Ships Nostalgia banner

21 - 33 of 33 Posts

·
Registered
Joined
·
45 Posts
Seem to recall that 9 m/sec is the preferable max piston speed, rather than 90 m/sec. The latter speed would really keep us engineers busy looking over the side for blown piston crowns & cylinder heads.
True, where did I get the 0 from. Don't trust an old memory.
 

·
Spongebob
Joined
·
9,328 Posts
Trying to think through all this fog , both steam and time gone by , is it correct to say that a two-stroke piston bottom end bearing is always loaded on the top half whereas a four-stroke cycle the load goes on to the bottom half during the induction stroke.
A excess bearing clearance in a 4/ might knock but in a 2/ the bottom half is only a keeper.
Tell me to get back to my gardening if you wish

Bob
 

·
Registered
Joined
·
6,271 Posts
Trying to think through all this fog , both steam and time gone by , is it correct to say that a two-stroke piston bottom end bearing is always loaded on the top half whereas a four-stroke cycle the load goes on to the bottom half during the induction stroke.
A excess bearing clearance in a 4/ might knock but in a 2/ the bottom half is only a keeper.
Tell me to get back to my gardening if you wish

Bob
Far too simplistic, Bob! Think about it and draw a simple vector diagram by crank angle, number of cylinders and what the other cylinders are doing! I have, somewhere, the B&W report on the 11 cylinder 98 Cm. engine. Also, in that case, the effect of the alternator and the overhang effect. Rather a splendid paper, but "for internal use only"!

Another interesting claim was a cracked crankshaft in a Wartsila V18 cyl. medium speed 4/S. Someone asked for an analysis of propagation of the crack versus rotations of the engine and why there were no auxy alarms to alert operators to the pending disaster. These engines have also an auxiliary system (cannot remember how it works, precisely) to confirm splash lubrication.

I did the analysis and the maths. Less than 2.2 seconds from crack initiation to failure! On "yer mudders grave" it was not predictable, merely fortuitous! It took out a couple of cylinders and required a new crankshaft, but I was quite taken aback how the engine resited more damage!

Rgds.
Dave
 

·
Registered
Joined
·
250 Posts
Bob, with boost pressures of up to 3 Bar in today's modern 4-strokes, the induction stroke hardly puts any load on the bottom end bearing bottom half when loaded at the top end of the rpm range. For a constant-speed 4-stroke (i.e., a lightly loaded generator), there is not much boost pressure acting on the piston crown, so there is the potential for more wear on the bottom half.
On a variable-speed 4-stroke propulsion engine, the boost pressure rises as the rpm rise, so load on the bearing top half is much greater on the induction stroke than, for example, a naturally aspirated diesel engine. The boost pressure of up to 40 psi exerts a good downward force on the piston crown when passing over TDC, thus reducing the reciprocating load on the bottom half of the bottom end bearing as the piston reverses direction.
 

·
Premium Member
Joined
·
10,202 Posts
#23 ,

Ah, Dave, a true disciple of the medium speed crap-heap. Who else would describe wrecking two units and a crankshaft as "fortuitous"? Only in the manner Capt. Smith might have regarded his last iceberg.

I do agree that there cannot be alarms for everything although I have often thought that crank case hot spots could be detected at a much earlier stage with a low vapour pressure (and low hazard) luboil additive and a more sophisticated detector targeting its vapour.
 

·
Premium Member
Joined
·
142 Posts
Bob, with boost pressures of up to 3 Bar in today's modern 4-strokes, the induction stroke hardly puts any load on the bottom end bearing bottom half when loaded at the top end of the rpm range. For a constant-speed 4-stroke (i.e., a lightly loaded generator), there is not much boost pressure acting on the piston crown, so there is the potential for more wear on the bottom half.
On a variable-speed 4-stroke propulsion engine, the boost pressure rises as the rpm rise, so load on the bearing top half is much greater on the induction stroke than, for example, a naturally aspirated diesel engine. The boost pressure of up to 40 psi exerts a good downward force on the piston crown when passing over TDC, thus reducing the reciprocating load on the bottom half of the bottom end bearing as the piston reverses direction.
About 30 years ago I had the misfortune to be involved with some Mirrlees MB275s ([email protected]) driving a CCP and a front end 1500KVA alternator and after about 15k hours (2 years) we started to get BE bearing failures. Mirrless said it looked a bit like light load overspeed but we were convinced that that could not be the case as, although the engines had huge load swings whilst they were driving the discharge system, we never saw anything above 1020rpm on our analogue instruments. Eventually we fitted a digital rpm recorder and to our horror we saw that in fact we had in excess of 1090 rpm for a few seconds every couple of minutes when the load came off. Working it out it seems that we were subjecting the engines to about 25 hours of very light load overspeed every year. Oops!
 

·
Registered
Joined
·
6,271 Posts
#23 ,

Ah, Dave, a true disciple of the medium speed crap-heap. Who else would describe wrecking two units and a crankshaft as "fortuitous"? Only in the manner Capt. Smith might have regarded his last iceberg.

I do agree that there cannot be alarms for everything although I have often thought that crank case hot spots could be detected at a much earlier stage with a low vapour pressure (and low hazard) luboil additive and a more sophisticated detector targeting its vapour.
The original fault was an inclusion in the crankshaft forging, below 2 microns, which is the limit of resolution detection. This caused the initiation of a crack which propagated over time with the crankshaft cyclic loading, until it ran completely out of the journal and through the web.

I have seen other faults on Vee engines where the piston rods share a journal, usually due to seizing but without damage to the crankshaft which, normally, only needs in situ machining, honing and the fitting of oversize bearings (They even get a very nice brass plate on the entablature of the particular crancase, below the access cover).
Rgds.
Dave
 

·
Registered
Joined
·
3,517 Posts
Far too simplistic, Bob! Think about it and draw a simple vector diagram by crank angle, number of cylinders and what the other cylinders are doing! I have, somewhere, the B&W report on the 11 cylinder 98 Cm. engine. Also, in that case, the effect of the alternator and the overhang effect. Rather a splendid paper, but "for internal use only"!

Another interesting claim was a cracked crankshaft in a Wartsila V18 cyl. medium speed 4/S. Someone asked for an analysis of propagation of the crack versus rotations of the engine and why there were no auxy alarms to alert operators to the pending disaster. These engines have also an auxiliary system (cannot remember how it works, precisely) to confirm splash lubrication.

I did the analysis and the maths. Less than 2.2 seconds from crack initiation to failure! On "yer mudders grave" it was not predictable, merely fortuitous! It took out a couple of cylinders and required a new crankshaft, but I was quite taken aback how the engine resited more damage!

Rgds.
Dave
Crack failure analysis is interesting, for my dissertation I did a project on hull life estimation for a bulk carrier. The really interesting thing from my point of view was that critical crack length is not dependent on structure size, and once it has been achieved propogation takes place at the speed of sound in the material under stress. Messers Miner and Basquin have theories and rules!! (But you knew that anyway.)

The finest engine I ever sailed with was the B&W 9K98FF, unusually it had second moment compensators at front and back ends and No 9 Exhaust Valve had shims under the springs to reduce vibration in the camshaft. That well built and engineered that all the liners were still original (although not in the same holes) and some of the piston rings were still original (or at least 20 years old, some of the early records had been lost) none of the crowns had been welded when she went on her final voyage at 27 years old.
 

·
Registered
Joined
·
6,271 Posts
Crack failure analysis is interesting, for my dissertation I did a project on hull life estimation for a bulk carrier. The really interesting thing from my point of view was that critical crack length is not dependent on structure size, and once it has been achieved propogation takes place at the speed of sound in the material under stress. Messers Miner and Basquin have theories and rules!! (But you knew that anyway.)

The finest engine I ever sailed with was the B&W 9K98FF, unusually it had second moment compensators at front and back ends and No 9 Exhaust Valve had shims under the springs to reduce vibration in the camshaft. That well built and engineered that all the liners were still original (although not in the same holes) and some of the piston rings were still original (or at least 20 years old, some of the early records had been lost) none of the crowns had been welded when she went on her final voyage at 27 years old.
Fascinating stuff, Duncan. I learned a lot with a metallurgist and in a specialized lab in Dallas.

A US company had bought an independent production plant in Quintana Roo, just outside the arqueological remains at Chichen Itza.

About two months after the sale, the "single shaft" generator shaft cracked in the location of the field bar bushings. Now, a single shaft generator is comprised of a gas turbine coupled through a reduction gearbox to an alternator and, on the other end of the alternator, is coupled a direct drive steam turbine (Alsthom design, cogeneration).

The electron microscope detected a red residue on the crack surface. It turned out that the original bushing had been fitted with crossed threads that had initiated a crack in the thread profiles. The final layer of dielectric paint had penetrated the crack, giving rise to the evidence of the smoking gun barrel!

The other telltale of failures are the tide marks, like an arrow pointing to the culprit.

I got my first lesson on crack propagation from my Dad, when we visited a Blue Funnel Liberty (Memnon or other "M") when I was about four years old. Visiting with the Sparks to his shack, my Dad pointed out the only riveted joint in the hull, abaft the wheelhouse, to ease the tensions and avoid "notch cracking". It has stayed with me all my life!

I must admit that my Dad, although dedicated to lecturing in Marine Engineering, probably harboured a desire to be a metallurgist. When he did his time with BF in all departments, there was a dedicated metallurgy lab in Odyssey Works.

My Dad was involved in the solution to cracking in cast cylinder liners due to the struvture of the spherical cast iron (crystal boundary tension). The usual fix were tensioning bands but the metallurgist devised a method to reform the crystalline structure. I cannot remember the details, but my Dad always refers to him as a very clever man!

I have the metallurgy book published by the Metallurgist, dated 46 or 47 and signed by the metallurgist and gifted to me by my Dad - And it is still a mine of information.

Rgds.
Dave
 

·
Registered
Joined
·
3,517 Posts
Liberty Ship cracking, another fascinating subject, solved by Constance Tipper, who lived about 5 miles as the crow flies from where I'm sitting now, always wondered if she was buried in the local churchyard, one of my failings is visiting the graves of the unrecognised!!

One thing I was told by one of the Chief Engineers on the B&W engined ship was that if you need a new stainless steel pump shaft machining ashore specify that the nut and shaft should come from different heat numbers of steel otherwise they will gall and seize - naturally pump manufacturers know this and will manufacture OEM spares accordingly.

Cylinder liners - centrifugal casting, ensured that the fine grains were on the wearing surface.
 

·
Premium Member
Joined
·
142 Posts
The original fault was an inclusion in the crankshaft forging, below 2 microns, which is the limit of resolution detection. This caused the initiation of a crack which propagated over time with the crankshaft cyclic loading, until it ran completely out of the journal and through the web.......
Dave
2 microns? How sensitive! We has a 75LB6 Doxford crank fracture across No 4 aft side crankweb, It had initiated in the internal fillet between web and crankpin and when we broke it open we found a real inclusion - the remains of a half-round file(K)
 

·
Registered
Joined
·
6,271 Posts
2 microns? How sensitive! We has a 75LB6 Doxford crank fracture across No 4 aft side crankweb, It had initiated in the internal fillet between web and crankpin and when we broke it open we found a real inclusion - the remains of a half-round file(K)
Ha ha ha! Like the old joke about the 2/E and the torch - and it was still on!

That would have been on the British bulkers with the "weld filling"!

Yes, MAN B&W have a limit of 2 microns.

Rgds,
Dave
 
21 - 33 of 33 Posts
Top