9.8 or 10.1?

[ChrisBennet]

Originally posted by jadeblaquiere
Not to doubt Steve's words... but why does too much deck hight promote detonation?

If I understand things correctly, having a compression ratio too high heats the intake charge to the point that the fuel spontaneously combusts.

I'm just curious as to what the dynamics are for too much clearance (and lower temps) causing the mixture to ignite...

--
Joe
1986 951
1977 911

Not intuitive is it?
The larger deck height creates a pocket out at the edge of the piston can promote detonation - or at least that's how Steve explained to me once.
-Chris

[Steve@Rennsport]

Joe:

Chris has an excellent memory,...

You are right about too much compression triggering spontaneous combustion, but there is FAR more to this story. The detonation thresholds of an engine depend on many things such as:

1) Compression ratio
2) Piston crown shape & design
3) Ignition timing
4) Spark plug position
5) Combustion chamber shape: hemi, pent-roof, or wedge
6) Squish band placement & shape
7) Natural turbulence or swirl (or not)
8) Fuel mixture and distribution in the chamber near TDC
9) Cylinder head temps
10) Outside air temps
11) Octane

These are just a few and a thorough discussion of each would fill a book (one day I might do just that).

With regard to the low-swirl found in hemispherical head engines and especially ones with offset plug locations, its critically important to force the air-fuel mixture toward the plugs approaching TDC as cylinder temps are rising. The goal of proper deck height is the prevention of any air/fuel pockets remaining out by the edges of the bores that can ignite BEFORE the plugs do,...

Maintaining close (spec) deck heights ensures that the majority of the air/fuel mixture gets squeezed toward and around the plug(s). Too much deck height will leave those pockets of air/fuel and too little deck height permits piston-to-head contact at high RPM.

I shoot for .040 (1mm) in street engines that will run on pump gasolines. For race engines where I'm looking for that last bit of compression ratio, I will reduce that clearance depending on the piston I am using.

Like I said, this subject is fascinating (at least to me) and could fill a book but I hope this helps clarify the issue for you,...

Analyzing burn patterns on piston crowns and heads for 28+ years gives one an education if one is willing to learn.

[kenikh]

Originally posted by Steve@Rennsport

Like I said, this subject is fascinating (at least to me) and could fill a book but I hope this helps clarify the issue for you,...

Please write a book. I will buy it!

[jadeblaquiere]

I agree... would make a good book. Thanks for the explanation. Sign me up for the 1st edition!

Until then at least we have the Pelican Parts board... I learn something every day here.

--
Joe

[Grady Clay]

I am very impressed with your tenacity to get repeatable
CR measurements. This is not as easy as many portray
and there are many places where errors can creep in.

As a final check, I prop the engine so the spark plug hole
is absolutely vertical and measure the CR with the piston
at TDC, the rings and valves sealed with grease and the
head torqued in place. In the CR range of 10:1 to 12:1
you can get accuracy of better than 0.05:1 directly
measured combined volume at TDC.

I have a couple of tools “Of Local Manufacture” to do this.
First is a sparkplug shell modified with a 30 cm piece of
½” drill rod. This allows me to use a level to make sure
the sparkplug is correctly positioned. The other is an
empty sparkplug shell with an O-ring and a calibrated
tube for filling with a burette. Have a calibrated step a
centimeter down and the ID small so you can distinguish
evaporating test fluid from a leak at the rings, valves or
head gasket.

For a measuring fluid I use Stoddard Solvent with some
red ink. I use a high quality glass burette. Use one with
markings etched all the way around. This will help
eliminate parallax and allow good interpolation to 0.10 cc



Steve and I have been in perfect agreement on this for 28
years (I thought it was way over 30).

In this discussion, Steve’s #6) is of particular importance.
The high dome angle at the transition from the flat
perimeter of the piston is particularly critical. The dome
actually “traps” mixture in the pocket formed by the flat
portion below TDC Compression while the piston is still
several mm down. This happens just before TDC
Compression and just after.

Most of the time we concern ourselves with simply the
piston-to-head clearance and try and have it everywhere
equal (and as close as reasonable for the application. In
the above situation the dome “closes off” the area for
gasses to move from (to) this “pocket” faster than the
volume of the “pocket” is reduced (expanded).

Below is an exaggerated diagram where the important
feature is the closing off of the gas path (yellow) while
the “squish” volume is still decreasing (red arrow).



The volume of the “squish area" above the ring will
continue to be reduces with nowhere for the gasses
to go. This is like a second combustion chamber with
a 30:1 CR. The mixture will detonate.
Things go downhill from there.

The “ideal” combustion chamber is almost spherical with
4-5 valves in the head, a central sparkplug, an equally
dished piston and a significant “squish” area around the
perimeter. This forces the gasses back into the combustion
chamber (green arrow).



The problem is that this isn’t conducive to high CR with our
bores, strokes, rod lengths, valve actuation and size and
spark plug(s) placement. Most important is cooling – or
lack of it. This is one of the reasons for the change to
liquid coolants.


The detonation result is classic for Porsche. Detonation
occurs above the compression ring. The aluminum piston
deforms downward and prevents the compression ring
from floating in the grove. Remember it is the gas
pressure behind the ring that provides the sealing force.
Once the ring is no longer able to seal, hot gasses pass
the side of the piston and bad things happen. Typically
the ring brakes up and pieces destroy the piston. In other
cases the piston overheats and expands to seize in the
cylinder. Or both.


What to do?

Almost everything in automotive involves compromise.
With this piston-head situation, you can’t use as much
CR as you might otherwise get. There must be an angle
transition in the head from the deck squish to the
combustion chamber. The piston dome can’t rise as fast
as the angle in the head. The net result is slightly less
CR but not forming a detonation pocket.



With your intended use, you won’t have any issue with
getting high enough CR. In fact you should concentrate
on the best possible combustion chamber, squish and
turbulence. I will still argue for twin plugs.

The “deck” piston-to-head clearance should still be as
little as reasonable given street use and the type piston.
Something slightly more than 1.00 mm seems to be
common. Should it be less than 1.00 mm?
How close do you go?

Best,
Grady

[pete917]

i have a 3.5ltr mahle piston and barrel conversion that I am doing on a eurospec 3.2c. I assume a 0.040" deck height is still required. The heads have been resurfaced and i have had some machining done on the cases to get the bigger bore cylinders located. Its therefore highly likely that each cyclinder will need slightly different shims to get the same deck height across all six.

Is this correct or OK to shim differentyl to get the same heights?

Cheers

[ChrisBennet]

Originally posted by pete917
i have a 3.5ltr mahle piston and barrel conversion that I am doing on a eurospec 3.2c. I assume a 0.040" deck height is still required. The heads have been resurfaced and i have had some machining done on the cases to get the bigger bore cylinders located. Its therefore highly likely that each cylinder will need slightly different shims to get the same deck height across all six.

Is this correct or OK to shim differently to get the same heights?

Cheers

The guy doing the heads should make them identical heights and the guy doing the case should "deck" (make flat and all the same height) the case spigots. The latter operation will change the deck height (see above) and may require a thicker than stock shim height to compensate. In my limited experience, decking the case spigots just made the deck height closer to "ideal" (.0035-.0040") EDIT: Oops, I meant .035-.040", Thanks Grady! with the stock shims.
-Chris

[Grady Clay]

Chris,

Aah, the good ‘ol misplaced decimal point. Chris, I assume you mean deck (piston-to-head clearance) of 0.035” to 0.040” (~0.9 – 1.0 mm).


Pete, WELCOME to the Forum. You will find a lot of help here. Where are you located? How far along is your project? Details, details.

These clearances are very demanding. What is your experience? What do you have for local resources?

A critical issue in building these is the cam housing must sit on three heads that are at the same height when torqued in place. This prevents warping the cam housing and no longer having a straight bore for the cam. Great care must be taken here.

At the same time you need to have your design deck clearance the same among cylinders and the CR the same among cylinders. This then all needs to be the same as the other bank of cylinders. There is the fit of the chain housing to the cam housing and possibly chain length. Very simple.

I recommend you do some serious research using the Search function. A lot has been discussed here and elsewhere but there isn’t a book or single manual. Bruce Anderson’s Porsche 911 Performance Handbook (available from our host) is about as close as you can get. Make yourself a detailed plan. Post it here for review. This not a “try it and see if it works” subject.

The first one of these (74.4x100) we built was about ’78.

Best,
Grady

[pete917]

Thanks Guys,

Im located in London, UK.

The project has progressed as follows:

base engine: 32.c euro spec;
3.5ltr mahle pistons and barrels;
suffle pinned engine case;
resurfaced engine case;
bore aligned engine case;
polished crankshaft;
conrods capped and honed and balanced;
twin spark heads;
resurfaced heads;
new valve guides;
new valves;
reseated valve seats;
EBS titanium springs and retainers;
ARP rods and bolts;
Resurfaced cam followers;
GTII turbo camshaft profile by Camgrinder;

Apart from being financially ruined :-( so far bottom end has been completely rebuilt, and and initial top end built to test deck heights........

and thats where are I am. Im not personally building, its being done by an engine builder local to me here in the UK.

I am going to have the deck height at .040". A question with regard to tolerance on the deck height in terms of allowable variance between all six. How much tolerance is there before it causes problems with camshaft bore alignment.

Cheers
Pete

[Grady Clay]

Pete,

Looks like a great project.

First the usual questions among us “…separated by a common language.”

“conrods capped and honed and balanced;”
What is “capped”?

“ARP rods and bolts;”
Do you mean “ARP rod nuts and bolts”?


OK.

suffle pinned engine case; resurfaced engine case; bore aligned engine case;

How much material was removed when the case was resurfaced? What is the sum of the amount off the mating surfaces and off the cylinder seats? This will give you some idea of the reduction in deck height.

How much was taken off the heads? Enough to have interference between the head and top cylinder fin? Were the heads cut there also for fin clearance?

All these numbers contribute to the narrowing of the engine and the range of easy deck adjustment.

What is the specified Compression Ratio (CR) of the pistons with this stroke? Have you made some actual measurements? How (what techniques)? What results?

What issues are there with this cam and valve-to-piston clearance? What valve cut-outs do these pistons have?

I am going to have the deck height at .040". A question with regard to tolerance on the deck height in terms of allowable variance between all six. How much tolerance is there before it causes problems with camshaft bore alignment.

I can’t give you an exact number there because there are too many variables (head mostly). You should measure the deck individually on all three cylinders. Assemble that bank using the cam housing to locate everything (have the cam installed with assembly lube) and to see if there is any cam bind during the head torquing process. If so - STOP. If not, remove the cam & housing and inspect the head-to-cam housing surfaces of the heads with a knife edge. You can spot differences <0.0001” this way.

Additional clearance deck differences in the ‘tenths (0.0001”) range are not of concern. If you are pushing the CR limit, having the CR all the same is more important.


More questions:
What is your planed use?
What car, weight?
What intake?
What exhaust?
What fuel do you plan to use?
Was this a good running engine before? Any prior trama?

How repeatable are your measurements of everything?

With you, your builder and everyone on the Forum, you should be able to have this exactly how you want it. We will be able to see the smile half a world away.

Best,
Grady

[pete917]

Hi There,

Its been a great [project and has consumed my life for what seems ages:

"Capped" is the resurfacing of the conrod mating surfaces, and then the centre honed out to ensure concentrivity and removal of any ovalness;

I have used ARP for the conrod bolts and nuts, and then also the main engine studs (great product);

our absolutely right, the removal of material from the case halves and heads, although small, has consumed most of the 0.040" deck height, but I will be shimming back to size.

I havent yet measure the CR (next on my list) and will use the method in Bruce or Waynes books. Mahle say that they are designed to retain my euro spec 10.3:1.

The cam profile I decided on i got from Doherty race cams http://www.drcamshafts.com which he has taken my entire spec into consideration. The Mahle pistons come with some deep pockets. Ill be spending some quality time checking P-V clearances in the near future.

I forgot to mention I have also added a Motec M48 Proc CIS to help map everything together.

I had no drama with the ending before other than normal wear and tear.

The intake system is going to be a set of Jenvey Throttle bodies and the exhaust is a custom made equal length stainless steel system from Hayward and Scott in the UK. My exhuast is featured on their website

http://www.haywardandscott.co.uk/case2.asp

everything else i have bought through EBS Racing www.ebsracing.com

The car will be road and track use. Its not been lightened in any way, but I may well do somthing when money permits. This project has a small fortune so far, so things may have to slow done some.

Ill keep you all posted and look forward to all your advice

Pete

[CBRacerX]

Since this has turned into a most informative thread, let me add "my issues" here. I have a set of RSR 3.8 Pistons (which appear to be the 993 version) and have measured my deck height to be 1.3mm using stock 964 heads machined for the larger jugs. Apparently, this deck height is about what they start at. So... Should I be machining the cylinders to get this down to 1.0 mm?

Pics here: Another 3.6->3.8 Rebuild Thread...

[Carrera3.5L]

Pete,

Are your heads still single plug or have you twin-plugged them? I believe that I have the same p/c set that you are using in my engine and I had some compression ratio & suitable deck height issues that needed addressing...

I orginally cc'd my twin-plug heads at 91.8 ml and the cylinders at 71ml. After test fitting a piston/cylinder on the case my deck height was only .5mm with the standard .25mm base gasket. The compression ratio calculated was slightly under 9.2:1. I wasn't happy with this result, so I disassembled the heads on the bench and had them milled .020" to gain back a few tenths of compression, reassembled them and than cc'd again.

Have you cc'd your heads and cylinders yet? I'll give you my numbers just for a comparison. Head volume (twin-plugged remember) after milling .020" was now 88.8 ml. Cylinder volume was 71 ml. With a thicker .5mm base gasket now in place of the .25mm unit I ended up with a .75mm deck height. If I would have stacked a .5mm and .25mm base gasket I would have ended up at 1mm deck height but I personally don't like stacking gaskets if I don't absolutely have to. Measured compression ratio with .75mm deck height was now 9.49:1.

Worked out okay so far (keeping my fingers crossed every time I turn the key)...

Ralph

[pete917]

Thanks Ralph. I have cc'd mine yet, but will do so and report back.

I have gone the twin spark route though. I noted on my 3.2 barrels that on the opposite to the spark plug, that the barrel coating had started to wear. With a plug either side it should even things oout and get evenly spread presure.

What other mods have you made to your engine?

Pete

[Carrera3.5L]

Originally posted by pete917
Thanks Ralph. I have cc'd mine yet, but will do so and report back.

I have gone the twin spark route though. I noted on my 3.2 barrels that on the opposite to the spark plug, that the barrel coating had started to wear. With a plug either side it should even things oout and get evenly spread presure.

What other mods have you made to your engine?

Pete

Very minor mods and I am still retaining the Motronic system. Polished the intake ports, Extrude Honed the intake plenums, enlarged throttle body, mild Web Cam 20/21 cam profile, 1 5/8" ceramic coated headers from George Narbel coupled with a Monty street muffler and custom dyno optimized chip from Steve Wong running on poor quality 91 octane "gasoline". We have stringent emissions tests in Southern California which resulted in a "compromise" motor and other than changing the exhaust every 2 years, the car easily passes the rolling dyno tests...

I'd be interested to hear what you end up with for static compression and how you got there...

Ralph

[pete917]

Sounds like a great car. Have you got some pictures you could post?

Im a bit luckier in the UK, I use Shell Optimax fuel which has more than 98RON.

It should only be a few more weeks and Ill be able to report back on all of the interesting stats.

[Carrera3.5L]

Originally posted by pete917
Sounds like a great car. Have you got some pictures you could post?

I've never been shy to post pics, although by searching you'll find the same ones over and over and over...

Apologies to Mr. Clarke, this thread kinda spun out of control and went from your question/problem to Pete's...

This 100mm Mahle piston/cylinder should be what you have as well...




Getting closer...




Ready to install...




In the car with approx. 10K on it...




Maybe we should start a new thread if Mr. Clarke deems appropriate?

Ralph

[pete917]

thanks Ralf. Very impressive.

Yes the MAhle's look familiar. I was wondering if the deck height with this replacement are problematic with regard to deck height. Mine is all gone, yours was initially 0.5mm... hhhmmmmm.

I'll try and dig out some photo's of mine as it comes together.

[pete917]

Ralph,

With regard to compression I wonder if your original result was because of the US versus European specs. I understand that the CR for a 3.2c is 9.5:1, wheres the CR for a euro spec 3.2c is 10.3:1. I wonder where the difference is in the two cars when original. If its in the heads then that might go some way to explain the differences.

Im about to measure my CR, so will report back with the numbers soon. Meantime, do you have the calculated volume of the dome of your pistons?

Cheers
Pete

[Carrera3.5L]

Originally posted by pete917
Im about to measure my CR, so will report back with the numbers soon. Meantime, do you have the calculated volume of the dome of your pistons?

Cheers
Pete

Hi Pete,

Here are all my pertinent calculations:

Deck Height - .75mm
Head CC - 88.8
Dome CC - 25.863
Swept CC - 584.04
Deck CC - 5.8875

Swept CC + Head CC + Deck CC - Dome CC divided by
Head CC + Deck CC - Dome CC = 9.49:1 CR

Andial (one of the Mahle distributors) claims this is a 9.5:1 set, whether using the DME or RSR style piston. The other distributor (SSF Auto Parts) advertises them at 10.3:1 but they are simply a parts reseller and pass on whatever info Mahle Motorsport spoon feeds them.

Good Luck! I'm interested on how it turns out for you...

Ralph