SEALING HISTORY

.......APEX SEALS

People still ask me this question; "What about the apex seals? I have heard that Wankel engines have trouble with the apex seals." That problem went away in about 1965. However that is not to say that the Mazda apex seals have not continuously been improved to this day. The apex seals are central to the power generated and the efficiency of the Wankel rotary engine. The latest Mazda apex seals are very sophisticated and beyond the manufacturing capability of just about anybody but Mazda. Mazda has spent millions of dollars over the past 50 years improving the apex seal. The latest Mazda cast iron seals are heated with a laser and chilled in liquid nitrogen.

Many people and companies have tried in the past to manufacture apex seals. The vast majority have no knowledge of the history or the technical nuances of the apex seal problem. An ignorant person does not know what he does not know. The first thing a person needs to do when he is trying to come up with a better mouse trap is study the history of mouse traps and why they are made the way they are. Jan Norbye's book, "The Wankel Engine" published in 1971 has a good chapter on the apex seal problem. Another good book on the subject is "THE WANKEL RC ENGINE" by R.F. Ansdale published in 1969. I highly recommend anyone considering manufacturing apex seals get copies of these books and read them.

Some after market apex seals seemed to work fairly well but the life was suspect. Some caused chatter marks as the exact alloy or surface finish was incompatible with the rotor housing wear surfaces. Some recently introduced apex seals destroyed the rotor housing surface in short order as they had a rough surface finish and were too hard. Some provided less combustion chamber pressure which translated into less HP and higher fuel burn due to slight distortion of the warm rotor housing not matching the distorted shape of the hot apex seal.

To make a better apex seal than Mazda does is harder than it looks. Mazda uses cast iron for a very good reason. Cast iron is porous and retains oil. Is also softer than the chrome plating so it wears rather than the expensive rotor housing. Steel is the wrong material for apex seals. Many after market seal manufactures will not tell you what material they are using. They are probably using an alloy of steel as it makes the seal cheap to make. Therefore they can lower the cost. If the steel is hard it will destroy your rotor housings sooner rather than later.

There is only one person that sells better apex seals than those you can buy from Mazda. Francisco Iannetti manufactures expensive ($200 each) silicon nitride apex seals that have, for all practical purposes, unlimited life and excellent performance.

The following scans are from Jack Yamaguchi book RX-8






2. Acicular Irons.

Carbon 2.9-3.2, nickel 1.5-2.0. molybdenum 0.3-0.6%. Copper can replace nickel up to 1-5%. This rigid, high-strength, shock-resisting material is used for diesel crankshafts, gears and machine columns. With the correct amounts of nickel and molybdenum correlated with the cooling rate of a particular casting the pearlitic change point can be suppressed and an acicular intermediate constituent (ferrite needles in austenitic matrix) can be produced with high mechanical properties. Acicular cast iron is very much tougher than any of the pearlitic cast irons of lower strength. The tensile strength of acicular cast iron with a carbon content of about 3.0% will vary from 380 to 540 MPa but these figures can be maintained in quite large sections. Phosphorus should not exceed about 0.15% in the presence of molybdenum. otherwise a compound is formed which impoverishes the matrix of molybdenum. Quite large variations in silicon content can be tolerated, but chromium in excess of 0.% is harmful. The structure changes rapidly at 600-750°C and these irons should not be used at temperatures greater than 300°C.

540 megapascal = 78 320 pound/square inch

380 megapascal = 55 114 pound/square inch


**********************************************************

Some comments from Jeff Spitzer from the Aircraft Rotary Engine Newsletter. Jeff worked for Paul Moller and his responsibility was redesigning the one rotor air cooled OMC rotor engine to a two rotor liquid cooled engine. Jeff now is the project engineer on the Predator UAV.

-------- Original Message --------

Subject: Home made cast iron tip seals?

Date: Tue, 08 Sep 1998 16:37:29 -0700

Jeff Spitzer wrote:

"Simple cast iron" will probably not make a suitable apex seal. The oil film between the apex seal and trochoid is practically non-existent. That is why your quoted text refers to a hardening process for the wear surface.

The apex seals that were in the OMC engines were made from M2 tool steel. They are fairly expensive to make because they must be ground on three to five sides (one of them a radius as you mentioned) after heat treat.

Another issue that must be addressed is the combined friction/stiffness effects at the trochoid boundary. It was found by OMC (which used a sprayed on carbide trochoid surface) that the heavy steel tip seals were needed to dampen chatter just after the major axis. Lighter seals seemed

like a good idea until this problem was discovered. If you change materials (e.g. ceramic, carbon) and reduce the friction then you can reduce mass and still not experience chatter. My only point here being that it ain't as easy as "x is better 'cause y" really this is a challenging area to delve into.

Jeff Spitzer

*****************************************************************

Yes, I finally realized that after running across that SAE paper on the subject.

BTW In rotary drag circles this chatter phenomena is refered to as "jump-off the hump" :-)

Paul

*****************************************************************

Not actually. If you read again I was talking about just after the *major axis*. The phenomenon you are referring to occurs at the minor axis and always gets better with lighter tip seals (for high RPM running)

Jeff Spitzer

*****************************************************************

Mazda claimed they had completely solved that problem back in the early seventies either with special cast iron apex and Ni and SiC plating or hard chrome plating and aluminum impregnated carbon apex seals.

Cast iron and tungsten spray coating also works.

Of course what has replaced the carbon aluminum seals for racing now-a-days is ceramic which works on the stock housing.

I have never seen a rotor housing with chatter marks.

Paul

*****************************************************************

Jeff Spitzer wrote:

Paul Wrote:

I guess the major axis is where the apex seals change angle from drag to push.

Is that right?

*****************************************************************

Yes, IIRC this is where the apex seal begins to push. The chatter is caused by the drag of the tip seal pushing the seal back up into the rotor groove hence lifting it slightly from the surface. This, of course, reduces the drag and the seal recontacts the surface. If this phenomenon is resonant you have excessive chatter. The exact location where this will occur is usually somewhat after the major axis on the combustion side leading up almost to the exhaust port. (The intake side has more oil, less drag) The factors that determine this location are seal angle (push), tip speed, friction, seal mass, seal spring rate, etc. Note that the seal is slowing down throughout this region causing a change in the normal force/friction among other things. This is one of those cases where if everything is just right it will go wrong :)

Jeff Spitzer

*****************************************************************

So it is real important to match the spring rate of the apex seal with the weight of the apex seal itself. Two mm cast iron verses three mm cast iron for one example.

Higher spring stiffness and lower apex seal weight would increase the resonent frequency of course.

Another good argument for light weight ceramic apex seals.

Paul

*****************************************************************

Jeff Spitzer wrote:

Yes, in general I think lighter seals are better. Don't forget that there are really two springs here. One is the apex seal spring. The seal spring is really secondary in importance to the trochoid surface "spring." The frequency of the spring mass system when just the seal and leaf spring are considered is probably so low that the seal would be off the housing for a major part of the cycle for resonance. This just can't happen. That's why I made the point originally that compatibility of seals and trochoid surfaces must account for all factors (seal mass, friction, surface "spring rate" etc).

Jeff Spitzer

*****************************************************************

I don't want to give the impression that this is a major problem.

Nobody has complained in decades. The only problem Abel mentioned with his 700 HP Mazda two rotor is jump off the hump at 10,000 RPM and not chatter.

Paul

*****************************************************************

Jeff Spitzer wrote:

I recieved a brochure in the mail for a company that does metallurgical analysis. They are SMTS in Riverside. 909-689-6144. No endorsement.

Jeff Spitzer

*****************************************************************

Funny you should mention that. I just got Hurleys apex seal done at UCLA. 3% Si, 92% Fe, 1% S, 1% Cu trace elements make up the rest. Don't know the percentage of carbon yet. Electron microscope. That is Doctor Sergey Prikhodko with me in front of the EM. Anyone recognize that alloy?

Paul Lamar


*****************************************************************

....... O-RINGS

In the early 1970's Mazda had some water O-ring problems. A person would over heat the engine for whatever reason like running it low on coolant and the rubber O rings used to seal the rotor housings to the end housing would melt and leak coolant into the combustion chamber. The engine would not stop but it was hard to get it started after that as the spark plugs became wet with coolant. The newspapers and the automotive magazine press could not tell the difference between an apex seal and a cooling system seal so it was widely reported that Mazda had trouble with the "seals". Mazda fixed that by simply changing from ordinary 240 degree F rubber to 500 degree F silicone rubber.

Later on the Teflon encapsulated silicone O-ring was developed that totally and forever fixed that problem. These TES O-rings are good to 500 F and the 800 F Teflon coating is virtually inert to all known chemicals. If the engine gets that hot you are running it with out coolant at high power and risk melting the rotor housing.

Another problem showed up in the 1970's. It was possible to over heat the oil at high continuous power and that would melt the rubber O-rings in the sides of the rotor. Mazda changed to a higher temp rubber but not as high as Teflon encapsulated silicone. Consequently I recommend all TES O-rings when overhauling the rotary engine. Paul Lamar