I'm (re) reading Bell's MB book. It is straightforward, but there are certain considerations that are clear but not as straightforward as is meets the eye.

It's obviously not easy to deduce what your compression ratio is, or what you'd like it to be. It's not too difficult to find the information on the internet; that is for a stock Suby engine, but when you go piecemealing parts together, the onus is now back on you to redo the calculations.

Then, I know for a fact that there are very few, if any, i-clubbers who have CR below 8:1 or above 10:1. According to this graph in his book, it looks like many of us who already have a forced induction car have gone beyond what the graph shows. (There isn't any discussion with the graph to the effect of 'this graph is for a particular car' or 'this graph is for a car with stock internals'.)


This doesn't compute for me. My CR on my EJ257 with WRX heads is supposedly 8.6:1. That would obviously go towards explaining why I blew the ringlands on my #4, but it's my understanding that there are quite a few folks over on NASIOC who have the same setup as me who haven't broken anything in their engine.


So a couple of questions I have are: Why do the NA-to-turbo guys run such low boost? Is it because they made the decision to go with stock internals? What are some opinions of the minimum amount of block/internal strengthening that a 2.5 liter RS guy would have to accomplish in order to run above 20psi? Would forged pistons be enough? Would the block have to be sleeved?

With regards to both versions of heads that the 2.5RS comes with, I don't even know what questions to ask as far as this topic goes. Port and polish? Harder springs? I'm not asking about increasing the rev, I'm just addressing strength. I am pretty sure that the phase 1 and phase 2 RS heads both provide much higher compression than the WRX, as they should. Therefore, I know that it would be wise to select some pistons that would give lower compression, although I've heard that on some setups, you then would run into a dome-to-valve clearance problem.

Thoughts?

 

MB is a good book with good explanations of the theory behind turbocharging. THe thing to remember though, is that when it was written engine management was a lot less sophisticated than it is now. An SFC and RRFPR was usually all the difference between NA and FI engines done aftermarket. That and the distributer was turned to retard spark.

About the EJ25 and turbos-

Basically, we have to run low boost because our compression ratio is so high. If you think about, basically what a turbo does is bump your dynamic compression ratio way up. Having high static compression with boost makes the engine more prone to detonate and get pre-ignition. The other thing to remember is that an RS-T with stock compression and 7-8 pounds of boost makes WAY more torque than an 8.6:1 WRX on 16 pounds of boost, although part of that is the extra displacement. A lot of people have done things like put forged rods and pistons, and even sleeved the EJ25. Personally, I think it's much smarter and more cost-efficient to just get an EJ257 block. an EJ22T block is also cheap, especially if you get one used, but it has some minor issues with RS heads.

As far as heads go, I know of one RS-T making 400+ hp and torque to the wheels using completely stock SOHC heads. They flow fine, although I would (and will) do a port/polish, maybe deck them, and I'm adding Cobb cams and TWE springs/retainers.

 

Wow, I didn't even realize that this forum was here. Wasn't my topic appropriate to 'conversions'?

Ban, you bring up a great point in that the info in Corky's book, while probably timless in some respects, is probably something that Corky himself would tell you in more definition nowadays (MAN! I'd love to talk to that guy without distraction for a week or so.)

DYNAMIC CR. That's something I'll need to research. I understand static compression ration is something a NA car has (as well as a forced induction car prior to the turbo/sc kicking in.) Is dynamic just mean the 'adjusted' CR of the engine after the added 'displacement' of a turbo kicks in? I understand that the use of the word displacement is probably inappropriate, but I think you'll understand what I'm talking about.

What you're saying about the torque variance between an RS-t and a WRX I think is right on, but do you think that an RS-t with 4-7lbs of boost will make enough power and torque to destroy the block? (assuming no knock)

My only experience is going around a road course with an RS-t, one in which the owner couldn't figure out his EM, specifically his boost, so he couldn't run beyond 3psi and that car was as impressive as my then st4!

I guess my only point is that I know I could have easily grenaded my st4 had I been irresponsible with the knock management, but assuming no knock, I doubt that I could have tuned that VF30/UTEC to power levels that would put my stock internals in jeoplardy. I know that 1-4psi in an RS is a much larger proposition than in a 2.0, but still.

 

The conversions forum is 99% EJ20s into GC/GMs. The Aftermarket Turbo forum is 99% RS-Ts.

Dynamic CR applies to every engine. MB probably has a section (the one on cams?) that will help explain it. Basically, even NA engines are "wasting" some of the charge mixture with valve overlap, plus some minor amounts of blow-by, etc. So, running your EJ25 at a given RPM, the actual cylinder pressure at TDC (is combustion weren't taking place) isn't exactly 10 times atmospheric pressure. Switch from vacuum filling the combustion chamber to positive MAP filling it, and the dynamic CR is much higher than 10:1. typically, RS-Ts don't break because of how much boost they run or how much powe they put down, not directly anyway. Pre-ignition and detonation break motors, and both are much harder to prevent in a boosted engine that is meant for NA. If you keep the AFRs right and timing retarded, and keep EGTs whee they should be, you can run 10-12 pounds of boost with the right turbo. That pretty much requires a true standalone system to work though. With band-aids like voltage clamps, RRFPRs, and ITCs, 6-7 pounds on a TD04 or T25 isn't that dangerous, just more so than a slightly modified EJ20 would be.

 

And last I checked, you had a seriously badass wagon that you haven't been able to drive to break... why the sudden interet in EJ25s?

 

I'm still putting the wagon back together but I've got a project loyale in which I'm figuring out which direction I want to go. I'm also thinking VERY hard about buying a pre-02 RS and seeing what I couldn't do with that.

Great info. Thanks for providing me some direction.

 

Pat, I found a link which explains dynamic compression far better than I can on my own. HTH.

 

Alright, but I still don't quite understand. Dynamic obviously means that it changes when complementing conditions change. DCR, following my reading, means that the effective CR changes based on cam choice and stroke (among other things.) I guess, if any of the factors that are part of the static CR equation change, then the DCR would obviously change.

It seems as though this is becoming a theme for me lately, but essentially, the DCR is like the 'economics' of the engine. The give and take. If you give a radical cam, then you take low end because it the engine will be pushing exhaust out of the intake valve, for instance.

And I failed to mention, but of course adding forced induction to the equation must be considered. I'd imagine that that is a very critical consideration to figuring out the DCR.

OT (for my purposes) so how does variable timing affect the dynamic CR?

Back on topic, I've almost purchased a 2001 RS. As I decide how I want to build it up (if I decide to build it up. Build would most likely be towards pistons, fabricated exhaust manifold, turbo, injectors, pump, DOHC...I think) I'm taking from that article that I don't want to set, say, 9:1 or 8.6:1 as my target CR so much as I want to figure out what my dynamic compression ratio should be, right? Basically, figure backwards and forwards using the equation from that page.

Say I know that the RS stroke is X, the volume of the squish area, the volume of the cylinder. Then I figure out the VE and choose a turbo (like I'd know how to do that and don't point me to a webpage showing how to do that. You want my head to 'splode?) Have I identified most all of the critical variables that I'd need to consider for an engine build? I know that there are also some other critical but ancilliary factors to be considered like water pump size and efficiency and if I'd need an oil cooler or not, but I'm just trying to address the initial build.

Thanks. Good luck making sense of that.

 

Okay, working backwards- presumably you want to run pump gas on this potential RS-T right? Given that, I'll leap way ahead and say that on a moderate sized turbo, say 250-275whp, You'll want to be between 8:1 and 8.5:1. STis with stock VF39s are capable of this level of power, on pump gas, with 8.3:1 CR, right? Something very close to that. Anyways, yes, you're on the right track by stating horsepower goals first, and working from there, rather than picking CR and fighting an improperly sized turbo. You're ahead of the game. For reference, the motor I'm slowly building (on paper ) is EJ257 with ported/polished/cammed SOHC heads. I'm going to be aiming for 9:1 CR, because it's going to be primarily a race gas motor, and merely needs to get me to the track and back on 91 without exploding. I don't care how much power it makes on pump. This is just a "best guess" on my part; the only way to nail down the exact best CR is to build the motor, test it like the author of that article did, then tear it down and change the CR. Not worth it IMO, since the actual power output won't change drastically.

Now as far as actually calculating your dynamic CR- I really don't see the need to. Why? Because dozens and dozens of really smart people have done the work for us. Quick, when I say "Subaru race motor" what kind of specs come to mind? Low (~8:1) compression forged pistons, head work, gnarly cams, big turbo, etc... right? Basically, the blueprints for success are already drawn. No need to re-invent the wheel. In my case, I'm more concerned with finding unique hidden problems my setup will cause and solving them before the motor is assembled.

Your basic thoughts on your engine build are sound. In my case, I'll probably initally use my Borla headers and an RS-T type uppipe when I turbo the stock motor, then switch to a turbo crossmember and WRX-type manifold and uppipe when I put the new motor in. This opens up a bevy of aftermarket equipment for you.

Okay, the effects of AVCS on DCR. You need to know the basic difference between high output NA cams and high output turbo cams. If you compare the specs of Cobb or TWE's "stage 1" or equivalent to their biggest, baddest, race cams for NA, you'll the trend is to more lift, and especially more duration and valve overlap. Turns out that having both intake and exhaust valves open at the same time early in the intake stroke improves VE. In order to fully explain to you, I'd have to know a lot more about fluid dynamics, so you'll just have to take my word for it. Now, under boost, the overlap just lets some of your charge mixture escape into the exhaust and do nothing for you, so turbo cams tend to have as little overlap as possible while still having decent lift. What AVCS does, is allow your cam profiles to act like Na at low RPMs off boost, and like turbo cam profiles on boost. Are you up on your pre-calculus? Imagine two inverse sine waves. Now, shift one wave to the right until the intercepts are above the X axis. This is what the plot of valve lift vs. time looks like in 2D. The area that lies above the X axis but below both waves is where you have overlap. I'd find a link with a pretty picture, but I got the hint that your brain is full already. ) So, what AVCS does, is shift the sine wave of the intake cam to reduce or enlarge the overlap area on the graph to match the characteristics of the engine at a given load and RPM. The extra cylinder filling at low RPMs, like an NA car does, is why AVCS equipped cars have more torque down low than non-AVCS cars. They are less compromised by their cam profiles. So yes, your intuition was correct- balancing your torque curve is in some respects a zero-sum game, but variable cam timing goes a long way to fixing that.

And, here's the crux of what I personally took from that article- MBT values for a specific engine layout are fairly standard; an EJ20 is an EJ20 is an EJ20 basically, regardless of ancillary mods. Determined mostly by the characteristics of the combustion chamber, i.e. slow vs. fast burn, knock resistance from quench, etc. So why do people with different setups have different timing tables? Because they're compensating for less than ideal conditions elsewhere of course, namely octane and CR. So, ideally, if you're building an engine from scratch, you figure out where the ignition timing needs to be on all points of the timing map to have peak cylinder pressure at 14 ATDC, and you balance compression ratio with your octane needs. I base this mostly on the Northstar engine example, where they tuned it for 87 octane and only lowered CR by .3 without changing ignition timing significantly, because that's a near-static figure for that motor. Actually fairly simple in theory, just takes a lot of experience in practice.

Finally, as for water pump sizing and oil coolers and such, I'm planning on overkill in that regard. Mike@Gruppe-S gave me a good tip in another thread about inexpensive, thermostat controlled oil coolers. I can't think of a single reason not to use one, but I can think of several ways it can improve engine life. Look for every edge you can get with that sort of thing.

 

So some other things I'm finding....

It looks as though, like you're talking about, that the ONLY considerations are available octane and power. It looks as though a jump from, say 8.0 to 8.6 to 1 is only 9 hp. On the other hand, I know how much of a pain it is to be adding xylene every tank as Hawaii only had 92 octane. Every single tank I had to be 95-96 octane. I didn't really mind at the time; I just told myself it was necessary and didn't think about it much more than that.

So, I'm thinking idly about what life would be like if I were to have a lifetime free giftcard at True Value and my butler had a degree in fuel mixing. Or maybe I get completely batty and I had a fuel cell that was plumbed into the fuel line with a separate fpr.

Nah, j/k.

Seriously, I'm thinking about the overall effects of choosing high vs. low comp. High = always having to put in a buttload of xylene. Cost is obviously prohibitive. But on the other hand, I wonder if I'd be leaving something on the table if I opted for lower compression. Am I giving up anything beyond power and high fuel bills? BAN, I imagine you already picked up on this, but I'm asking on behalf of both of my Imprezas. Actually since there are currently only pipe dreams for the RS, and the low comp CPs are already paid for for the WRX, this vein of thought is completely idle...for now at least.

Lastly, I'm seeing that there are multiple ways to play with CR, static at least. Machining a few .001's off of the deck, HG thickness, piston dish, p&p'ing the combustion chamber. Money seems like the only issue.

Now for some chitchat that's a little more timely. Once I get the WRX up and running (I have a hoist now, so I should be out in the garage right now) I will have an almost entire WRX longblock to install into the RS. I think I'd just be missing injectors and a fuel pump filter. I'll have to do a search to see if anyone has cut the tips off of the end of the RS injectors. I'd also need an ECU. I'd imagine a piggyback would be out of the question because I don't know of any. Of course, I don't think I'll put the WRX engine in without some pistons.

One more option is like how you're doing. jdmeis is selling some 2.5L JE pistons and crower rods on ebay. http://cgi.ebay.com/ebaymotors/ws/eB...tem=7946875850 I am probably making a mistake by posting the ad here, but I haven't yet bid because I haven't put any thought into turboing the (what's the RS's engine code? EJ251?)

I wonder if I could just utilize my stock WRX exhaust manifold and a turboback, injectors and a fuel pump?

 

This is what I'm going to do, for two reasons- one, less manifold volume is good in FI applications, and two) it allows for a larger topmount mounted more properly under the hood scoop. I've been researching this extensively lately, and I've figured out that if you have a standalone like a Hydra, it WILL work. If you have stock management, the issues are making the RS MAP sensor work, and wether or not the WRX and RS TPS voltage ranges are the same (should be). Beyond that, it's typical Subaru hot-swappable. I'll go over all the details as soon as I get mine on.

 

This is a great thread. I wish I had something monumental to add but... I really was needing all this information. Thanks and good luck with the builds.

cheers

garrett

 

I split out the NA discussion fellas.

 

You're absolutely right. Sorry.

 

Eh, it was cool for the first question or two. Then we started gabbing about it like girls in a junior high hallway.