Tech Articles

Ford’s Legendary Y-Block—Still In Style

(Image/Jim Smart)

Ford’s first overhead valve V8, the Y-Block, remains the engine of choice for a lot of hotrodders. By modern standards, the Y-Block is dated with its stacked intake ports and solid tappets that must be installed first.

It is also—by far—the most handsome V-8 ever done. Introduced in 1954 to replace the time-proven flathead and displacing 239ci (Ford) and 256ci (Mercury), it was a bold first step by Ford to get away from the venerable flathead. It was long overdue.

In 1955, Ford pumped up the Y-Block’s displacement to 272ci and 292ci. In 1956, Ford grew the Y-Block to an optional 312ci as a bold displacement option in a competitive market. Ford had renewed respect from the competition and the close attention of car buyers because it finally had a powerful modern V8. In 1957, Ford offered two optional 312ci performance engines with either two four-barrel carburetion or a McCullough supercharger for those with additional cash on their hands during a bad economic recession. Having an “E” or “F” engine code Ford was real status in 1957.

When the Ford FE Series big block was introduced for 1958, the Y-Block took something of a back seat to these new mills, considering the larger FE engine’s displacement and power. There was also the Ford MEL Series fat block V8 for Mercury-Edsel-Lincoln luxury carlines, which was in a heavyweight class by itself. The Y-Block lived in production through 1964 and was a hard-working heavy hitter because it found purpose in so many applications. To this day, enthusiasts love the Y-block for its throaty sound and chatter of 16 traditional solid tappets. There’s just nothing else like it.   

Two engine builders stand out more than any other when it comes to Ford’s Y-Block—Ted Eaton and John Mummert. There’s no one else in the world who knows more about the Ford Y-Block than these two gentlemen. Mummert calls San Diego, California home and is well known and respected for his Y-block efforts, including aluminum cylinder heads and induction systems for these legendary Ford V8s. Ted Eaton in Lorena, Texas knows the Y-block like few others. He builds a lot of them.

John Mummert of John Mummert Machine tells us block identification numbers are found on the side of block above the oil filter on Cleveland blocks (“CF” logo). Blocks cast at the Dearborn iron foundry (“DIF” logo) have block identification numbers near the distributor in back or above the generator. Most Dearborn foundry blocks were used in trucks, yet no Dearborn Y-blocks were produced after 1957, which adds to the confusion. Heavy-duty trucks were fitted with steel cranks and used C1AE or C2AE blocks, which were produced for both car and truck lines.

Nearly all Dearborn blocks cast after 1954 were 272s. Most 292 and 312 blocks were cast at Cleveland. John tells us it is generally believed no 312 blocks were ever cast at Dearborn. He also confirms 312 block casting numbers as ECZ-6015A, ECZ-6015B, ECZ-6015C, EDB-6015E, B9AE-6015F. Of these the ECZ-6015A and ECZ-6015C are the most common.

John also tells us some 312 service replacement blocks are numbered C2AE-6015C. What’s more, he adds 312 main caps are always marked “ECZ” while all other Y-block main caps are marked “EBU.” He stresses this is the only positive way to correctly identify a 312 block.

Although the 312 remains the most popular Y-block, you can get by with a 272 or 292 block with great success. John says nearly any 272, 292, or 312 block can be used for performance applications if massaged correctly. The 272 block can be bored to 292ci without consequence. Since all other internal components are the same across the board, you can cross pollenate. He adds that 292 pistons are easier to find, cheaper, and have a better ring selection.

John goes on to tell us the 292 blocks from 1955 to 1964 are easier to find. In fact, improvements were made to 292 and 312 blocks in 1959 with deeper drilled main cap threads for added strength. The 1961 to 1964 C1AE and C2AE blocks have additional material in the main webs for added strength. These blocks typically don’t sonic check as thick as earlier blocks according to John. Therefore, he adds, if a good early block is found, drill and tap the main bearing cap threads deeper and use the early block.

Cylinder Heads

The ECZ-C cylinder heads were used across the board on the 272, 292, and 312 engines. It is good to opt for stainless steel valves and hardened exhaust valve seats for use with unleaded fuels. The most desirable Y-block heads for increased compression are the 1957 through early 1958 ECZ-G castings with intake valves sized at 1.927 inches, according to Mummert. Combustion chamber size is approximately 69cc.

For slightly lower compression for today’s pump gas is the 1958 to 1959 “5752-113” casting. These heads have the same 1.927 inch intake valves and a slightly larger chamber, which lowers compression. For low-compression engines the 1959 “5750-471” truck head is also a good choice, Mummert adds, with a 1.927 inch intake valve and 80cc chamber. It is the same head casting as the “EDB” supercharger head, yet with a different casting number.

Mummert stresses some very important points you need to pay attention to. Never throw away your old Y-Block camshaft until you’ve captured the thrust spacer. The new aftermarket cam will not come with one. If you’ve already thrown it away, check with John—he may have one.

John explains there appears to be two different length head bolts for the Y-block—five short bolts near the spark plugs and five longer bolts under the rocker arm assemblies. The five short bolts near the spark plugs are identical but the five under the rocker arms are not the same. Two of these cylinder head bolts are slightly longer and installed at the outer ends of the cylinder head where alignment dowels are located.

It is good advice to lay all 10 long bolts next to each other on the bench and you will find four longer and six that are about 1/4 inch shorter. Installing the longer bolts in the center three holes can cause them to bottom in the block, which can be disaster. Late production Y-block engines have only long and short bolts.

John adds both Y-Block cylinder head gaskets are identical. It might seem that the same face of the gasket would go against the block and the opposite face would go against the head on each side. Mummert stresses this is not true. What is critical is that the open coolant holes are located at the back of the head and the blocked portion of the gasket is at the front. Otherwise you will experience overheat due to no circulation. Look for the word “FRONT” on the head gasket and place it at the front, even if it looks upside down. This places one of the gaskets face up and one face down.

Note there is a square corner at one end of each gasket. The square corner must be at the front of the engine. This can be checked without removing the heads. If you are having overheating problems, check for these square corners at the top front corner of the head near the intake gasket to confirm installation.

John stresses if you’re using a camshaft with a cross-drilled center journal you must use 1955 through early 1956 cam bearings designed for cross-drilled camshafts. If you’re installing a cam with a grooved center journal you must use the late 1956 to 1964 cam bearings. If your cam will not fit in the block check it for trueness. John has seen some cams with up to 0.010 inch runout, which should not be used. Another issue seen is the front cam bearing cocked in the bore. Install the front cam bearing from the rear to ensure proper installation.

Never throw away your old Y-Block camshaft until you’ve captured the thrust spacer.

Mummert repeatedly stresses installing rocker arm shafts right side up for lubrication purposes. He adds rocker shaft stands are identical and will bolt down either way. However, the oil hole in the shaft must be aligned with the hole in the shaft stand and is at the bottom when the stands are bolted down. Get this wrong and you will starve the rocker arms of oil and wind up with valvetrain failure and engine damage.

The 312’s main caps are taller than the 292’s main caps. However, the 312’s #5 main cap is at the 292’s height to clear the rear main seal holder and the oil pan rail. This makes it possible to install any of the 312’s longer main cap bolts in error from the first four main caps in the #5 cap. They will bottom out. Some blocks are drilled deep enough to accept the longer bolts in the rear cap. Do not do this. There have been a few instances where the rear main saddle of a 312 cracked during assembly because the longer bolt was used and bottomed out.

Another issue has long been the incorrect torque specification of 120 ft.-lb. main cap bolt torque according to John, which was printed in all 1956 Ford Shop Manuals. This figure is excessive and has undoubtedly cracked main webs in 312 blocks. Torque main cap bolts to 95 ft.-lbs. It is also critical to check the amount of thread that will be engaged in the block before torquing.

Do not use main cap bolts in any Y-block that don’t reveal at least 7/8 inch of bolt thread when placed in the main cap. This may require running a bottoming tap into the main bolt holes. Later 292 engines have significantly longer main cap bolts, an indication that Ford engineers realized this need. Do not use bolts or studs that engage more than 1-1/8 inches of thread because the oil passages to the main bearings will be blocked.

Mummert also tells us not to use excessive-length bolts for the intake manifold. The intake manifold bolt holes in the cylinder head intersect push rod passages and too long a bolt can interfere with the push rods. Also be certain that the bolt holes in the heads at the rear of the manifold are plugged. These are the threaded holes that are unused but are drilled through into the push rod passage. Use short bolts for this application, with a 1/2 inch of thread to avoid the push rods.

Timing cover bolt length is yet another issue. If these 5/16 inch bolts are too long they will contact the front two cylinders and do severe damage. Apply sealer to bolts that enter the water jackets or you will have leakage. They are the two bolts above and below the water passages.

Although Y-block cylinder heads can be interchanged to either side of the block, don’t do this. Years of exposure to coolant and the resulting corrosion, the 0.906 inch holes at the front of the intake surface will not accept a freeze plug. When you’re installing cylinder heads, make sure the corroded 0.906 inch hole is located toward the front of the engine. The corroded hole can be reamed to a larger size and an oversize freeze plug installed. Remove all oil galley plugs and the oil filter adapter before having your block cleaned. Always have the block decks surfaced, main bearing bores align honed, and head decks surfaced. These castings distort after years of operation and heat-cycling.

John tells us Ford used two different thrust washer thicknesses and cam plates, which calls for your close attention. If you use the wrong combination of thrust washer and cam plate, there will be zero camshaft endplay and failure is certain as the engine heats up. Ensure you have at least 0.004 inch camshaft endplay.

And finally, some replacement camshaft cores typically have a glob of metal between the last lobe and the distributor gear. On high lift cams, this glob can be higher than the base circle of the lobe, which can do incredible damage. Place a lifter on the last lobe base circle and be sure the lifter clears this glob.

Engine Masters Challenge

Ted Eaton of Eaton Balancing entered Ford’s Y-block in the Engine Masters Challenge at the University of Northwestern Ohio many years ago to see what could be done with the Y-Block in brute competition. For the 2007-08 Engine Masters competitions years back, he built a 312ci Y-Block with .022 inch overbore and a stock stroke to get 316ci. The following year, there were major rule changes that affected his agenda. One rule change was increasing the rpm range in these Engine Masters dyno pulls with compression as high as 11.5:1.

He told us although the 316 looked good in the 2,500 to 6,500 rpm range, the rules placed his Y-block much lower in the field due to the allowance of roller camshafts and increasing rpm to 3,000 to 7,000 rpm. Ted added that he was already at a disadvantage by having to use stock iron heads given these revised guidelines. He just wasn’t going to do it.

Ted’s plan for this engine was a “square” 4.000 inch bore and stroke to get 403ci. He told us this combination probably would not have given the best results but would have given him impressive peak power numbers. Ted commented the 4.000 x 4.000 inch bore and stroke combination created water problems when he went wide open throttle at low rpm under load on the dyno. He opted for a smaller 3.859 inch bore coupled with a 4.000 inch stroke to get 375ci along with additional cylinder wall thickness from a “B9AE-F” 292 non-steam hole block. Ted told us though 3.700 and 3.800 inch strokes had been tried successfully with Y-blocks, he wanted to know how 4.000 inches of stroke would work with a 292 block. It was a risky proposition.

Ted went with a 292 crankshaft offset ground for mock-up purposes to get the 4.000 inch stroke coupled with a 1.889 inch Honda connecting rod journal with 3/8 inch rod bolts. This sounds like monkey motion, but it worked well for Ted. He cut a single rod journal down to Honda size and went to work with the mock-up phase. His concern was getting 3/8 inch rod bolts to clear the camshaft while still getting the lift he needed.

Ted found a smaller base circle camshaft was not going to work because he was already planning a camshaft with 0.376 inch lobe lift to get the 0.600 inch plus lift valve lift he wanted. Anything smaller than 1.150 inch would result in the lifters falling out of their bores. He added the 1.150 inch base circle camshaft would clear rod bolts only if he opted for 0.337 inch maximum lift lobe.

When Ted did the mock-up and confirmed he could get a 4.000 inch stroke inside a 292 block, he ordered what he needed to build and flog this engine. He ordered a 4.000 inch stroke Moldex 4340 billet steel crank with 2.4980 to 2.4988 inch Y-Block main journals and 1.889 inch Honda sized rod journals—obviously a custom grind. Instead of a Y-block crank flange, he opted for a FE/385-series big block crank flange to make it easier to find an SFI approved steel PRW flywheel mandated for the Engine Masters Challenge.

Ted went with 6.750 inch long Oliver connecting rods with Honda sized journals to get a 1.68:1 rod ratio. His wanted to make this engine less vulnerable to detonation. To get the 375ci desired, he went with Diamond ceramic-coated pistons with special low-friction skirts for reduced friction. Compression height was 1.000 inch, which put the wrist pin into the ring lands. This meant he needed an oil ring support rail to prevent oil ring distortion. A 0.866 inch wrist pin and 9cc reverse dome got Ted the desired 10.1:1 compression ratio. Actual working compression was 8.0:1 with the cam advanced 1-1/2 degrees and valve lash at 0.025 inch hot. With the decided dated Ford iron heads, this was how Ted did it with 91 octane fuel. With aluminum heads, his 2009 Engine Masters entry could have enabled a higher compression ratio.

Ted tried seven different intake manifolds with a variety of carburetors and spacers. The final choice was the Mummert intake manifold and a Holley 950HP (834 cfm) and stepped headers (1.75 and 1.875 inch tubes feeding into modified 3-1/2 inch collectors and mufflers. Ted managed to get 462 to 464 horsepower and 446 to 449 lbs.-ft. of torque through the mufflers on the dyno, which was good considering 10.1:1 compression on pump gas. Ted tells us they tried a 1,050 cfm Holley Dominator and got tremendous torque from this combination.

At the 2009 Engine Masters Challenge, the engine fired nicely and ran smoothly at 900 rpm. He jetted the Holley carb at 77/89. It had been jetted at 75/87 in Texas. The jet change was a good decision with the Eaton engine making 433 horsepower and 416 lbs.-ft. of torque. A year later, Ted had the Mummert aluminum head advantage at the Engine Masters Challenge.

Ted felt it was time to bolt these Mummert heads, fully ported, onto his Y-Block test bed to see how these heads would do over fully ported iron castings at the 2010 Engine Masters. He gained another 72 horsepower over the ported iron stockers. He gave his 375 inch Y-Block seven quarts of fresh Valvoline 20w50 and put it back on the dyno for some road work. He managed 519 horsepower with the Mummert aluminum heads and induction package.

The Mummert heads were the only change. He then tweaked ignition timing to 32 degrees total to establish a baseline and without detonation. That said, he was able to push total timing to a whopping 38 degrees BTDC. He installed a modified and ported intake from Mummert with a carb spacer. The 375 suffered a minor loss in power at high rpm yet gained torque and average horsepower in at 2,500 to 6,500 rpm.

Ted took his 375 Y-Block to Ohio again for a series of floggings. With total timing at 38 degrees BTDC, Ted became concerned over potential detonation issues with such aggressive ignition timing. The decision was made to risk detonation and leave ignition timing where it was. Engine Masters rules in 2010 allowed for up to 11.5:1 compression. Static compression was 10.8:1. Ted opted for an Isky cam with 254 degrees intake duration, which figured in at 8.5:1 working compression.

Ted decided to let the oil get hotter in the sump while allowing fresh coolant into the engine at the same time. This netted Eaton 523, 521, and 524 horsepower for a total score of 2205.7 points—eye-opening performance from a vintage Ford Y-Block. The 375 came in 13th overall at the 2010 Engine Masters Challenge against a lot of tough competition (including Chrysler’s Hemi).

In the years since, Ted has returned to the Engine Masters Challenge with a succession of Y-Block entries—a 362-inch in 2015 with modified aluminum Mummert heads, a dual-plane manifold, and a Holley Dominator that came in third in a highly competitive field. Ted cammed these Y-Block entries for a street or drag racing, not necessarily dyno competition. In 2016, Ted returned with a 375ci Y-block conceived strictly for the Engine Masters Challenge. He came out on top with high honors.

In 2018, Ted returned to the Engine Masters again with a Y-Block engineered to turn in impressive numbers. He ordered up a Moldex 4.300 stroke crank. Custom made Jim Cunningham 6.500 inch connecting rods with a 45 degree rod cap were produced specifically to get the rod bolt to clear the camshaft. This enabled the tight confines of the block to handle a 4.500 inch stroke without rod bolts hitting the camshaft.

Ted kept the bore at 3.862 inches to achieve 403ci. Compression ratio was 13.0:1 with flat top pistons. Isky cam specifications were 0.629/0.639 inch valve lift with 256/260 degrees of duration at 0.050 on a 102 lobe centers. The real work was on top in the Mummert cylinder heads, which were ported along with a change in valve geometry to 16 degrees. Spark plugs were moved higher to near the roof of the chamber, permitted by the use of a smaller spark plug.

The Y-Block heads were originally designed with stacked intake runners to achieve equal-length runners. Where it got wild and crazy was custom work by Joe Craine to turn the Y-Block’s induction into a tunnel ram topped with Holley 660 cfm center-squirters. Where Ted got into trouble in 2018 was detonation issues on the dyno and the loss of the 403 engine. Still—he managed roughly 550 horsepower, which is remarkable for Ford’s venerable Y-Block.

Ted managed to take his Y-Block where no Y-Block had ever been before.

This is a B9AE-F 292 block with 3,859-inch bores, perfect for Ted’s efforts to build real power. It is the rugged nature of the 292 block that makes it an acceptable platform. (Image/Jim Smart)
Here’s another look at the 292 block, which was the best choice for a high-horse Y-Block. (Image/Jim Smart)
From this side, you can see the cross-bolted main caps and a Ted Eaton-fabricated billet stud girdle. (Image/Jim Smart)
Ted opted for a custom Isky Racing Cams grind with 109-degree lobe centers, which was advanced 1.5 degrees to 107.5 degrees intake lobe centerline. That and duration at .050 inch at 254/258 intake/exhaust and lobe lift at .370/.376 inch intake/exhaust with 1.6:1 rocker ratio. Ted dynamic balanced this cam and sprocket for added smoothness. Note the use of moly lube on the lobes and engine assembly lube on the journals. (Image/Jim Smart)
Head on, we get a look at the Y-Block’s skirted demeanor. Seems ironic Detroit got away from skirted blocks to save weight, yet it is doing skirted blocks again in 2021. The old tried and proven skirted block worked. (Image/Jim Smart)
Ted saw a real need for bottom end strength. He studded and crossbolted the mains and fabricated his own billet stud girdle. (Image/Jim Smart)
The beauty of Ford’s classic Y-Block is its simplicity. A simple straightforward American V-8 much as the flathead was. Ted Eaton has taken this design and improved on engineering with a bulletproof bottom end able to handle 500 horsepower. (Image/Jim Smart)
Ceramic-coated Diamond forged pistons will be able to handle any potential detonation issues. Ted would tell you an ounce of prevention. (Image/Jim Smart)
Ted looked to Moldex for custom cranks machined for Honda rods that ultimate led to 375ci and 403ci. He specified FE/385 Series flanges for these cranks for a broader selection of flywheels.. This billet crank isn’t fully counterweighted because Ted’s goal less reciprocating mass. He balanced this crankshaft to a 1630 gram bobweight including considerable overbalance or an increase in the percent value used for reciprocating parts. (Image/Jim Smart)
A trio of Y-Block connecting rods. From left is the custom Oliver rod with Honda dimensions. The two on the right are stock Y-Block rods. Shot peening connecting rods also makes them stronger. For 272 and 292 engine stock builds, use the 1962-1964 “C2AE” forgings. For 312 engines use the 1961-1964 “C1TE” heavy-duty truck rods if you can find them. Cast pistons will work fine for most street applications. If you’re going to push it, opt for forged pistons. (Image/Jim Smart)
To infuse displacement into his Engine Masters build, Ted had these custom Oliver parabolic rods made—ultimately arriving at 375ci and 403ci. They provide strength and manage to clear everything in the process.
Ted opted for Diamond custom pistons for this unique Y-Block stroker. Total Seal rings were 1.2mm wide top and second with oil rings at 3.0mm. The wristpin is 0.866 inch with a 9.0cc reverse dome to keep compression at 10.1:1. Dynamic compression is 8.0:1 with the camshaft 1.5-degrees advanced and valve lash at 0.025 inch hot. (Image/Jim Smart)
The Y-Block cast iron head is such an odd duck when it comes to American V-8 cylinder heads. Stacked intake ports versus traditional side-by-side. By contrast, the larger Lincoln Y-Block (1952-57) head sports side-by-side ports. Ted chose 1959 vintage 292 iron heads (#5752-113), enlarging valve sizes to 2.02/1.56-inch intake/exhaust along with in-depth port work. (Image/Jim Smart)
Aggressive port work on the exhaust side of these 292 heads improves scavenging. (Image/Jim Smart)
These heads have also been milled, downsizing the chambers to 65cc. (Image/Jim Smart)
The Mummert aluminum heads are game changers for the Y-Block movement. Just aluminum alone improves heat transfer and reduces the risk of detonation. It also enables you to run greater compression. (Image/Jim Smart)
These heads offer redesigned intake and exhaust ports along with 65cc chambers. They accept stock manifolds and valve covers. (Image/Jim Smart)
We like these high swirl 65cc chambers, with vastly improved quench and 1.940/1.740 inch stainless steel intake/exhaust valves. Valve shrouding is minimal. (Image/Jim Smart)
Cylinder head gaskets on the Y-Block must be installed where cooling passages are at the rear and “FRONT” is in front. The left bank will look upside down with the word “FRONT” facing down. Right side, this side, will face up. (Image/Jim Smart)
Mummert offers two basic dual-plane intake manifolds for the Y-Block—standard height for 1955-56 Ford/Mercury and reduced height for hood clearance for 1955-57 Thunderbird. All of your standard Holley flange carburetors will fit. Ted has reworked the Mummert plenum as shown including dimples that help keep fuel droplets in suspension. (Image/Jim Smart)
Much longer than stock valves have been employed, which resulted in the rocker shafts being relocated 0.315 inch higher to get rocker arm geometry right. Both Dove and Rocker Arm Specialist 1.6:1 roller rockers have been tested. (Image/Jim Smart)
The Y-Block’s oiling system is first rate and more than adequate for anything you want to do to a Y-Block. Chamfer the oil galley transitions and go with a high-volume oil pump from Summit and you’re good to go. Use conventional 10w40 oil along with a zinc additive. You must run a zinc additive. (Image/Jim Smart)
Seven different intake manifolds were tested on the 375 a long with a variety of carburetors and spacers to achieve best results. (Image/Jim Smart)
The final engine parts combo was the Mummert aluminum intake manifold and a Holley 950HP (834cfm) carb along with the stepped headers (1.75 and 1.875 inch tubes feeding into modified 3.5 inch collectors and mufflers. They managed 464 horsepower and 449 lbs.-ft. of torque through mufflers. (Image/Jim Smart)

Tags: , , , , , ,

3 Comments

  1. tom marrapode says:

    had a stock 312 in a 49 coupe, fun car.
    wish i had niw

  2. In my opinion, the Y-block Ford certainly was NOT the most handsome V8 EVER PRODUCED.
    I can think of several that were better looking!
    I happen to think the Buick Nailhead is more appealing, but I also really love the Chrysler Hemi, and consider the Chevy 348/409 the top of the line, when it comes to looks!

  3. I love the old y-block but a fe 427 built right will always be bad to the bone.

Leave a Reply to Mike Cancel

Your email address will not be published. Required fields are marked *

*

This site uses Akismet to reduce spam. Learn how your comment data is processed.