Comparing the 278A to the 567B Internally

One thing I am often asked by railroad friends when discussing the Cleveland 278A, is “Just what is the difference between an EMD 567, and what parts crossover?” Well, its a simple answer, They are two very different animals, and have zero parts crossover. I figured I would throw this gallery together of doing a full gasket renewal on a 278A, which shows the differences, at least in the cylinder and head area of the engine. Something to keep in mind – the 201A was the father of both this engine, as well as the EMD 567. EMD (EMC) engineers went off and designed the 567 from the mistakes of the 201A with the goals of a railroad engine, and Winton went off and designed the 248 which was the marine service engine, which evolved quickly into the 278/278A as I have mentioned in past articles.

The 278A is best compared to the EMD 567B, in that it uses a water deck style liner. The 278A uses an individual water deck area specific to each cylinder, not one section of the block per say, both of which are sealed with O rings, which cause leaks. Leaks are bad, especially when water gets into the oil side of things. Lets dive into a 278A and go through the process of finding a leak, and how to fix it, and compare the engines along the way. Click on all of the below images for larger versions.

The first sign of trouble, is generally when you see water coming out of the airbox drains. This at least narrows it down to one side, so then you start by pulling all of the covers off, and looking for Niagara Falls.

The leak itself can come from two very different issues – The liner O rings are leaking, as seen in this image. Follow the brown trail of water from the bottom of the airbox, leading up to the liner. A second way of spotting the trouble, is when blowing down the engine, you will get water vapor (or solid streams) coming out of the blow down. This is a second area that leaks, the O rings between the head and the liner. When this happens, water will run down into the liner, and out the intake ports in the liner into the airbox – if the piston is down. This will also cause the expansion tank/water side of the engine to pressurize. More on this shortly. A cracked head can cause this same issue, which can be a bit more troublesome to pin point, especially if it only does it when they are warm.

A third source of water issues on a Cleveland can be in the exhaust elbows. EMD’s have an integral exhaust path, whereas the Cleveland’s have individual exhaust jumpers between the head and the manifold which are all water jacketed. Look closely and you will see pitting in the elbow, causing water to leak into the head and liner through the exhaust valves.

The fourth area of concern, and this is NOT Cleveland specific – is when a liner lets go. In this case, the engine was not blown down before startup, and the liner violently let go (it is a sound I will never forget). The side of the liner pushed out, thus the entire water system dumped out in a hurry into the airbox, once again, Niagara Falls resulted. Now, Cleveland’s have a pressure relief valve on each head, thus when this happened, the valve opened (and shot a solid stream of water out) preventing the connecting rod from bending. EMD’s don’t have this. The only damage from this was the piston rings were broken by the chunk of liner that failed. And on that note – If your running an old engine, take the 5 extra minutes and BLOW IT DOWN every time! It can save you some serious trouble!

Now, we drain the engine down..

…and start to take it apart. Fuel and overspeed lines are removed, rocker assembly removed, exhaust jumper removed. In this case we pulled the injector also, but it is not required.

Sometimes it takes some creativity to get all the tools in there in certain spots on these engines. In this case a chunk pf bent pipe from a handrail was used as a cheater bar. Hey, it worked! 4 nuts hold down the head to the block, and 6 bolts hold the head to the liner. EMD’s the head sits inside the block, but the Cleveland’s sit on top of the block, and do not use shared crab nuts.

No difference between them here – Lots of rags, and piles of parts!

With the head removed, we see one of the biggest downfalls of the 278A. There is a half moon shaped groove that fits a round rubber seal. When these seals start to go, the engines leak oil, and badly. In the 1950’s, the US Navy devised a tool and a process to push the head back onto this seal before you torque them down. This helps, but not that much. This, is why almost every Cleveland is covered in oil. The valve cover gasket seals (two per cover) are not any better, and are very temperamental. EMD built a box around their oil leaks…

Bottom of the head. Nothing special here. The left is toward the center of the engine. The large hole to the right is where the injector control rods goes into the head.

Now we have the head off. Pulling the head is typically the hardest part of the operation, as the stud holes will fill with oil and sludge, as well as carbon – more on that one shortly. Cleveland’s use direct air start, meaning no starter motor. 300PSI (up to 600 was used on the Sub’s) is directly admitted to the head in order of timing in 8 of the heads – that’s what the small line is just above the right most, lower head stud.

With the head off, we are at a crossroads. If the liner O rings are leaking, you need to forge ahead and pull the liner. If you are getting water through the liner, and the expansion tank is bubbling off – that means you lost a fire ring, which is a solid copper ring that seals between the head and the liner. When this fails, it burns out any number of the 12 small rubber O rings that seal the water side of the head and liner on small copper ferules. We dubbed these the little rubber douchebags. If the liner is still sealing good, you don’t HAVE to pull it, but it is typically good practice to just replace them all while its apart.

To pull the liner, you need a liner lifting plate (we had to make one, we now have an OEM one..). The piston crown has a small tapped hole for an eyebolt, thus you can secure the piston to the plate, and lift them out as one assembly.

But first, you need to pull off the bearing shell from the connecting rod. Cleveland’s use an individual rod and bearing for each cylinder, EMD’s double up with their fork and blade style assembly.

In this one, we pulled the piston out as the rings needed to be changed. 278A’s use a traditional wrist pin assembly, whereas by the later 567A engines, EMD switched to the floating piston.

Pulling the liner out – Cleveland’s have two liners available, a cast style, and a later fabricated style, which are interchangeable. The two ferrules that are closer together are the markers telling you this is the outer edge of the liner.

Liner and piston/rod is out. The rusty area is the where the water enters the liner.

Looking down through the airbox, you can see the water deck area. Water enters through the water manifold that runs through the airbox into the liner. The EMD 567C and Cleveland 498 engine simply used a bolted on extension from the manifold pipe directly to the liner, eliminating these O rings. These O ring seats can actually be changed, but it is an enormous project. You can see here how Cleveland’s have individual connecting rods on each connecting rod journal.

A full gasket kit for a Cleveland, which encompasses all new O rings for numerous things, cork seals for the covers, exhaust jumper gaskets and many other small gaskets.

With the O ring grooves cleaned up, the new O rings are installed and set in a bead of silicone, which helps seal them, as well as keeping them from pulling out when installing the liner back into the block. The liner is lubricated with dish soap to help it slide in.

With the liner back in, a new copper head gasket is installed, new fire ring, and new O rings for the water jumpers and oil seals on the rear of the head. A large cork circular gasket seals the injector control rod hole and the head. EMD’s essentially have all of this combined onto one gasket. Note that the liner is not fully seated down – This will be remedied when the liner is bolted to the head, and it is all cinched down.

With the head back on, everything else goes back together fairly quickly. The head is bolted to the liner with 6 bolts, which are torqued to 175 ft lbs, (290 on the 567B), and the 4 bolts holding the head get torqued down to 650 ft lbs (1800 on the 567B). The exhaust elbow uses two copper clad gaskets, as well as two smaller gaskets for sealing the water side (Cleveland’s are water cooled exhaust). It is a bit tricky to put them on, as the lower bolts are fine thread, and the upper bolts are coarse thread, so you cant mix the bolts up. You essentially have to roll the jumper on, starting with tightening the lower bolts first, in order to compress the gaskets, and line the top up.

Hopefully this answers some more of the mechanical questions of how the 567 and 278 engines compare internally. Down the road I may do something a bit more detailed on this and cover the other portions of the engine, and how they are different, such as the blower, oil and governing systems. Something to note about the Cleveland’s – they require no special tools to take them apart, outside of a torque multiplier.

One thought on “Comparing the 278A to the 567B Internally

  1. I was paging through this post when I realized that my dad would be going ape shit over this blog. Then the memory came back – nine years old handing my dad tools as he worked on a diesel power plant onboard one of the party boats in Sheepshead Bay (NYC). He was a marine engineer who swallowed the anchor. He’d swap maintenance work for fishing trips – the reason I was never seasick when I went to sea in the Navy. thanks for a wonderful blog.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s