The H6 Impreza Resource Thread

[Storm]

If anyone is looking for a medium-low milage (86k) EZ30D, PM me. I'm going to go a different direction with my Forester and won't be using this motor. EJ25 trades welcome.....


Jay

* Registered users of the site do not see these ads.

[Marnix]

Originally Posted by Patrick Olsen

Anyone have the '01-04 EZ30 FSM? I have a little bit of it that I downloaded 5 years ago (I'm not really sure why I would have done so back then), but it would be nice to have all of it. Obviously I can sign up for the Subaru Techinfo site, but I figured I would check here first.

Pat

I have all the '01-'04 wiring diagrams in PDF form. Drop me a PM if that's what you're after.

[Patrick Olsen]

Originally Posted by Marnix

I have all the '01-'04 wiring diagrams in PDF form. Drop me a PM if that's what you're after.

PM sent.

Originally Posted by dragoontwo

Care to elaborate on this?

Sure. I sent Delta my heads and cams in March 2011 to be ported and get the cams reground. I had previously discussed my project with Ken (when I was living out there in WA), then talked with him again on the phone prior to shipping the parts. It is now January 2013 and I don't have them back. Upon getting the heads, Ken said he wasn't willing to do the weld & grind route on the cams since they're gun-drilled (and thus I guess could be warped out of line by the heat involved in the process?). So, he planned to grind them to a smaller base circle. OK, got it. Then it took about 6 months to find someone to supply longer valves with +1mm heads. Meanwhile, it was taking forever to get the heads ported by whoever it is does their porting for them (since Delta doesn't do that in house). They didn't get any "before" flow data, as I had specifically asked , so God knows what the porting has done. After a year or more they were having trouble sourcing springs; I told Ken that Supertech has them in their catalog , so I ordered the springs and had them sent to Delta. Then apparently they can only install one spring about every 20 days or so , because every time I called I was told that they "just needed to finish putting things together so they could spec the cams" or something along those lines. After about 20 months, Ken decided he wasn't willing to grind the cams to a smaller base circle because there's not enough information about the AVCS lifters to know whether they would work right and he didn't want an issue with that to ruin the heads. I understand his misgivings on that, but I'm not sure why it took so long to come to that conclusion.

Now they're supposed to be putting the ported heads together with the stock cams and the +1mm valves (which will have to be trimmed, since the longer stems are no longer needed with the stock base circle cams). Of course, that's been the status for 6 or 8 weeks now.

[Marnix]

The lifters should be fine with 11 mm lift cams, that is what Porsche is also running on the same lifters.

[Patrick Olsen]

Originally Posted by Marnix

The lifters should be fine with 11 mm lift cams, that is what Porsche is also running on the same lifters.

Hmm, interesting. Can you provide some more details? Which Porsche motor are we talking about, and they no kidding use the exact same lifter setup?

[Baldturbofreak]

^^ pretty sure I read that subaru licenced said technology from porsche. I think in the Perrin thread it mentions that it says "Made auf Deutchland" on the bottom of said lifter.

That sucks to hear about Delta, sounds like they are stuck on the easy money to keep the doors open. Guess I'll send my spare Eg33 cams to Webb if I cant source a place for billet cores. (suppose I could cad/cnc them myself in spare time. lol)

[dragoontwo]

Originally Posted by Baldturbofreak

^^ pretty sure I read that subaru licenced said technology from porsche. I think in the Perrin thread it mentions that it says "Made auf Deutchland" on the bottom of said lifter.

That sucks to hear about Delta, sounds like they are stuck on the easy money to keep the doors open. Guess I'll send my spare Eg33 cams to Webb if I cant source a place for billet cores. (suppose I could cad/cnc them myself in spare time. lol)

While it's not billet cores, there is a company in Poland that makes new cams for the EG33. The only person they will deal with is member SilverSpear on the SVX forum. The post concerning that is here. They aren't exactly cheap, but they are new 42CrMo4 cams with tho option to be rifle drilled for extra.

Patrick, are you doing an N/A build or turbo? It sucks to hear about your experience with Delta Cams.

[Marnix]

EG cams do not fit EZ cams.

The lifters are exactly the same as what Porsche uses on the 997 engines, it even has a Porsche partnumber on it:















According to Wikipedia, these Porsche engines use the Variocam Plus system:

2006-2008 3.596 litre 325 hp water-cooled flat-6 (VarioCam Plus) (911 997 pre-facelift Carrera series / Targa 4)
2006-2008 3.824 litre 355 hp water-cooled flat-6 (VarioCam Plus) (911 997 pre-facelift Carrera S series / Targa 4S)
Carrera S powerkit: 3.824 litre 385 hp water-cooled flat-6 (VarioCam Plus)
2007-2009 3.600 litre 480 hp water-cooled flat-6 (bi-turbocharged) (VarioCam Plus) (911 997 pre-facelift Turbo)
2008-2010 3.600 litre 530 hp water-cooled flat-6 (VarioCam Plus) (911 997 pre-facelift GT2)
2008-present 3.614 litre 345 hp water-cooled flat-6 (VarioCam Plus and DFI) (911 997 facelift Carrera series / Targa 4)
2008-present 3.800 litre 385 hp water-cooled flat-6 (VarioCam Plus and DFI) (911 997 facelift Carrera S series / Targa 4S)
2010-present 3.800 litre 530 hp water-cooled flat-6 (bi-turbocharged) (VarioCam Plus) (911 997 facelift Turbo S)
2010-present 3.600 litre 620 hp water-cooled flat-6 (bi-turbocharged) (VarioCam Plus) (911 997 facelift GT2 RS)

So if you can find the valve lift or cam lobe height of those engines you know what the lifters can handle. For sure you should upgrade the valvesprings when running more lift.

[dragoontwo]

Originally Posted by Marnix

EG cams do not fit EZ cams.

If this was directed at me, I know this. I'm an SVX owner with an EZ30R torn apart in my garage.

[Patrick Olsen]

Originally Posted by dragoontwo

Patrick, are you doing an N/A build or turbo? It sucks to hear about your experience with Delta Cams.

For the most part I've been planning to just do a N/A build, but I occasionally waver and think about doing a low pressure turbo setup of some sort.

Interestingly enough, I spoke with John (I think) @ Outfront Motorsports a couple weeks back. They use the stock EZ30R cams, even with their big power turbo builds.

Originally Posted by dragoontwo

If this was directed at me, I know this. I'm an SVX owner with an EZ30R torn apart in my garage.

I think Marnix may have missed that you were commenting on two different discussions running through this thread in recent days - Baldturbofreak's comments about EG33 cams and my comments about EZ30R cams. Maybe we should mix in some ER27 discussion, too, just to really confuse this thread!

Originally Posted by Marnix

The lifters are exactly the same as what Porsche uses on the 997 engines, it even has a Porsche partnumber on it:






According to Wikipedia, these Porsche engines use the Variocam Plus system:

2006-2008 3.596 litre 325 hp water-cooled flat-6 (VarioCam Plus) (911 997 pre-facelift Carrera series / Targa 4)
2006-2008 3.824 litre 355 hp water-cooled flat-6 (VarioCam Plus) (911 997 pre-facelift Carrera S series / Targa 4S)
Carrera S powerkit: 3.824 litre 385 hp water-cooled flat-6 (VarioCam Plus)
2007-2009 3.600 litre 480 hp water-cooled flat-6 (bi-turbocharged) (VarioCam Plus) (911 997 pre-facelift Turbo)
2008-2010 3.600 litre 530 hp water-cooled flat-6 (VarioCam Plus) (911 997 pre-facelift GT2)
2008-present 3.614 litre 345 hp water-cooled flat-6 (VarioCam Plus and DFI) (911 997 facelift Carrera series / Targa 4)
2008-present 3.800 litre 385 hp water-cooled flat-6 (VarioCam Plus and DFI) (911 997 facelift Carrera S series / Targa 4S)
2010-present 3.800 litre 530 hp water-cooled flat-6 (bi-turbocharged) (VarioCam Plus) (911 997 facelift Turbo S)
2010-present 3.600 litre 620 hp water-cooled flat-6 (bi-turbocharged) (VarioCam Plus) (911 997 facelift GT2 RS)

So if you can find the valve lift or cam lobe height of those engines you know what the lifters can handle. For sure you should upgrade the valvesprings when running more lift.

Thanks very much for the info and the pics, Marnix, particularly the ones showing the lifter actually in operation. Part of my problem in figuring this out is that I never really played with the heads at all myself - once I had them off the engine, they got boxed up and sent out to Delta. I've never even pulled one of the lifters out to look at how it works.

When it comes to using different cams with a smaller base circle and higher lift, in my mind there are two issues:

– First, the lifter has to be able to span the difference in lobe heights between the low and high lift lobes. So, hypothetically (not having the FSM in front of me here at work), if the stock lobes were 4mm lift and 10mm lift, and the new lobes were 5mm and 13mm, that center portion of the lifter would have to “reach” 2mm farther to maintain contact with the low lift lobe. While I don’t know what Ken @ Delta would plan on for lobe heights, looking at the first and second pictures I’ve quoted, I’d say there’s no issue there – looks like the lifter can cover a pretty big spread in lobe heights, much bigger than I would ever expect to actually run.

– Second, when the valve is shut, with a reduced base circle, the lifter will need to rise up out of the lifter bore higher to maintain contact with the camshaft. There’s a danger that the lifter could become unstable if it is pushed too far out of the bore. I don’t know at what point that would become a concern.

Does that make sense to others? I’m just trying to think this through a bit, and again without having the heads here to play with it’s hard for me to visualize.

[Marnix]

The lifter being higher up its bore with either a longer valve or a thicker shim would not be a concern. It won't rock in the bore.

As long as the low lobe and high lobe are increased by the same amount (ie. the difference in base circle stays the same) then there is no issue. My feeling says that a bit bigger difference would be acceptable, if the base circle is almost the same I guess there would be difficulties with the system switching from low to high lobe (then there would not be enough 'time' to push the plunger in the bucket to lock the bucket)..

[Patrick Olsen]

Originally Posted by Marnix

As long as the low lobe and high lobe are increased by the same amount (ie. the difference in base circle stays the same) then there is no issue. My feeling says that a bit bigger difference would be acceptable, if the base circle is almost the same I guess there would be difficulties with the system switching from low to high lobe (then there would not be enough 'time' to push the plunger in the bucket to lock the bucket).

OK, at home now, so I can look at the FSM specs.

Cam base circle diameter: High - 32mm; Low - 31.84mm
Cam lobe height: High - 42.14mm; Low - 38.19mm or 35.49mm

You raised an interesting point that I hadn't thought about - ensuring there's enough time for the lifter to lock into the high lift mode. So if you increased the duration of the high lift lobe by a lot, but didn't really touch the low lift lobe duration, the two lobes might not have enough time on the back of the lobe to get the plunger to be pushed back in and allow the lifter to lock into high lift.

And the factory specs do have a little bit of a difference in the base circles, I suppose just to keep the low lift lobe from riding on the lifter when it doesn't need to.

[Patrick Olsen]

I talked with Ken @ Delta last night, and forwarded him the pics that Marnix had posted. In addition the concern about how the lifters actually function (which I think will be addressed with the pictures), the other issue is that Ken didn't think they could cut the center lobe. Basically, they don't have a cutter that would fit between the two high lift lobes. If they can't grind the center lobe to a smaller base circle, then they can't grind the high lift lobe down to a smaller base circle, either. He said he would take another look at things to see if they could figure something out.

In other news, I picked up a cheap EZ30D. Part of me wants to do the "easy" swap and put it in the car with the factory ECU, and part of me just wanted to pull it apart to compare it to the EZ30R I've already got.

I had been under the impression in the past that the two EZ30 variants used the same short block, that it was just the heads and cam stuff that had changed. At some point somebody pointed out that the short blocks do have different part numbers, but that doesn't necessarily mean there are functional differences. Well, even without pulling the heads yet, I've already determined the short blocks are different - the EZ30D has just a single knock sensor (on the passenger side of the block), while the EZ30R has a knock sensor on both block halves. You can see that the mounting boss for the knock sensor isn't there on the EZ30D, so the casting of the block half has been changed.

Some other items of note thus far...
- The timing chain setup is essentially identical. The only difference (as far as I can see) is the cam sprockets. I knew the intake ones would be different, as the 30D doesn't have AVCS, of course. The exhaust sprockets are also different. The driver side intake and exhaust sprockets are actually the same, but with markings for both (for chain alignment). I first saw the exhaust marking on the intake sprocket, which made me go , but then I saw both sprockets had dual markings.
- The first gen engine has an interesting coolant cross-over that goes from the back of each head and crosses over underneath the intake manifold. The EZ30R doesn't have anything like that; the AVLS solenoids occupy that space on the back of the R heads. It seems strange to me that such a thing would be necessary, as there's already a coolant cross-over at the front of the engine.
- The 30D front timing cover has 60 () bolts holding it in place, as opposed to only 40 on the 30R.
- Rear timing covers are pretty similar - 45 bolts on the 30R, 46 on the 30D. In both cases, there are 7 different types of fasteners holding on the rear timing cover.
- The 30D has separate components for the oil pressure regulator and the oil pump cover, while the 30R has that all combined into 1 piece.
- The 30D oil pump has 9 teeth on the inner rotor, and is 11mm thick. The 30R oil pump has 7 teeth on the inner rotor, and is 13mm thick.

This 30D is filthy compared to the 30R I've got. Other than some typical carbon deposits in the combustion chambers, the 30R is spotless. This 30D, on the other hand, has serious oil staining and crud build-up. Here's the center of the timing chain assemblies for the two engines:





(You can also see the 1-piece oil pump cover + oil pressure regulator on the 30R vs. the 2 separate pieces on the 30D.)

And the inside of the front timing cover:


I plan to get the heads off tonight or at some point this weekend, at which point I'll have the 2 short blocks sitting in my garage to compare.

[dragoontwo]

Can you check the intake manifold compatibility between them just for fun if you remember?

[Marnix]

At least the mating surface/bolt pattern for the manifold is the same for both engines.

The EZ30D and EZ30R have different oil- and coolant passages on the heads so you can't swap heads. Almost nothing is interchangeable within the block, only the rods are the same.

My EZ30D engines (and wiring diagrams) show two knock sensors...

[Patrick Olsen]

Originally Posted by Marnix

My EZ30D engines (and wiring diagrams) show two knock sensors...

I stand corrected. I took another look and one of the mounting bosses on the driver side is clean, and there's been something cut off the wiring harness. So, it would appear that knock sensor was chopped off prior to my getting the engine.

Originally Posted by dragoontwo

Can you check the intake manifold compatibility between them just for fun if you remember?

I only have the EZ30D heads, so I can't do the swap you want me to do - EZ30D intake manifold on EZ30R engine.

As I said before, I tried that combination a few years back and it didn't work. As I look at the bits and pieces in my house, though, I think I remember what the issue is - the fuel rails. (By that I mean the lines that run from one side of the engine to the other, not the actual rails that mount on top of the fuel injectors.) The 30R fuel rails mount to the short block along with the wiring harness. The low-hanging humps on the bottom of the 30D intake manifold will interfere with the 30R fuel rails. The 30D fuel rails across the intake manifold in front of those low-hanging humps. The obvious answer would be to use the 30D fuel rails.

It may also be the coolant hard lines that were in the way - you can see those as well in the picture I linked above. As previously discussed in this thread, I seem to have a misfit engine when it comes to those hard lines, so maybe on a "normal" EZ30R those hard lines wouldn't be an issue?

[Marnix]

Maybe this thread should be separated in an EG33-H6 thread and an EZ30-H6 thread as there's a ton of information for both swaps in here but it can get confusing and7or hard to find information someone's looking for.

[Patrick Olsen]

Originally Posted by Patrick Olsen

Some other items of note thus far...
- The first gen engine has an interesting coolant cross-over that goes from the back of each head and crosses over underneath the intake manifold. The EZ30R doesn't have anything like that; the AVLS solenoids occupy that space on the back of the R heads. It seems strange to me that such a thing would be necessary, as there's already a coolant cross-over at the front of the engine.

Turns out the "interesting coolant cross-over" isn't for coolant at all - it's actually part of the crankcase ventilation setup. The passenger side head has a port at the top of it that connects to the crankcase. That cross-over connects the two heads; the driver side head does not have a direct connection to the crankcase. However, it does have ports in the front cam bearing cap that connect the head to the air volume contained between the front and rear timing chain covers, and then there's an opening at the bottom of the rear timing chain cover into the top of the oil pan.

As a result of the differences in the head configuration, the 30D head gaskets are different on each side of the engine, while the 30R head gaskets are the same on both heads. Based on my limited experience with engine assembly (a couple EJ25s, a SBF, and an LS6), having different head gaskets on a V or H engine is not common.

I also found something that mystified me, and I posted about it on another forum. The fuel injector ports on the heads are all joined together by a header. The headers on each head are connected by a cross-over, and that cross-over then connects to the throttle body. Pictures to explain...

Here's a shot showing the passenger side head, with the fuel rails and fuel injectors just placed in their normal position (but obviously not connected or bolted in place). Fuel supply and return are at the rear of the head (left side of the picture), and the cross-over I referred to is at the front.


Zooming in, you can see that where the fuel injector sits there is a port, and that port ties into a header that runs along the head, connecting all 3 injectors.


A cross-over connects the two heads together, and also runs to the throttle body:


Here's a shot of the cross-over going over to the driver side head:


A blurry shot of the TB. The hose from that cross-over connects to the nipple at the top of the TB. The black solenoid thingy on the right is (AFAIK) the IACV. The nipple connects to either side of the throttle plate.


So, what would this system be? The only thing I could figure is that it provides an atmospheric pressure reference for the fuel injectors, but I don't know why that would be required. As you can see in the driver side picture above, there's a normal fuel pressure regulator at the front of the driver side fuel rail, with the vacuum reference hose tapping into the intake manifold somewhere underneath. On the other few engines I've messed with, that's all there is, just that vacuum-referenced fuel pressure regulator.

One of the guys on the other forum replied:

The Germans used to run air-shrouded injectors tips for emissions reasons. There would be a small air channel to each injector that was routed to pre-throttle body air pressures. When the intake manifold and ports are under vacuum, air rushes in through the air channel and then into the port, rushing out around the tip of the injector. The rush of air atomizes the fuel better, making more of it burnable and cleaner emissions.

Apparently some Honda engines used a similar setup, called a Fuel Injection Air Control System. Interesting stuff!

[OrbitalEllipses]

On my EJ205, there was a coolant hose that went to the throttle body near the IACV just like that. It was used to prevent the TB butterfly from sticking shut/freezing in the cold as the coolant acted to warm it up.

[Storm]

Likely similar to the air assist system on the EJ251?


Jay

[Patrick Olsen]

Originally Posted by OrbitalEllipses

On my EJ205, there was a coolant hose that went to the throttle body near the IACV just like that. It was used to prevent the TB butterfly from sticking shut/freezing in the cold as the coolant acted to warm it up.

This isn't that.

Originally Posted by Storm

Likely similar to the air assist system on the EJ251?

Could be - I don't have any experience with those. It wouldn't surprise me if EJ25s from the same timeframe have a similar setup.

[Marnix]

In the parts catalogue it is described as a vacuum system. Maybe to draw vapour from the injectors? It does not seem to be connected to the pressure side of the injectors.

[Patrick Olsen]

Did a bit more work on a whim a few nights ago. I was going to start pulling the piston pins, but I need to get some long-ass needle nose pliers, and by the time I wandered out to the garage it was about time for the stores to be closing. So, instead I decided to pull off the oil pan.

As seen in earlier pictures, the oil pan is two parts. There's a large, cast aluminum piece that bolts to the block and is really a structural part of the block (with oil and coolant running through it, the water pump mounted to it, etc). Then on the bottom there's a stamped steel piece that bolts to the cast aluminum piece.

Here's the "lower" oil pan. The circular bit in my hand is a magnet that sits on the bottom of the pan (around that bump you can see):


Here's the "upper" oil pan, looking up from below. The big manifold running north-south (which is actually side-to-side on the engine) is for coolant. The oil pump pickup feeds the port that's at 12 o'clock.


And here's the upper pan looking down from above. The flash reflecting off that center flat portion kind of ruined the picture - I'll have to try and get a better picture. You can see the two big coolant ports, and the kind of J-shaped port is where the oil pickup feeds. I don't understand all the oil flow paths, so I need to pull out the rear timing chain cover and and oil pump so I can map it out in my brain.


And here's the bottom of the block with the pan removed:


I just bought an '01 ECU on Ebay. It was a bit more expensive than others I've seen, but this one includes the harness plug and about an 8" pigtail, which I figured would be very good to have.

Has anyone ever seen the factory exhaust manifolds advertised for sale anywhere, either for the 30D or 30R? I never have. My guess is that the scrappers just chop the cats out of the manifolds because they can get good money for the cats. I don't know that I want to use them, but I would like to see some pictures of them, just to know how they're configured.

[zephyrus]

Does anyone know what the diameter of the ez30d throttle body is?

[Patrick Olsen]

Inlet is ~76.5mm, tapering down to ~66mm at the throttle blade and at the backside of the TB.