subaru's decision with the current model to go to 2 litre engines in the manuals and the XT is a retrograde step. The later SG engines are definitely better than the early ones in their performance (and fuel consumption).

I remember hearing at the time the reason stated for the difference in the power outputs was due to exhaust system differences. Certainly, I think the weak link in my SG is the engine. For the performance offered the fuel consumption is unacceptably high.

 

[Ratbag]

Gidday Guzzla

Thanks Ratbag, you're a font of knowledge.

Thanks for the compliment ... :biggrin:
But sorry, but I can take no credit for that - my brain does it whether I like it or not!

So this accounts for the reason my brother's SG manual S1 never feels as powerful as the SG manual S2 that I drove. In fact the whole car is actually better including the seats and trim (and looks).

What I really didn't like about the Series I Foresters that I test drove was the 4 spoke steering wheel. I have never liked this, and none of the cars I have bought have had one. The dual arm rests in the front look very flash, but don't work very well, IMHO. The single one in the SII is covered in nasty Naugahyde that almost immediately causes my arm to sweat all over it - UGH! I currently have an piece of old sheepskin seat cover attached to it with a rubber band! I do plan to remove the cover, and put a real suede leather cover over the top.

Apart from that, it only took one look at a SII to convince me that it was the direction I wanted to go. At that stage I knew absolutely nothing about engine differences - zero, zilch, nada. Didn't even know that there were any!

However, the SII that I drove, the first and only one - couldn't go past the price; full service history by Wippells (I checked with them ... ); low mileage, 100,000 Kms (low for any Subaru - just run in at around 75,000 Kms IME ... ); AND the incredibly tractable performance from such a (relatively) heavy car with a (relatively) small engine.

What I like about the N/A engine compared with my turbo is the evenness in delivery without the turbo lag at low revs and the almost frightening rush and spike of power that can cause jerkiness at times. It just doesn't require the skill of operation that my turbo sometimes does.

My feelings too. Having had some fairly close second hand experience with turbo engines, I didn't like the idea of this. I also avoided getting one for the same reason I didn't buy a concours condition E-Type Jag back in the early 1980s - my driving licence wouldn't last 5 minutes ... .

Buying a s/h turbo anything without knowing its full history is plain fraught IMNSHO. It has almost certainly been thrashed, and a new snail is more than expensive enough, let alone a full engine rebuild!

That smoothness of delivery makes mine an excellent vehicle for general use, and fantastic off road when towing a trailer (about 500 Kgs, loaded).

You're certainly right about the flat, wide spread torque curve. It never ceases to amaze me the incredible drivability of that little 1.4 turbo in our new Golf with its 200Nm of peak torque from 1400rpm to 4100rpm, even with its tall gearing.

Imagine what the 1.8 turbo that VW, Audi & Skoda share that has 250Nm (and some models 300Nm) of peak torque from 1250rpm to 5100rpm must be like! Basically in these engines the peak torque lasts all the way through to where the peak power starts & is spread over about a 1000rpm rev range. Engine design and management has sure come a long way in a few years.

That last sentence is spot on, IMO. The "tiny little things" changed from the SI to SII engine were impossible in my youth. The SII engine has a torque curve that's flatter and wider than a conventional, thumping great US V8, but without the headline numbers, of course. That's just nothing short of incredible out of a N/A 2.5L four cylinder engine IMO.

I really do not like turbo engines ... Others may love them, and that's fine too. I don't.

 

[Ratbag]

Gidday Rally

I drove an SH XT, and found the power delivery to be very linear. While the XT did not feel very quick, the linear power delivery would certainly make it better for off road use.

Of course one can get this sort of linear performance and high, low rev torque from any turbo engine. Just witness the huge number of cars in the last couple of years that demonstrate this. The cost is that one has to be prepared for the inevitable consequences of an 'always on' turbo. The stress on an engine is proportional to the square of the increase in inlet pressure (boost). Thus a 7 psi boost equates to a 125% increase in stress ((14+7)/14)^2 = 225% of the stress on the un-boosted motor. Dropping the compression ratio a bit helps to reduce this, but the increase in stress is still very high.

The advantage of the older turbo engines with a cut-in at around 3500 rpm is that one can choose to drive them without using the turbo all the time. The downside is the very non-linear torque curve.

The 3.0L H6 has around the same maximum torque as the turbo H4 2.5L. I do not know what shape its torque curve is, but it almost certainly employs the same design features of the SII 2.5L in the later models, so the torque curve should be flat, and up there with a turbo 2.5L.

There is also a trade-off of the power used running the turbo vs the extra power it gives at low revs. While nothing like the problems associated with running a supercharger on a small capacity engine, it would still be there, I reckon.

As Ratbag has on many occassions pointed out, torque is king and this is where Subaru's decision with the current model to go to 2 litre engines in the manuals and the XT is a retrograde step.

Completely agree.

The later SG engines are definitely better than the early ones in their performance (and fuel consumption).

Ditto.

I remember hearing at the time the reason stated for the difference in the power outputs was due to exhaust system differences.

There are probably all sorts of fine tuning and tweaking in the later engine. Exhaust is a definite possibility. Fuel pump pressure and control is another. Completely sealed fuel system on the SII (I don't know if this was the case on the SI or not??). The length of the inlet manifold may also have been changed, as it has between the EJ-253 in my Forester and the EJ-253 in SWMBO's. Same engine, different inlet manifolds and exhaust system.

Certainly, I think the weak link in my SG is the engine. For the performance offered the fuel consumption is unacceptably high.

I wouldn't call it a weak link, Rally. Beats the living carp out of having the very best 2.0L donk in it! Just that Subaru did it better with the SII engine. As someone mentioned in another thread, it's impossible to retrofit VDC or X-drive to an existing car, it would be extremely difficult to fit a SII engine into a Series I Forester - too many control systems have been changed. The VVLT is done using oil pressure via some arcane magic overseen by the ECU ... I have only the vaguest notion of how this is done, but it is dependent on engine load and gear selected as well as revs, as far as I can work it out at all.

 

[guzzla]

Weirdly that 1.4 turbo engine in my Golf shows no indication that it's a turbo though (unlike my Forester's which shows all the typical turbo engine characteristics). There is no sound of the turbo spooling up, no spikes in power delivery, absolutely no turbo lag and no flat spots or jerkiness. And it starts instantly hot or cold, immediately settling into a 700rpm idle even when cold. I have never driven an engine with such benign perfect behaviour, even my Camry V6, which was a paragon of smoothness & civility. Compared to it my Forester turbo feels like Subaru never did get their engine management controls sorted out properly. And getting the more powerful reflashed tune still didn't fix that.

Talking of fuel consumption, my 1390kg Camry with a 141kW / 279Nm engine is measurably more economical than a 1390kg Forester X with 121kW / 226Nm engine. Gearing, AWD & aerodynamics come into play there though. I do know that it was quite a bit more economical than a Gen1 Liberty 2.2 engine that a friend owned at the time too. And it was smaller, lighter, less powerful and only 2 wheel drive. Some things are just hard to explain.

 

[Rally]

The stresses on a turbo engine are a function of boost, timing, and revs. Mild boost at low RPM are quite different to high boost, lots of timing and lots of revs. Indeed, it could be argued that low boost turbos offer less stress on an engine than a NA engine, and this is reflected in their superior fuel economy to normally aspirated engines offering otherwise similar outputs.

In order for an NA engine to offer the same performance as a turbo engine, it must either be revved harder (lots of stress there), or be of significantly larger capacity (More fuel usage there). However, in the case of the Forester, the 2.5 SG1 and 2.5 turbo SH, the SG is less tractable around town and my understanding is that it uses around the same amount of fuel, maybe more, than the SH when driven at similar speeds.

Comparing the 2.5 litre SG1 with my 2.0 litre WRX, the WRX is actually more tractable even off boost, and is much harder to stall. It offers substantially more power/performance than even the XT, for about the same amount of fuel or even less as the SG1 when driven at the same pace. Trust me, the 2 litre in my WRX is a million miles ahead of the 2.5 in almost every way.

 

[Rally]

I don't know about you, but I am totally addicted to the turbo rush! I love the sound of the turbo spooling up and how the car becomes totally manic as the power comes on. It's a case on hold on, this is going to be FUN and you have to manhandle all this crazy grunt as the car fights you the whole way while your head gets shoved back into the head rest as if you've just being kicked in the face. :biggrin:

Maybe the XT has changed between SG and SH- because I could barely detect the turbo coming in at all.

Weirdly that 1.4 turbo engine in my Golf shows no indication that it's a turbo though (unlike my Forester's which shows all the typical turbo engine characteristics). There is no sound of the turbo spooling up, no spikes in power delivery, absolutely no turbo lag and no flat spots or jerkiness.

 

[Ratbag]

^ Guzzla, the reason it shows no "turbo characteristics" is that it is using a constant boost system. By this I do not mean a constant pressure boost, but a turbo that's always providing boost at some level.

The scrolling up sound is therefore not separable from the normal engine noise, as it is always there ...

There is always some kind of explanation for variances in things like fuel economy. Some are arcane, some obvious.

With Number 1 Colt, folding the external mirrors back against the body gave it measurably better fuel economy at highway speeds. It had a coefficient of drag of 0.29. Compare this with the 911 Porsche of the day at about 0.31 ... It was a slippery little beast!

Subaru were still making a 2WD version of the '93 Impreza, and it turned in nearly 1L/100 Kms better fuel economy than the AWD version. Same engine, drag (same body), gearing etc. Here it's obvious that the AWD was making the difference observed. TANSTAAFL ... much as we would all like to think we got the occasional one ... :lol:.

Our 2.2L Camry gave excellent economy on the open road, around 8.5L/100 Kms (by the standards of the day; or reality ... :iconwink: :lol, even fully loaded and running the air-con most of the time. Also an AT. Around town it was as bad as my Forester ... And the Forester is a good deal heavier; faster; more powerful (as in more torque); a far less aerodynamic body; etc; etc. Now the Camry engine was excellent design for its day (1994 model), but couldn't even start to compete with the sorts of technologies built in to the SII SG engine.

On the open road, SWMBO's 4EAT turns in around 8.5L/100 Kms, whereas Roo2 gets around 7.3-7.5L/100 Kms over the same road (Melbourne to Ballarat freeway and return). Both cars have far more in reserve than the Camry ever had, even though I would not criticise the Camry except that its engine was put in the car in such a way that it was extremely difficult to remove and refit ... .

So many variables. Is it any wonder that people are given to making absolutist statements about these things. Our whole species is rather partial to such behaviour ... :rotfl:.

 

[Ratbag]

The stresses on a turbo engine are a function of boost, timing, and revs. Mild boost at low RPM are quite different to high boost, lots of timing and lots of revs.

Quite so. I used the example I did as one almost needs to use factor analysis and calculus to work the factors out for a non-static situation (i.e. real life use). However, the highest boost modern turbos give is usually approaching a factor of two at full boost, rather than the old touchstone for superchargers of using a fixed boost of between 4 and 7 psi. This 14 psi boost at full boost equates to 4x the stress ...

The MGA Twin Cam 1600 that set the then world record at Bonneville flats was running a 34 psi boost supercharger (IIRC) ... Bloody amazing that the engine lasted for the two way run!!

Indeed, it could be argued that low boost turbos offer less stress on an engine than a NA engine, and this is reflected in their superior fuel economy to normally aspirated engines offering otherwise similar outputs.

In order for an NA engine to offer the same performance as a turbo engine, it must either be revved harder (lots of stress there), or be of significantly larger capacity (More fuel usage there). However, in the case of the Forester, the 2.5 SG1 and 2.5 turbo SH, the SG is less tractable around town and my understanding is that it uses around the same amount of fuel, maybe more, than the SH when driven at similar speeds.

This is where the VVLT, tumbler valves, and the like make such a difference in the SG SII engine. It doesn't need to have its rings revved off, ever - even though I have been known to give it the occasional squirt ... :iconwink: .

Comparing the 2.5 litre SG1 with my 2.0 litre WRX, the WRX is actually more tractable even off boost, and is much harder to stall. It offers substantially more power/performance than even the XT, for about the same amount of fuel or even less as the SG1 when driven at the same pace. Trust me, the 2 litre in my WRX is a million miles ahead of the 2.5 in almost every way.

I can't and won't argue with that, Rally.
Engine technology really does make a huge difference, as Guzzla said upthread. When Subaru decided to put all the turbo inlet tech into the N/A 2.5L, it made a very substantial difference to the tractability and drive-ability of the SG SII.

Who would have thought that tumbler valves would make such a difference? It makes sense when one thinks about it. It means that the valve timing, lift and duration can be changed to give higher torque at low revs without making the engine more likely to stall. Putting in the VVLT allows the use of a high lift, long duration cam for high rev use, without all the problems associated with such a camshaft at low revs. Combining the two technologies allows the torque curve to be pushed out in both directions, while flattening it as well.

Add in the MAF sensor, and this becomes even better.

Add in the latitude such mechanical changes allow with the re-programming of the ECU, the gains are very substantial.

My hotrod Morris 1100 was running a 25/65 camshaft (each valve open for 270 deg ... ) with 0.4" lift at the valves. It was literally un-drivable in heavy traffic - had to pull off the road and remove and clean the N3 plugs that would carbon up at the drop of a hat! The BMC Works racing 1340 cc Cooper S at the time was using a 20/60 cam with a 0.38" lift at the valves, for comparison ... :iconwink: :biggrin:.

 

[Ratbag]

I don't know about you, but I am totally addicted to the turbo rush! I love the sound of the turbo spooling up and how the car becomes totally manic as the power comes on. It's a case on hold on, this is going to be FUN and you have to manhandle all this crazy grunt as the car fights you the whole way while your head gets shoved back into the head rest as if you've just being kicked in the face. :biggrin:

The E-Type Jag I drove a bit in my youth was like that. Geared at about 33 mph/1000 revs in 4th, O/D. Push the pedal to the floor at about 15 mph, and you got pushed back into the seat, the long nose rose about 6", and off you went. That's from less than 500 rpm in top (not over-drive) ... :biggrin: .

It would do around 96 mph in second ... at 5500 rpm, IIRC.

All the while it behaved like a big cat cat ... :iconwink: :ebiggrin:.

 

[Rally]

It's much easier to compare the fuel efficiency of Camry's on the freeway. Seeing as they always do 10-15km/h below the speed limit- normally in the right hand lane next to another car doing the same speed- all Camry's of the same model get pretty much the same economy. Slip roads are another area where Camry's will get the same fuel economy. In this instance, they all share their economy as they enter the freeway, signposted at 110, at between 45 and 55 km/h, depending on the length of the slipway.

Around town it is a bit different. Fuel economy is affected by the time it takes to take off after a green light- this can vary in Camry's from 10- 20 seconds, and the amount of time they stop at a roundabout. Some Camry's won't proceed if another car approaches a roundabout within 250 metres, while others will not move if any car is approaching within 400 metres. Many are still there awaiting an official invitation from the Queen before proceeding. For many it is so confusing as to what they are to do at roundabouts that they have, in frustration, broken off their indicator stalks wondering what they were there for anyway.

When parking, the time it takes to find a double parking spot means endlessly going round in circles at slow speeds, plus the 15 attempts (on the rare occasion an attempt is ever made) to park it squarely between the lines.

All this affects fuel economy and thus vary from one Camry to the next.

Our 2.2L Camry gave excellent economy on the open road, around 8.5L/100 Kms (by the standards of the day; or reality ... :iconwink: :lol, even fully loaded and running the air-con most of the time. Also an AT. Around town it was as bad as my Forester ... And the Forester is a good deal heavier; faster; more powerful (as in more torque); a far less aerodynamic body; etc; etc. Now the Camry engine was excellent design for its day (1994 model), but couldn't even start to compete with the sorts of technologies built in to the SII SG engine.

 

[guzzla]

I don't know about you, but I am totally addicted to the turbo rush! I love the sound of the turbo spooling up and how the car becomes totally manic as the power comes on. It's a case on hold on, this is going to be FUN and you have to manhandle all this crazy grunt as the car fights you the whole way while your head gets shoved back into the head rest as if you've just being kicked in the face. :biggrin:

You're sure right about that! What an addictive experience. I never tire of listening to it and even now after all these 7 years that manic devil comes out in me when I start to play and I use all the gears up some of the back mountain roads around here. That whining whooshing sound and the almost unexpected push is WOW! People comment on how quiet the engine is in normal driving and then scream when I play. And they can't get over how quiet the exhaust is. It's literally never heard and that's how I like it. And the reflashed tune has just added another layer to the whole thing.