Crankshaft rear oil seal

[rubyfloriley]

Hugo,
The guy doing my engine has done many oil seal conversions and swears by the SNG seal. He thinks conversion seals fail because they are incorrectly or badly fitted. ( i assume Bill Terry thinks the same as he has gone to the trouble of making a video showing the correct fitting procedure for his seals) So based on this i have gone with the SNG seal. With my XK i am assuming it to be the oil seal as the leak has appeared after winter lay up which i understand is when they appear, ie when the rope has dried out. I have run it for a few miles but it didn't go away or get less, must get it on a ramp all this e type work getting in the way.

[Hugo]

Do let us know how you get on. Out of interest, I just found this on eBay.au - item no. 191791421033. It's only 20 quid so I've ordered one just to see what it is, really.
I'm still torn between the original rope seal and probably Terry's mod. I need to make a decision so I can get the crank re-ground.
I have in the meantime received a private message on this topic which is very informative, but the writer doesn't want to say anything negative about certain suppliers in a public forum.
I will post some excerpts if I can do so without incriminating anyone!

[Hugo]

#### I received this (edited) comparison of the SNG and Terry's kits in a private message, as the poster, who clearly has experience of both kits, did not want to write anything critical about any suppliers on a public forum. Anybody care to agree/disagree (with the content I mean!)?

"The fundamental difference between the Terry kit and the others is the seal itself. Bill Terry elected to use a GM/Federal Mogul seal with a long service history. The designer of the SNG/XKS appears to have sourced a design specific seal.
The SNG seal is a conventional looking spring/lip seal with one "cut" on the radius. I believe the cut is actually a purpose designed gap as there would be a space in the circumference if it was simply sawed through. To install this seal the seal is spread apart at the gap and slipped over the crankshaft. The one in the kit I evaluated wanted to stay slightly spread apart even before any attempt to force it over the crankshaft. Once the seal is in place the assembly is lowered onto the bearing shells in the block. Then the lower bearing halves, then the bottom half of the seal carrier. Then the fun part...the spring for the lip must be fished around the lip of the installed seal and the ends hooked together. I believe the instructions call for super gluing the two faces of the "rubber" portion of the seal earlier in he process.
The first place for an error is getting that spring correctly placed on the seal The groove is tiny and buried out of sight for the most part. Also, as you know, the "rubber" portion of the lip is quite thin, and it's not under compression. I believe this is a source for potential leaks. Also the outer radius of the seal could be a leak source.
Terry's two part Federal Mogul seal is very thick and stiff around the outer radius (see video). It's listed as a "VITON" seal in the literature, as used in racing engines. The two faces are in slight compression when installed...and the surface area of the two faces is relatively large. The sealant for the outer radius of the seal goes on the raised radius of the carrier where it's easy to get it right. Overall an easy do-it-yourself installation with minimal chance for errors. Also a much more robust seal design."

One comment one super-glueing the seal together - you can buy split cv boots today that are super-glued together along the split, up & down all the concertina folds from one end to the other. In my view that is probably more work that fitting the proper boot, but my point is that these modern glues seem to do the job.
There is another factor that nobody seems to have taken into consideration - crankcase pressure. If you have a worn engine and/or blocked breather, the crankcase pressure would, I reckon, easily blow right past the rope seal, whereas a lip seal would tend to tighten up under pressure.
Not that this should affect me, as I'm rebuilding the whole engine.

[politeperson]

Very interesting comparison.

However I still dont' see the point.

Jaguar would not have fitted it in the first place if it did not work. After all we won the war didn't we? T he UK is the cradle of the motor car (maybe along with the others).

Coventry must have more than some idea what they were doing in the 1960's? I think so

As for a rope seal being ancient technology, yes. It is.

Many of the best things in the world are ancient technology. For example the pen and paper, the missionary position, alcohol, to name but a few.

I still think a correctly fitted rope seal will last a long time in a car that is used regularly.

[steve3.8]

As for a rope seal being ancient technology, yes. It is.

Many of the best things in the world are ancient technology. For example the pen and paper, the missionary position, alcohol, to name but a few.

All causes of messy leaks ?

[politeperson]

Good point Steve.

Yes they have that in common. Apart from a correctly fitted rope seal of course.

[christopher storey]

I still do not regard ANY seal modification which requires an 8.4% reduction in the cross-sectional area of the most critical part of the crankshaft ( it carries ALL of the torque and all of the load reversals ) as good engineering . It is axiomatic that the effect of the reduction is an undesigned increase of 9.2% in the rotational stresses in the tail of the crank . I imagine that Heynes/Hassan/Baily would have very long faces if they could have foreseen this totally unauthorised modification to their carefully designed engine

[Hugo]

You know, I hadn't even thought of what should have been a bleedin' obvious point. Thank you for doing my thinking for me - that is indeed a major consideration.

[64etype]

This is all a rehash of a J-L thread that I initiated back in 2012. The fact is I initially purchased the single split seal kit, was talked out of it by the rope seal crowd, and then took a close look at the Terry's seal. At that point I initiated a lengthy discussion about the strength/stress on the machined surface. Ray Livingston, a highly respected mechanical engineer with experience in this area spoke very convincingly that the area in question is lightly loaded and would be the last place one would expect a failure. Bill Terry echoed exactly what Ray Livingston was saying. Again, he runs these seals in very highly stressed racing engines....never a crank or seal probem in that area according to Bill. I remain convinced that they are correct. Contributors to that thread had used the Terry seal with no leaks...at least one for ten years. Search the Jag Lovers site in 2012 time frame. This the text of Ray's post:


John,

Peak stress will generally be at peak torque output - high
RPM, high load, as that is when the crankshaft is
transmitting maximum torque. A poorly balanced engine may
have a different peak, due to the additional stresses
induced by vibration.
What matters is the stress, which is force per unit area.
Double the cross-sectional area, and you halve the stress.
If you look at pretty much any crankshaft on any engine, the
most highly-stressed area will almost certainly be at one of
the rear-most bearing journals - either the last rod journal
or main journal - as the rear-most journals must transmit
the power of ALL forward cylinders, and the bearing journals
are pretty much always the smallest cross-section on the
whole crank. Due to torsional vibration, it may also be at
one of the intermediate journals on some
engines,particularly those with very long crankshafts like
our XKs. The flywheel flange and seal area are, in
comparison, MUCH larger cross-sectional area, and,
therefore, MUCH lower per-unit-area stress. So, you'll
probably never see a crankshaft fail anywhere OTHER than at
one of the journals, barring some really serious damage or
major material defect. Googling ''broken crankshaft'' then
clicking on ''images'' will, I think, prove this point.
Bearing journals will always be radiused at the ends, to
reduce the stress concentration. To do otherwise WILL
significantly weaken the crank, making it much more likely
to fail at that point.
The other point is that the crank DOES twist and bend under
load, and the strain induced by this twisting and bending
will be concentrated at the weak points - again near the
bearing journals. If there is ever to be a fatigue failure
on a crankshaft, it WILL be at one of those high-stress
points. The flywheel flange is, in comparison, a MASSIVE
MUCH stronger, stiffer structure, operating at a MUCH lower
level of stress, and will, therefore, not be where a failure
will occur. It would be exactly like wiring your house and
doing most of the wiring with 12 gauge wire, then using 4
gauge wire for the last foot before the outlet. If you
over-load that circuit, where is it going to fail? There is
a roughly 0% chance of failure within that 4 gauge segment.
Furthermore, if you look at the bolts attaching the flywheel
to the crankshaft, you'll see that they are FAR more highly
stressed than the crankshaft itself, making them also FAR
more likely to fail than the crank flange.
Fatigue also requires a tensile load, and there is almost no
tensile load on the rear flange of the crank (other than in
the area of the flywheel bolt threads). There is a large
torsional load, but a quite small stress, due to the large
cross-sectional area, and a compressive load, due to the
flywheel bolts. Also, for iron and steel, there is a
minimum stress required to introduce even the possibility
of fatigue. The stress in the flange area is certainly well
below this limit, again, other than around the bolt threads.--
Ray Livingston - '64 OTS Santa Cruz, CA
Santa Cruz, CA, United States
--Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]--

[Hugo]

This is fascinating, isn't it? One query with this analysis however; "Double the cross-sectional area, and you halve the stress."
Is that actually correct? I would have expected the ratio to be squared, i.e. double the load and you need four [2 squared] times the cross sectional area. That's just my instinct though - I'm not saying you're wrong.
And yes, I must admit, the main bearings look HUGE (as Mr Trump might say) in comparison with other engines, and as the writer says, they don't seem to break in service, do they, even when modified and on the track?
Hey ho .....

[Hugo]

Oh, there was another thing - I think you said the highest torque was at high rpm - I think that's incorrect - the highest individual torque loadings on the crank will be at low revs - i.e. peak torque. You will, it is true, get more of them at high revs, but each individual 'punch' will be less. Yes / no?

[Jeremy]

This is fascinating, isn't it? One query with this analysis however; "Double the cross-sectional area, and you halve the stress."
Is that actually correct? I would have expected the ratio to be squared, i.e. double the load and you need four [2 squared] times the cross sectional area. That's just my instinct though - I'm not saying you're wrong.
.....

Stress = force per unit of area. N/m2
Double the cross-sectional area for constant force = half the stress. No square function here.

[Hugo]

If you say so. But area = radius squared, that's why I thought that. It's like calculating the difference in pressure when you alter the size of a hydraulic cylinder (or as you say it isn't like that!); i.e. if you increase the diameter by 25% i.e. 5/4, the difference in pressure or travel becomes 25/16, or more than 50%.

[rubyfloriley]

Hi
Well as has been pointed out we won the war and within a very short space of time the wonderful XK engine was born, so why some 22 years after its arrival were the British motor industry Leyland to be precise building a wonderful car called the e type that was designed so it would rot from within its own structure aided by all the little crooks and crannies where water and salt could lie to aid the process. Maybe it wasn't designed to still be around 50 plus years later and waxoil hadn't been thought of.

Well i have fitted the SNG seal and will have to live with it, i dont intend racing with the e so hopefully it will stand up ok (assuming it has been fitted correctly) The perceived oil leak to my XK prompted me to do the conversion.

[Hugo]

You've done it already? 73 replies to my question and I'm no nearer a decision than I was when I asked the question! Actually I am. I think I'm going to stick with the original design, simply because if it does leak it's unlikely to be catastrophic, whereas if I do one of the mods I will have burned my bridges as it were. Once I've altered the crank there's no going back. My impression is that both Terry's and SNG's kit are pretty good, but if there is an issue with the fitment they can leak and leak badly, and it's a heck of a lot of work to have another go at it. And as it's me doing the job it's me who's going to have to put it right. I should say as it is I who am doing the job, it is I who am going to have to put it right. But that sounds weird.

[Hugo]

Hi
Well as has been pointed out we won the war and within a very short space of time the wonderful XK engine was born, so why some 22 years after its arrival were the British motor industry Leyland to be precise building a wonderful car called the e type that was designed so it would rot from within its own structure aided by all the little crooks and crannies where water and salt could lie to aid the process. Maybe it wasn't designed to still be around 50 plus years later and waxoil hadn't been thought of.

Surely the answer is salt? When did they start putting salt on the roads instead of sand? I think it might have been around that time? I might be completely wrong of course.
I just found a thread from last year about white vs black rope seals. The writer queried why jaguar changed from white to black for OEM, and which were better. Surely that's when asbestos went out of favour? I would guess that the white seals were asbestos whereas the back seals were not, and were graphite impregnated instead. Just another educated guess though.

[JagWaugh]

Hi
Well as has been pointed out we won the war and within a very short space of time the wonderful XK engine was born, so why some 22 years after its arrival were the British motor industry Leyland to be precise building a wonderful car called the e type that was designed so it would rot from within its own structure aided by all the little crooks and crannies where water and salt could lie to aid the process. Maybe it wasn't designed to still be around 50 plus years later and waxoil hadn't been thought of.

BL (or as we say in German: "British Elend") may have perfected the biodegradable car, but it was Alfa who did the pioneering design work.

[Jeremy]

Hugo, I took Rob Beere's advice and went for his one-piece seal. It's not cheap, as you need a new flywheel. But crucially it doesn't involve reducing the dia between journal and flange. I have no doubt the original designers knew exactly what they were doing. But as Rob has successfully rebuilt hundreds if not thousands of XK engines for road and racing over the past 30 odd years, I figure he's the best in the business today, and that's good enough for me.

[PeterCrespin]

Nice job. Looks like the later AJ6/AJ16 alloy sixes from 84-97, and the V12 from about 89 onwards.

Don't forget the pilot bush, or does the conversion include a needle roller in the flywheel, which is also what the above models use? :-)

[Jeremy]

I will check with Rob, but rest assured I won't forget it. 35 years ago when restoring my TR4A I had reinstalled the engine and gearbox and was feeling v proud of myself after a laborious weekend, when I spotted this little brass bush thing lying on the floor. What the hell is that, I asked myself.....