3.0 twin plug MFI teardown

[speedo]

I am down to the short block and the reason the engine is apart is that the PO was frustrated with the engine after it was diagnosed with broken exhaust side head studs...repairs were made, but the engine never ran right afterwards. As I tore it down...the throttle bodies and intakes were opened to 40mm, the exhaust to 38mm. The MFI pump looks to be a stock 2.2 S injector pump. The cams are a turbo 4 bearing core ground to a Andial "modified S " profile. The case is a 930 case with a 70.4 crank. The heads have been flycut 3.5 mms to reduce the cc's of the combustion chamber. Everything looked fine until I took the cylinders off the pistons. Scoring on both at the 12 and 6 o'clock positions. I initially thought the scoring was due to detonation and poor cam timing. Now I am thinking the problem was the use of the Cosworth pistons in Alusil cylinders. The scoring looked like overheating due to detonation...but it also resenbles what might have occured if the "K" Alusil cylinders were used with the aluminum Cosworth pistons instead of Nicasil cylinders. Questions...anybody use the 95mmm Cosworth pistons back in 1990s...and were the ring lands the same as the Mahle RSR pistons. Were the pistons designed to be used with the Nicasil cylinders?
I have asked around to try to find the shop that might have been building these engines off stock 3.0 CIS engines, but no luck. The build list looks great...with the exception of the P and C combo. Help would be appreciated. I am looking for a twin plug 3.0 dizzie as the Electromotive HPV-1 this came with is going away.

Thanks

Speedo

[lindy 911]

I did a little research before building my 3.0 twin-plug and found that the Alusil cylinders need a special piston AND rings. There are metallurgy requirements to make the combination work. I chose to use JE with Mahle cylinders because of the nickle plating or Nikasil. Alusil is a completely different animal than Nikasil. The JE pistons and rings are designed to run in the nickle bore and the make up of the aluminum in the JE is designed to expand and contract similar to the Mahle cylinders. The scoring at noon and six would be skirts getting to big for the cylinder bore.

I don't think detonation is the cause of this. Detonation will break parts before it heats them to this point. My experience is that the piston crown, head and con-rod bearings are gone way before the piston has a chance to heat to the point of seizure. I definitely think new piston and cylinders are on you build sheet.

There are some fellas here that can give you a much better explanation than me but that's the skinny of it.

FWIW, I'm using BURN-BROS twin dizzy and MSD 6AL with twin coils. I looked at cheaper stuff but came back to this.

Lindy

[jpnovak]

You are correct that you can not mix and match between Alusil and Nikasil.

Nikasil is commonly regarded as a high silicon content aluminum. This is a hard material and requires rings to match. Relative to the cylinder walls the pistons are soft.

Alusil is commonly described as soft relative to the Fe piston coating.

Hard Nikasil rings create problems on soft Alusil cylinders.

If the pistons spec out then source some new 95mm Nikasil cylinders, re-ring and have fun with a healthy new motor.

[Henry Schmidt]

I think we're seeing a lightly confusing description of the difference between Nikasil and Alusil cylinders.
The aluminum alloy used to produce Alusil is actually a very high silicon alloy and is very hard. .
Much higher than the silicon content in Nikasil cylinders.
Imagine if you can a cylinder wall that is lined with little beads of glass. Than is in essence how Alusil works. The silicon beads (glass) in the alloy are isolated via an etching process that removes aluminum from around the beads and then the beads are honed to produce a relatively smooth surface.
The Nikasil cylinder is actually a lower silicon content alloy (softer) that has a plating of high nickel chrome.
When parts of similar material operate in a friction environment the materials tend to stick (gall) together.
With Nikasil you have chrome in contact with aluminum where with Alusil you have aluminum in contact be it slight with aluminum so to prevent galling the piston in the Alusil cylinder is plated with iron thus producing dissimilar material.

[jcge]

Alusil is a hypereutectic alloy (AlSi17Cu4Mg or 390 alloy) in which silicon crystals are uniformly dispersed throughout the material. The cylinder wall is (factory) honed to expose and "round" the sharp silicon crystals. Iron plated (ferrocoat) pistons are required.

Nikasil is a thin plated layer over an aluminium cylinder. Electroplated nickel with a silicon cardbide dispersion. Aluminium (unplated) pistons are used.

There is no chrome (metalic chromium) in Nikasil. (Henry - mybe your thinking of Chromal - plated chrome subsequently mechanically "dimpled" ??)

John

[Henry Schmidt]

Originally Posted by jcge

Alusil is a hypereutectic alloy (AlSi17Cu4Mg or 390 alloy) in which silicon crystals are uniformly dispersed throughout the material. The cylinder wall is (factory) honed to expose and "round" the sharp silicon crystals. Iron plated (ferrocoat) pistons are required.

Nikasil is a thin plated layer over an aluminium cylinder. Electroplated nickel with a silicon cardbide dispersion. Aluminium (unplated) pistons are used.

There is no chrome (metalic chromium) in Nikasil. (Henry - mybe your thinking of Chromal - plated chrome subsequently mechanically "dimpled" ??)

John

Perhaps the more correct description would have been "chrome like plating".

[speedo]

While I do appreciate the technical aspects of why you cannot use Alusil cylinders with pistons expecting Nikasil...the reality is it happened and I inherited the problem. That is not by any means intended to be a smart-a$$ remark...I asked for and appreciate the feedback. I called Cosworth today and talked to a technical guy who wants to see pics of the damage. I will send them to him along with the part #s to see what the actual CR was intended to be. Which leads to the next question. My heads have been cut...by how much, I am not sure. I stated 3.5 mms, but that is this distance

Can someone measure a stock 3.0 head to get the distance from the bottom of the straight-edge to the top of the dome?

I sat with Grady on Saturday and compared it to a 935 head...and the difference was 3.5 mm. But a 935 head is different than a stock 3.0 head. My chain housings have been cut, but the cams were not centered in the seal...off by maybe 1mm. When the engine goes back together I will trial assemble the pistons and cylinders and heads with the cam towers and measure the deck height and valve clearance...the see how much more I will need to machine off the chain housings to center the cams in the seals. What is the best source for oversize idler arm sprockets cause I am probably going to need a couple?

Thanks guys

Speedo

[Jesset100]

Originally Posted by speedo

What is the best source for oversize idler arm sprockets cause I am probably going to need a couple?

Smart Racing Products - Home of SmartStrings, SmartCamber, and other quality performance auto parts
Page 26 in the catalog

[tom1394racing]

Lars

I just did a rough check on a stock 80 SC head

As shown in your pics:

To the top of the dome = 27mm

To the sealing surface = 2.5 mm

[lindy 911]

Lars,

Can you post the pictures of the galled pistons and maybe the cylinders? That would tell a lot about what is going on. If the pics above are the heads off your problem motor, I don't see any indication of detonation. What I have seen in the past is that there is a very clear area somewhere on the head void of carbon, almost as if a spot were cleaned off. The piston will also have a very clean spot that matches the spot in location with the head. So I think you're right to rule out detonation as the cause of the piston / cylinder gall.

Also, the information offered by Henry and others would suggest that if indeed your set up uses an aluminum piston without a coating in the Alusil barrel, creating similar metal contact in friction, that galling would be expected. It seems reasonable to assume that in your case the barrel would have deposits left behind from the pistons because the pistons are softer than the surface of the barrels. You could tell if this were true if the barrel galling were raised instead of a gouge or a scratch in the barrels surface. On the other hand, the pistons would have gouges or scratches in the surface of the skirts where the material was lost and deposited on the barrel wall.

One thing I have learned about this site is that although posters may get off subject from time to time, the information they offer is all related somehow. Knowing how the parts work beyond my questions builds my understanding of the big picture. You may already know all this crap but a lot of readers don't, and they benefit from your thread too. We're all on a giant learning curve, some ahead of others.

Thanks for the thread and hope you don't mind some of us using you as part of our "curve".

Lindy

[euro911sc]

Originally Posted by speedo

Can someone measure a stock 3.0 head to get the distance from the bottom of the straight-edge to the top of the dome?

I sat with Grady on Saturday and compared it to a 935 head...and the difference was 3.5 mm. But a 935 head is different than a stock 3.0 head. My chain housings have been cut, but the cams were not centered in the seal...off by maybe 1mm. When the engine goes back together I will trial assemble the pistons and cylinders and heads with the cam towers and measure the deck height and valve clearance...the see how much more I will need to machine off the chain housings to center the cams in the seals. What is the best source for oversize idler arm sprockets cause I am probably going to need a couple?

Thanks guys

Speedo

Originally Posted by tom1394racing

Lars

I just did a rough check on a stock 80 SC head

As shown in your pics:

To the top of the dome = 27mm

To the sealing surface = 2.5 mm

Tom: he needs your straight edge width or did you take that out of the measurement already?

Lars: Why machine the chain boxes instead of space out the cylinders with a thicker copper gasket? Seems spacing it out avoids a host of other problems...

Also, Are there not drawbacks using a straight 2.2 pump on this 3.0?

Henry: You have built a few of these 3.0 MFI wonders. How did you set up the pump and what would be the best pump to start with as a core ro does it matter at all?

Best regards,

Michael

[speedo]

Originally Posted by euro911sc

Tom: he needs your straight edge width or did you take that out of the measurement already?

Lars: Why machine the chain boxes instead of space out the cylinders with a thicker copper gasket? Seems spacing it out avoids a host of other problems...

Also, Are there not drawbacks using a straight 2.2 pump on this 3.0?

Henry: You have built a few of these 3.0 MFI wonders. How did you set up the pump and what would be the best pump to start with as a core ro does it matter at all?

Best regards,

Michael

Tom- Thanks for the measurement. Mine measure 32 from the mating surface to the dome and 28.5 from the flycut surface which makes sense because the flycut is 3.5 mm from the mating surface....hmmm, I thought mine would be less than yours, not more?

Michael- If you space the cylinders out using the base gasket, you change the deck height , piston to valve clearance and CR. Proper way is to machine the chain housings. The pump looks to be a stock 2.2 S pump, although it could have been modified. Jury is still out as to whether this is a good, bad, or acceptable setup. I have heard pros and cons. Most cons suggest the volume of fuel put out by the 2.2 S pump will be insufficient for the 3.0's thirst. Others suggest that the mixture can be tweaked and that will suffice. The ideal pump would be the later RS pump...but I think the smaller pumps can be made to work.

Lindy- Good to chat with you the other morning. Here are the pics....I honestly think the pistons and cylinders were equally affected. Hard to say if the cylinders hurt the pistons or the other way around.







Not pretty...but the good news is that the damage was limited to the P and Cs. If the rings had broken and thrashed around...I would have debris distributed all through the engine. If a piston had come apart, my twin plug heads would have suffered. If the engine had overheated and seized...whole "nuther" story. I don't feel fortunate, but it just goes to show you...if you have any concern about an engine and lack history, better to tear it down than run it. Engines with issues don't tend to heal themselves.

I am taking the heads to get checked out tomorrow...valves, seats, guides, seals and get the springs checked. Cheap insurance.

[jcge]

From 911 FWSM Volume 1 page E78 circa 1965 for a 2.0L


(C) Porsche AG

I believe this dimension was retained through all models to 1989 - others might verify or offer alternate info.

If you beleive the heads have been cut - consider the top surface of the head as your datum....(less frequently machined ex factory)

Can also confirm 27mm total dome depth that Tom measured above (taken from a 3.3T head)

John

[euro911sc]

Originally Posted by speedo

Michael- If you space the cylinders out using the base gasket, you change the deck height , piston to valve clearance and CR. Proper way is to machine the chain housings.

True, but you will be buying new Pistons and cyls anyway. You could spec the piston as needed to get the compression you would like. Just seems the easier route.

-Michael

[James Renfroe]

Hi,

It looks to me like the JE pistons were used with Alusil cylinders. Based on the aluminum scoring on both the pistons and cylinders, I would bet that is the case. The harder ferro-coat pistons needed to be used in these Alusil cylinders. Thanks!

James Renfroe

[HawgRyder]

Henry...are the Alusils something like the old 4 cylinder Vega engines?
I remember them having to be "etched" after boring to raise the silicone to the surface...or perhaps more correctly to lower the softer aluminum.
Bob

[Henry Schmidt]

Originally Posted by HawgRyder

Henry...are the Alusils something like the old 4 cylinder Vega engines?
I remember them having to be "etched" after boring to raise the silicone to the surface...or perhaps more correctly to lower the softer aluminum.
Bob

Yes, Alusil is basically a development of Reynolds 390 aluminum-silicon alloy. The same technology developed for Jim Halls Chapperall Can Am cars from the sixties.
The technology is a three-part system consisting of the hypereutectic 390 , compatible pistons (generally tin or iron coated) and a special cylinder bore finish.

The alloy contains round silica beads. When the bore is honed the beads present a flat surface surround by aluminum. The aluminum is etched away leaving the flattened beads as a contact surface for the rings to engage.

This technology is used by Mercedes Benz, BMW, Mahle, and Kolbenschmidt just to name a few.