EG33 build for high boost stage 3 supercharger

[longassname]

The stage 3 supercharger system comes set up to run well on a stock svx engine but we can rebuild the EG33 engine to optimize it to run higher ratios of atmoshperic pressure on street gas. Namely, we need new pistons. In this post I'll be laying out the details of the Eg33 combustion chamber along with photos and explaining the details of what makes a good piston for forced induction.

There has been very little information made public about the EG33 engine's combustion chamber to date and what little information has been publicly available thus far has been confusing. The threads talking about pistons for the EG33 thus far have focussed on whether the 2.2L turbo pistons will work or not and what the compression ratio with them would be. Trying to compare the #'s estimated for the pistons with the #'s estimated for the head they just didn't add up. This turns out to be because the SVX uses a thicker head gasket and the pistons extend "out of the hole." Because the SVX pistons extend "out of the hole" and the ej22 pistons do not we can forget about using them in the svx as they would very negatively effect the performance of the squish area. I'll explain more about squish later on.


The SVX combustion is relatively large. 11cc larger than that of the Ej22 engines which have the same bore, stroke, and thus per cyllinder displacement as the EG33. Here are a couple of photos from both angles.
http://www.ecutune.com/posts/eg33/DSC_2380.jpg
http://www.ecutune.com/posts/eg33/DSC_2382.jpg

Before getting things messy I made a rubbing of the squish area on the head to be used in designing the matching squish area on the pistons that we will be having made. To do this I put a piece of paper over the combustion chamber and then put the head gasket over the paper and lined it up. I then used my finger to drag carbon from the edge of the head gasket to make the rubbing. I then traced the outline and put in measurements taken with a pair of calipers directly on the metal. The squish area is the area inside of the bore of the head gasket but outside of the combustion chamber. When the piston comes up to top dead center it squishes some of air fuel charge inside of this area and creates turbulence which mixes the fuel charge with the air charge inside of the combustion chamber--a feature which is important to the efficiency of combustion. You want to shape the top of the piston to create a matching squish area to the head. At top dead center the piston should be as close to the head as tolerances for expansion, piston rock, and carbon build up will allow. This prevents the air fuel charge inside of the squish area from igniting keeping the edges of the piston cool to prevent detonation and failure while maximizing the heat generated in the center of the piston early on to create the most power. Squish area is very important.
http://www.ecutune.com/posts/eg33/DSC_2384.jpg



I'm continueing with this write up but splitting it up so that I don't get any errors while posting. Please don't post anything in the middle of my write up while I'm continueing to type the next post.

[longassname]

We need an acurrate measurement of the volume of the combustion chamber in the head so I did what is called "cc'ing the head." I wiped a thin layer of wheel bearing grease accross the mating surface of the head surrounding combustion chamber to create a seal and then layed a piece of plastic over the combustion chamber with two small holes to allow for filling and air to escape. I then used a 10cc syringe to inject mineral oil into the combustion chamber in order to measure it's volume. The volume of the SVX combustion chamber is 51cc.
http://www.ecutune.com/posts/eg33/DSC_2385.jpg

I then turned the crank on the engine to advance the piston in one of the cyllinders to tdc. It was obviously "out of the hole" meaning it extends beyond the cyllinder wall and occupies some of the volume inside of the head gasket. Beyond the effect this has on the compression ratio this is a very important measurement as it greatly effects the squish area. Using my calipers I measured how far out of the hole the piston extends. I took the measurement along the axis of the pin which is important as a cold piston rocks. You can see in the follwoing picture what I mean as the bottom of the piston extend further out of the hole than the top of the piston. The correct measurement is made in the middle along the axis of the pin. The piston extends out 0.018" out of the hole. I then measured the thickness of the used head gasket where it had been compressed by the cyllinder wall. The height of the compressed head gasket is 0.054". Subtracting the out of bore height of the piston from the height of the head gasket gives the height of the squish area which is .036". A little suprising for a factory engine but then this is Subaru's best. .036" is what you would expect to find in a performance engine. Some builders might go as low as .030" others might go as high as .040" On a factory engine I wouldn't have been suprised to find .060"
http://www.ecutune.com/posts/eg33/DSC_2387.jpg

Now we need to know the volume of the entire combustion chamber from which we can also determine the volume of the piston if we do it right. In order to do this I cleaned up the cyllinder to match the combustion chamber I had already cleaned up. I then wiped the edges of the cyllinder with wheel bearing grease and advance the piston to top dead center and wiped off the excess grease. Doing this filled up the gap between the piston and the cyllinder wall above the 1st ring. I then reaffxed the cyllinder head, turned the engine on it's side, and filled the combustion chamber with mineral oil using the 10cc syringe until the oil reached the begining of the spark plug threads. The total combustion chamber volume is 61cc.
http://www.ecutune.com/posts/eg33/DSC_2388.jpg
http://www.ecutune.com/posts/eg33/DSC_2389.jpg
http://www.ecutune.com/posts/eg33/DSC_2390.jpg

Again I am continueing this write up but breaking it up into multiple posts to avoid errors with the website. Please do not post in the middle of my write up while I am typing the next section.

[longassname]

Ok now we have all the #'s we need to calculate what we need to do to make pistons to achieve the desired compression ratio with the squish design we want.

bore: 96.9mm
stroke 75mm
volume at tdc 61cc
combustion chamber volume: 51cc
piston volume: 61-51-10.5-3.37= -3.87cc
head gasket thickness: .054" = 1.3716mm
head gasket bore: 3.888" = 98.755mm
head gasket volume: pi * r squared * h = 10.5cc
piston to deck height: -.018"

--> factory compression of 9.94:1

we want a 9:1 compression ratio for our build so we want our pistons to have a dish of 11.1cc

[oab_au]

Ah, I was farly close. :)

A bit hard to measure off the picts, but if 0.5mm was machined off the top of the lump, it would drop to 8.9:1. This assumes the lump is 77mm X 25mm??????

Harvey. ;)

[longassname]

We are getting forged pistons made by JE.

[TomsSVX]

Do let us know how much they are gonna run you. Would also be great if they were available to the public afterwards... I love the progress you are making

Tom

[longassname]

Thanks,

Somebody had to do it. As it turned out I guess it was probably beyond the abilities of the average enthusiast anyway.

[TomsSVX]

Being as short funded as I am right now, pimpin aint easy :rolleyes: Anyway yeah it is great to hear someone took the initiative to get this done properly.

Tom

[SVXtasy]

this is so cool. i love how people are doing stuff rather then talking about it. thanks mike.

[longassname]

http://www.subaru-svx.net/forum/attachment.php?attachmentid=6656&d=1135706924
Thanks to some measurements Tom has taken of the dome on the SVX piston I was able to calculate the volume of the dome.

The cross section of the dome is an isosceles trapezoid, the sides of which are made up of the valve cut outs. By Tom's measurements with a digital micrometer the height of this trapezoid is .088 centimeters. The top of the trapezoid, a, is 2.9 centimeters. The base of the trapezoid, b, is 3.45 centimeters. The formula for the area of the trapezoid is A= 1/2 (a+b) h. This gives us an area for the cross section of the dome of .2794 squared centimeters.

The short sides of the dome are curved and also angled. Tom took all 4 measures and I averaged them to get a length of 7.688 centimeters. This is good enough. I could calculate the area of the curved section since we have all the measures to give us the radius of the unit circle and the length of the cord for the section but we really don't need to bother. The difference between doing the real math and averaging the 4 measures is insignificant and it's a lot of math.

Multiplying the length of 7.688 centimeters by the area of the cross section 0.2794 centimeters gives us the volume of the dome which is 2.15 cc's. Take that with the volume measurements I made earlier and we see that our valve reliefs have a volume of 6 cc's.

[longassname]

Tom, to answer your question about what machining the dome off of the piston would give you for a compression ratio it would be 9.6

[longassname]

Ok, we don't have the final design back from JE yet but we are going with a piston for a 3.818 bore. This bore gives us propper ring fit with the JE rings while allowing any svx block that hasn't been damaged to be easily and propperly honed. From the factory the blocks have a bore between 3.8144 and 3.8155.

Since people are beginning to do head work and cam work on these cars we are increasing the size of the valve reliefs to accomodate bigger valves and more lift. We will try to get our 5 cc's of extra volume beyond removing the dome there.

[TomsSVX]

Quick Q. Are they simply machining old pistons or are they making all new ones

Tom

[longassname]

We are having JE forged pistons made.

[TomsSVX]

Thnx

Tom

[longassname]

i looked at the specs in the factory service manuals for the eg33 and ej22 and determined that they use the same main bearings just in different #'s. Clevite makes bearings for the ej22 so I ordered 2 sets of those to get the bearings for the eg33 build. I also got clevite rod bearings.

[TomsSVX]

i looked at the specs in the factory service manuals for the eg33 and ej22 and determined that they use the same main bearings just in different #'s. Clevite makes bearings for the ej22 so I ordered 2 sets of those to get the bearings for the eg33 build. I also got clevite rod bearings.

Keep pouring on the good news...

Tom

[SVXRide]

i looked at the specs in the factory service manuals for the eg33 and ej22 and determined that they use the same main bearings just in different #'s. Clevite makes bearings for the ej22 so I ordered 2 sets of those to get the bearings for the eg33 build. I also got clevite rod bearings.


Michael,
Great news! How much did the full set of bearings end up costing you?
-Bill

[want-a-fast-svx]

this is awesome news since my engine definitely has crank walk ;( Well least when i have some money i can find the right parts to build the engine right. Cant wait to hear more, really appreciate all the great work lan.

[Tycho]

Question: There are two part numbers listed for the rod bearings (one for EG33s mfr'd pre 09/92 and one for engines after that date). What changed? I have what I believe to be a 92 engine and I've mic'd out all the journals (rod and main) just for kicks. Would you happen to have journal diameters for a post 92 engine?

Cometic makes MLS EG33 head gaskets BTW. I will be getting a pair myself sooner or later.

[SVXRide]

Question: There are two part numbers listed for the rod bearings (one for EG33s mfr'd pre 09/92 and one for engines after that date). What changed? I have what I believe to be a 92 engine and I've mic'd out all the journals (rod and main) just for kicks. Would you happen to have journal diameters for a post 92 engine?

Cometic makes MLS EG33 head gaskets BTW. I will be getting a pair myself sooner or later.

Tycho,
So, how's your EG33 rebuild going? I don't believe there should be anything different about the dimensions for any aspect of the EG33 over the years it was manufactured. Subaru may have just changed suppliers, thus the change in part number...
-Bill

[Tycho]

Oh...it's coming along swimmingly.

The parts actually sat in a box all summer while we tinkered with other projects. I just got cracking on it again over break a couple weeks ago (getting the deck surfaces milled and freshening the cylinders with a hone).

We've only recently decided how we're going to build the engine and what it will go in. Looking like basically a stock rebuild (skosh higher CR) with some good headgaskets and some of my headwork. Have some solid lifter parts also, but we probably won't sweat that (or cams). Induction options range from a low boost super quick spooling turbo setup to a N2O setup to ITBs. We'll decide that later. I recently had a very positive experience with ITBs on a 4 cylinder motor of mine as far as engine response was concerned. The guy I'm working with on this has a lot of experience with N2O on SBC/BBC combinations that make some pretty stout numbers.

[longassname]

Well we have our subaru engine overhaul seal and gasket set and our clevite bearings. The pistons aren't done yet but should be around the time we are done using the engine for fabricating the 1st run of stage 3 kits.

[svxsubaru1]

Great job :).