Fixing Crankshaft Seal Leaks by Jack Norris (EAA 225165)

After 35 years of near problem free Luscombe ownership and a year's work carefully majoring my C-85-12F, refurbishing the plane and engine compartment, and then repairing some serious wind created damage, I encountered a real nightmare of a problem, just when I thought I was again ready for serious flying. After 30 hours, the crankshaft seal began to leak, became worse and then apparently impossible to fix. After the sixth installation of the fourth seal, all of the tricks, combinations and smarts of every reachable expert in the U. S. had been exhausted. In compact form, the following chronicles the practical points as well as the obscure technical essence of the problem and solution.

11613 Seminole Cir.

Northridge, CA 91326

SEAL SPRING INSTALLATION TOOL — %z" MODEL AIRPLANE MUSIC WIRE

45° PATTERN (180 GRIT EMERY

ALL "NON LEAKING" SEALS OF

ALL TYPES WORK BECAUSE THE CONTACT STRESS AT THE SEALING SURFACE IS HIGHER THAN THE

FLUID PRESSURE BEING SEALED — AND THOUGH WETTED "BOUNDRY"

LUBRICATION

EXISTS,

A

TRUE

FLUID FILM AT THE SEAL INTERFACE DOES NOT EXIST. IF YOU GET A LAMINAR FLUID FILM, YOU ARE IN

TROUBLE — THE SEAL WILL LEAK! There are some weak, incorrect garter springs in the system, but a proper stiff garter spring on a C-85 lip seal stretches about 7/16" and develops about 2 Ibs. tension load. With two

pounds on each side, the 4 lb. hoop load develops approximately 28.5 psi contact stress on the 1.873" x .075 seal contact projected area. The 28.5 psi contact pressure is over 600 times as high as the 1-1/4 inches of water (.045 psi), actual crankcase pressure that I measured by putting a plastic manometer pressure probe tube all the way up the oil breather tube. The 28.5 psi seal stress is vastly higher than the case pressure, that's not the problem, but it is still a very low seal stress compared, for example, to a 3,000 psi hydraulic system "O" ring and it is all too easy to develop a fluid film under the seal, float it up and leak. That is the problem!

The spring can be shortened to raise the contact stress but not much. It is a mistake to shorten the spring by much more than about 1/4" because the spring simply yields, and can become distorted, thereby reducing the contact pressure locally. You can only gain about a 50% contact stress increase — helpful, but still not a big contact stress.

LOW CONTACT STRESS, ROUGHLY 30 PSI, CAN BE UNSEATED BY LAMINAR FILM FORMING FORCES. IF FILM OCCURS, SEAL LEAKS.

You can imagine that modern oils with wetting agents, high pressure additives, et al, are very, very oily oil and have a huge ability to develop a film, so you have a formidable adversary. If you have a problem seal, you must take specific, knowledgeable action to destroy the fluid film. Continental has a little known bulletin, M76-4, to put a 30 degree off centerline helical pattern of fine scratches on the crankshaft in the same direction as the prop pitch, which can both break up the film and pump the minute quantity of oil back into the engine.

I ultimately found a 45 degree pattern to be superior, and used 180 degree grit emery as Continental recommended. Since my shaft was nitrided in rework, it was nearly impossible to get a meaningful pattern into the shaft, especially with the recessed seal area and the interfering prop flange. Rather than the max finger pressure Continental recommended, I used a 3/16 x 3/8, 1 /8 thick neoprene rubber pad epoxied to the flat side of a long 3/16" blade screwdriver. Working from the top, I put a long 1/2" strip of 180 emery paper in from the 10 SPORT AVIATION 57

o'clock position and the screwdriver in from 2 o'clock and worked in the recommended 30 degrees off center line helical pattern. Shade or a darker hangar, out of the sunlight and a goose neck peanut light, lets you see what you are doing and also clean up carefully to keep all emery out of the engine. All of that and a tightened spring still didn't fix my problem, so here briefly is the final obscure technical key. It has just been learned in the past few years, that mechanical face seals can be film floated for very long life — but with small leakage — by a minute angle or fluid wedge radially across their face from the pressurized OD to the atmospheric pressure ID. The laminar leakage film flow pinching at the ID, a cube root phenomenon, causes a very non linear pressure drop across the face and roughly 80% of full system pressure to be exerted across the whole face area, which thereby floats the balanced face. Wedges as small as 100 or 200 millionths, .0001/.0002", across the face, will do this! THUS, VERY, VERY, VERY SMALL TAPERS OR GROOVES CAN BE MATHEMATICALLY DANGEROUS ON OIL SEALS!

Whereas my crank had been ground .010 under and nitrided in rework, the seal area had been polished smooth with emery, not ground, to clean up the old wear groove, as is common practice. It had been beautifully reworked and really had too smooth a surface and looked perfect, like a perfect diameter at installation. However, after

SEAL INSTALLATION

Seal tends to jump in crooked, shearing

at split joint. With two people armed

with 3 screwdrivers and one board chisel (at the split), slowly slide seal in place to avoid stick/slip phenomenon. Screwdrivers pivot against prop flange

for smooth installation.

58 FEBRUARY 1987

BENT SCREWDRIVER SEAL EXTRACTION

INSERT AT OD * ADJACENT TO PUT

TWIST, OUT

some running, I could again see the wear pattern of the old seal and a minute groove and undersize area (taper) mostly at the rear of the old seal's witness mark. My new seals tended to sit deeper in and overlap the old mark, as a result of shaft face rework, so I had shimmed some new seals forward various amounts in trying to get away from the overlap of the washout, which I knew was dangerous. Unfortunatley, this made a wider, new wear pattern and made it difficult to place the new seals accurately. It was almost impossible to do accurate measurements in the inaccessible area. Though impossible to measure the diameter installed, the groove or washout may only have been of the order or .001, more or less on a side and had not been noticeable at installation as polished. I knew that anything that looked like even a minute taper to the fluid could be deadly. Though the crankcase pressure was very, very low, too low to float the seal by itself, any hydrostatic pressure feeding into even a minute groove and taper under the rear of the seal lip, could combine with the impressive oiliness, wetability and viscosity of good oil and even the possibility of hydrodynamic film generation due to the shaft rotation to lift and float the seal. All the physics of films became the potential enemy trying to generate a film despite my helical pattern and I only had a weak 28.5 psi (x 1.5) trying to keep the seal seated. I went back and using max force on the screwdriver, worked a 45 degree pattern into the shaft. It was still not deep in the nitrided shaft, but it was a definite pattern that could be seen, that looked like it should survive some service life. I shortened the spring 1/4", formed

the new hook end very carefully so the garter load would be UNIFORM around the circumference. I then took my best shot at shimming the new fifth seal forward so it sat on its old mark, rather than straddle the minute groove and taper. The seal still leaked the first try, but this time clearly only at the top, adjacent to the split. Seeing a little mismatch this time at the seal split, which could cause local trouble, I evened the joint up and used a screwdriver bent 90 degrees to get back into the seal lip area to even up its seating. The seal immediately sealed and proceeded to run "normally". I breathed a huge sigh of relief. By that point I had zero interest in being smart, I just wanted to get it fixed. The bottom line conclusions: 1. A laminar fluid film is your enemy. Use a 45 degree pattern in the same direction as your prop's pitch to destroy the film and pump the excess fluid back into the engine. Boundry lubrication will still protect the seal face at the low contact pressure involved. 2. If your crankshaft is in for a major, get it ground in the seal area so it is a true diameter. No grooves, crowns or tapers, no matter how minute. Only a thousandth or even a few ten thousandths has the mathematical potential to put you in trouble if you get an "unlucky" combination of fits, location and phenomena. If your crankshaft must be ground a few thousandths undersize to clean up, don't worry about the unavailability of an undersize seal. I found my seals were actually bigger than the normal shaft in the free state but went roughly .070" under shaft size with the garter spring in place. That will more than handle any reasonable shaft clean up needed, .010 or anything of that order. (Continued on Page 62)

FIXING CRANKSHAFT SEAL LEAKS...

(Continued from Page 58)

After many seal installations, I got adept at the difficult installation and removal and I quit using silicon or permatex at the OD and the split since in a good, clean installation, sealant proved to be unnecessary and could get into the engine or interfere with a good fit. I was, however, careful to get seals with good split faces that would seal and was careful to remove any face grinding flash or burrs that could get in the joint and create a leak path. The illustrated installation and removal tools shown proved most effective. Needless to say, while this applies specifically to a Continental C-85, all the fundamentals apply equally well to any crankshaft seal problem. I realize full well that when we're talking about miniscule grooves and tapers, it sounds impractical and that one would

LETTERS TO THE EDITOR__ . . . (Continued from Page 4)

TRANSPONDER COMMENTS

Dear Mr. Poberezny: I am proud to be a member of EAA and am very impressed by its activities and programs in the general aviation field. My flight from Seattle to Oshkosh this year and my participation in the Convention has to be the highlight of all my flying experiences. However, I am afraid that I do not agree with the EAA stand on Mode C transponders. I do a fair amount of instrument flying, and travel throughout the West on business and pleasure. I have lost count of the times ATC has informed me of a contact, "altitude unknown." The busier the area, the more reports and the most of them, not surprisingly, occur in TCAs and TRSAs. If all aircraft in terminal areas such as TCAs, TRSAs and ARSAs had Mode C, it seems to me that the controllers would have to call only those contacts with a proximity and an altitude conflict. The workload of controllers would be less, the frequencies would have less traffic and the traffic calls would really get your attention. I am convinced that safety for everyone would be enhanced. I am aware of the argument that some aircraft simply are not able to carry transponders with Mode C and many owners cannot afford the cost. It is an unfortunate fact that a Mode C requirement would eliminate some GA aircraft from TCAs, ARSAs and TRSAs. Flying, and especially flying safely, has never been inexpensive! On the other hand, we as general aviation pilots have a responsibility for other's safety. In my opinion, flying in highly congested terminal areas without Mode C capability sig62 FEBRUARY 1987

think that a rubber seal should simply adapt to minute irregularities. Of course, it does in most cases. The practical answer is that if you look at the seal closely, the spring does not necessarily end up centered on the seal lip and the lip contact stress can be quite non-uniform. With powerful forces trying to form a film, some of the cases get in trouble and have serious problems. My unnitrided crank ran for 23 years before its recent nitriding and meticulous engine overhaul and ended up with a significant wear groove before overhaul, but always worked perfectly. We've all seen cases where pretty awful looking parts work perfectly well. The lesson to be learned here is that though poor parts can run perfectly, there are subtle, powerful physical forces that are able to give you nearly unsolvable problems with apparently beautiful reworked parts.

If you have a leaker — and have an otherwise good seal installation — the fundamental fact is that you are developing a laminar film under your seal. Your task is to figure out how it's developing and kill it. If you have any type of groove or taper under the seal, you can have big trouble. Understand the subtle film physics that you are up against here and the weak contact pressure and you can be the master of the problem even if you've become the unlucky one with the unusually tough case.

nificantly adds to the risk of midair collisions

their turns landing. We had not been waiting too long when the Captain announced over the public address system: "Ladies and gentlemen, I am sorry for the delay, but it looks as though every private pilot in Southern California is practicing their flying today. All you voters remember this when you go to the polls." He then apologized again for the delay. I couldn't believe what I had heard. I thought perhaps I would have cooled down by this time, but the more I have thought about his remarks the more angry I have become. We general aviation pilots

and all the tragedy that goes along with these disasters. The vast majority of our country's airspace would still be open to aircraft without Mode C or even with no transponder at all. It seems to me that the additional safety afforded by Mode C in terminal areas with radar coverage is worth the cost to us as GA pilots. I just do not think that such a requirement would be infringing on the rights of the general aviation pilot. I hope that you will publish this dissenting opinion, which is one that is shared by many of my flying friends. Sincerely, Philip K. Swigard 2017 Fairview Ave., E. Seattle, WA 98102 DISILLUSIONED PASSENGER

(Editor's Note: The following letter was shared with us by a fellow airline traveler and EAA member.) Mr. Frank Lorenzo Chairman of the Board Continental Airlines Houston, TX 77210 Dear Mr. Lorenzo: It is with great displeasure that I feel compelled to write you this letter, but as a general aviation pilot, secretary/treasurer of EAA Chapter 767 and a member of the 99s, I cannot in good conscience ignore what transpired on one of your Continental Airline flights. On December 3, 1986, I was returning to Wyoming from Southern California on Continental's Flight 588 departing John Wayne Airport at 2:50 p.m. which was filled to capacity. As we taxied out to the ramp I couldn't see how many planes were ahead of us, but behind us there were several. There were numerous planes, large and small, taking

About the Author Jack Norris, a Management Consultant in Los Angeles, has founded and managed companies, is a graduate engineer with 35 years experience in the aircraft and high technology industries. He has seven of his products in the Central Hall of the National Air and Space Museum.

have suffered the most because the public

hears and reads about this kind of rhetoric and believes it to be the gospel truth. General aviation does not deserve or need this. We are having a tough enough time trying

to stay in the air because of cost, FAA's

boondoggling with airspace, and then to

have a fellow pilot give a spiel such as he

did. It also makes me wonder where the Captain gained all his experience in aviation. I don't recall hearing about a birth in the cockpit of a 737! I have flown Continental exclusively since 1971 when your airline flew my daughter, who was seriously ill, from Denver to Los Angeles. The crew was most helpful and I never forgot how gracious they were when I needed moral support. However, if this is the new Continental Airlines, you can bet your last flight that I will not be flying nor recommending Continental in the future. It is ironic that the seriously ill child your airline flew from Denver in 1971 is now a private pilot working on her ratings to become a corporate pilot. I hope that she remembers you don't burn your bridges behind you as you may have to return over them someday. Your Captain should remember that if it were not for general aviation, he would not be where he is today. Sincerely, Barbara J. Croy, EAA 181197 7030 W. Redrock Dr. Gillette, WY 82716