A 235 For Severe Service

These notes describe the effort to build a 235 that would hang together during a 1300-mile trip through the mountains of eastern Mexico. That objective was achieved with only one unscheduled stop to clear dirt from the carburetor inlet needle.

In the larger sense, the project failed.  Knowing better, I entertained the hope that careful fitting and long nights spent thinking about technical problems might be enough to transform the 50-year old relic into a bulletproof, zero-problem engine. That did not happen. Below 4400 rpm, oil consumption was almost nonexistent. Exceeding that speed cost nearly a quart of oil an hour. Valve clearances opened to an alarming degree for reasons that are still not clear. In addition, the distributor cap, while remaining functional, exhibited severe spark erosion at the undersides of the terminals. These malfunctions are discussed below.

Readers may disagree with some of the choices made. And there are some things that I would do differently the next time around. But the old girl held together at speeds of between 4200 and 4400 rpm for hours at a stretch.

The 235 had been run only sporadically since it was professionally rebuilt in 1975 and installed in a 1950 ½-ton pickup. Work that had been done in the interim had been conditioned by the notion that the truck would be used for short hauls, puttering around town, that sort of thing. My job was to undo the effects of years of idleness and the history of expedient, "good enough" repairs.

Removing the cylinder head to replace a leaking gasket brought encouraging news. There was little carbon in the chambers and the ridge that defines the upper limit of piston-ring travel was almost invisible. It looked as if rebuild was holding.

Now that the head was off, it would have been the time to send it out for valve-seat inserts, advertised as protection against the effects unleaded fuel. Yet, the old man was skeptical about that, since the most of the inserts he's seen were made of cast iron, just like the head. Stellite inserts cost $10 each and installation can be problematic. The worst Pemex gasoline could do would be to accelerate valve-seat erosion; a loose insert would be a catastrophic in rural Mexico. Running the carburetor slightly rich seemed a better alternative.

While the engine was down, I had the manifold assembly milled and replaced the water pump (good insurance), the alternator (noisy bearings), fan belt and radiator hoses. The thermostat was tested in boiling water and worked okay. A Flex-Fan was installed to reduce noise and, hopefully, improve mileage a bit. Mexican gasoline costs US$ 2.50 a gallon.

Then the Saginaw four-speed transmission, installed ten years previously, set up a howl in all forward gears. New main and cluster-shaft bearings cured the problem, but in the course of the work, I noticed that the flywheel pilot bushing had a quarter-inch or more of slop in it. No wonder the transmission failed.

The OEM bushing was brass; aftermarket bushings are made of something harder and obviously less durable. An alternative might be to substitute a needle bearing of the type used on GM diesel cars. A quick test of the input shaft hardness with a file scotched that idea. The shaft was soft, too soft to survive direct contact with needles. An aftermarket bushing it would be.

Upon installation, the fit between the input shaft and bearing was still sloppy as it must be on a part intended to fit all input shafts, regardless of wear and manufacturing tolerance. In a better world, one would machine a bearing from brass, drive it home, build an alignment fixture and ream the ID for .0015" or so of running clearance. I packed the bushing with grease and hoped for the best.

A 50-mile test drive confirmed that the transmission worked fine. The next morning there was a puddle of oil on the driveway. The rear main seal had failed.

In the old days, people replaced rope seals from under the car. I have done it myself. But attempts to replicate that youthful feat failed. The engine would have to be pulled.

As a point of interest, the 235 has a reported weight of 630 lb, which means that it is 40 lb heavier than first-generation V-8 small blocks. A standard, car-parts-store engine stand is barely up to the job, even with the hold-down bolts replaced by Grade-8 fasteners.

Engine Identification

The date the block was poured is recorded on the right side of the engine, just above the starter motor. Look for a raised (not stamped) alphanumeric code consisting of a letter followed by two or three numbers. The letter signifies the month: A means January, B, February, and so on. The one or two-digit number than follows is the day of the month. The last digit is the year, i.e., 2 represents 1952. The block in question was labeled E 12 4, which meant that it was cast on n May 12, 1954.

Oil Seals

Because the engine had spent so much of its life in storage, the crankshaft flange was severely rust pitted. I mounted the crank in a lathe and polished out most of the pitting, although some remained.

No modern full-circle seal is available for the 235 and adapting one looks like it would be a major engineering job. At any rate, Chevrolet was not overly concerned with rear seal leaks. The rope seal was introduced on the 216 in 1940; prior to that time, the crankshaft slinger was solely responsible for oil retention. Older readers might remember the oil streaks down the middle of concrete highways.

A special, and expensive, rope seal was used on the 235 during 1941, its first year of production. Subsequent 235s reverted to the 216 seal. Today, Fel-Pro markets a replacement seal for all 1940-54 216/235 engines as PN 5048. The Victor equivalent fits the same range of models and seems easier to install than the slightly wider Fel-Pro part.

Old-time mechanics soaked rope seals overnight in oil to assure adequate lubrication during startup. Fel-Pro warns against this practice and says that a few drops of oil prior to assembly are enough. My experience with oil-swollen seals has made me a believer. Once the rear cap is torqued down, the seal can be flooded with a squirt can.

When these engines were new, dealers had access to factory seal compressor tools. Contemporary practice is to roll the seal halves into their grooves with a ¾" rod or dowel pin, working from the center out. Fel-Pro and Chevrolet say that the seal ends should be cut flush with the bearing mating surfaces; most working mechanics trim the ends to stand 1/8" proud above the cap and block to provide material for additional compression when the crankshaft is bolted down.

I followed shop practice and left the seal ends protruding. The surplus material acted as a shim, opening the No. 4 main bearing clearance to .0035" or .0015" greater than the average for the other mains. In retrospect, it would have been wiser to torque down the crank over the seal ends, remove the crank and trim the flattened ends flush with the bearing faces. A dab of RTV sealant would have assured oil tightness.

A new front seal was installed and centered with the homemade nylon tool.

Upon assembly, the front seal remained dry under hard usage. The rear seal leaked a bit initially, but settled in after a few miles.

Crankshaft Bearings

235 rod bearings were babbitted from the introduction of the engine in 1941 through 1954 standard-shift models. As listed in the table below, insert bearings are available for these engines, although expensive.

Auto parts houses can usually supply rod-bearing inserts for 1954 and later 235 and 261 engines in standard and .010," .020" and .030" undersizes. The forged crankshaft may tolerate more grinding, since 040." .050" and .060" under bearings were once catalogued.

Rod Bearings

3.562 x 3.938 235 CID (2.9L) & 3.750 x 3.9838 261CID (4.3L) Engines

Main bearings were available in Std. and 10-20-30-40-50-60 undersizes.

Main Bearing Dimensions
3.562 x 3.938 235 CID (2.9L) & 3.750 x 3.9838 261CID (4.3L) Engines

The original rod and main bearings were well within the generous clearances show in the tables above. But inspection showed that the bearings were impacted with dirt. (Thick babbit shells were the last line of defense for engines without oil filters.) In addition, the thrust bearing faces were dented, as if by a ball-peen hammer. The resulting drag had nearly sheared the locating pin. I polished the crankshaft and replaced all bearings, lubricating them with assembly grease.

Oiling System

The rebuilder had peened over the bleed ports on the big ends of the connecting rods, apparently in an attempt to boost oil pressure. Whatever the motivation, blocking off these ports starved the camshaft and reduced the flow of oil to the lower cylinder bores. I drilled them out to some approximation of their original size.

The rings looked good, and I didn't replace them. As mentioned previously, oil consumption on the trip was almost nil, so long as speeds were held below 4400 rpm. Above that speed, the engine drank oil, consuming 3 ½ quarts during one five-hour period. New rings might have reduced that consumption although the 235 was not designed to live at speeds approaching 5000 rpm.

Some 235's came with bypass filters hung off the intake manifold and connected in parallel with the main oil gallery. Only a fraction of pump output goes to the filter, but the house always wins. For example, suppose that 1/10 of pump output is filtered and dumped back into the sump. Now we have oil that is 90% unfiltered. The next pass gives us 81% unfiltered oil, then 73%, and so on. Elegant.

Modern cars use inline filters, in series with pump output. A Google search will show how the 235 oiling circuit can modified to send the oil through a filter before it goes to the bearings. In the example I found, oil to the inline filter has to pass through the same 1/8" pipe thread fittings that Chevrolet provided for a bypass filter. The pressure drop through these small-diameter fittings would restrict flow and cause the pressure relief valve at the pump to open sooner than it normally would. When GM added inline filtration to the 261, they used ¾" ID hose.

Probably the best approach would be to use a very fine bypass filter in combination with a free-flowing inline filter, after the example of Cummings.

I constructed a bypass filter, using a Ford element and hose with a 250-psi burst strength. Initially, the filter, which shunts pump output to the sump, dropped cold oil pressure by about 8 psi at 3000 rpm. A restrictor plate with a 1/16" orifice reduced the oil pressure loss by about half.

Before the filter was installed, fresh oil would turn black after a few days puttering about town. With the filter, the oil remained translucent after 1300 hard miles.

Because money was getting tight, the original oil pump, a 30-year veteran, was retained. Castrol 10W-50 weight oil gave hot pressure readings of 8 psi at idle and 28 psi at 4400 rpm. I was not overjoyed with those numbers, but took consolation in remembering splash-lubricated engines that thrived at high rpm. The journal itself generates pressure in the form of a standing wave; all the pump has to do is provide adequate volume to prevent the oil from overheating. The lighter the oil, the less viscosity and the less the heating effect. Idiot lights on modern cars come off with pressures in the 5-7-psi range, which may have more to do with hydraulic lifter collapse than with crankshaft bearing requirements.

At any rate, the engine showed no sign of bearing distress. Rockers received adequate, but not generous, lubrication during hot idle.

Camshaft and Lifters

The forging number 383023 steel camshaft was a solid lifter cam, which meant that the 1954 235 had begun life behind a manual transmission. This cam has the large (2.154") No. 1 journal, believed to have been standardized sometime during the middle of the 1954 production run. Earlier 235 cams had a 2.029" journal.

The original steel cam exhibited extreme lobe wear, but could have been salvaged by sending it out for welding and regrinding. In the interests of saving a few bucks, I opted to purchase a PN 251 replacement cam from Sealed Power.

As shown by the chart below, PN 251 is a mechanical cam, which means that it incorporates low-acceleration ramps, .010" to .015" long, to take up lash as the valve is opened and to ease it down on its seat. Hydraulic lifter ramps are shorter, between .005" and .010," since there is less mechanical slop to cushion. Running a mechanical cam with hydraulic lifters radically changes the valve timing. And running a hydraulic cam with mechanical lifters results in rapid failure of the lifters, cam and rockers.

I was stuck with mechanical lifters because, unlike later 235's, this engine does not the oil gallery outboard of the lifter bores necessary to support hydraulics. The presence of the gallery is indicated by a welch plug on the back of the block and visible when the bell housing is removed.

Sealed Power 235/261 Camshaft Specifications

 

Like most modern camshafts, these Sealed Power replacement parts are cast iron, not steel as Chevrolet had specified until about 1956. According to one source, all domestic manufacturers had switched to iron by 1960.

Iron and steel camshafts are easy to identify when you have examples of both in hand. Struck with a hammer, steel rings; iron makes dull thump. The chip produced by a chisel against iron crumples; steel curls with a sharp edge. Destructive testing of this sort should be confined to a noncritical area, such as the portion of the cam that bears against the expansion plug.

Iron camshafts require hardenable iron lifters, which are heat treated over their whole surface to be moderately hard. Steel cams run on chilled-iron lifters, manufactured by pouring the lifter into a honeycomb mold with a chilled plate at the bottom. OEM chilled iron lifters are made of three parts: the rubbing surface is glass hard, the body cylinder file soft, and the pushrod socket hard. Mismatching lifter and cam material results in rapid failure of both components.

Patrick's lists solid lifters in their catalog, but was unable to supply them at the time this work was underway. I purchased a set described as 235 replacements from Kanter's. The parts turned out to be NOS PN 839263 lifters, intended for 1940-47 216s. As beautiful as they are, these chilled iron lifters would destroy a cast-iron camshaft within minutes of startup. Ditto for NAPA PN 2131603, catalogued as replacement lifters for the 1954 truck.

A call to Clevite produced the information that PN 2131657 are hardened lifters with the requisite .8895" diameter. Intended for 1956-62 235 truck engines, these one-piece lifters are .400" shorter than the originals and have an hourglass configuration with a deep groove at the center. The cutaway is part of the rocker oiling circuit on late model engines. Longer pushrods were, of course, mandated.

Custom pushrods are available from Smith Brothers (541-388-8188) at a very reasonable price of $5.70 each. But the new lifter cups have a slightly different profile than the originals, and I decided to go with Clevite pushrods intended for 1956-62 235 truck engines.

Camshaft Timing Gear

As a rule, 235 truck engines came with aluminum timing gears and the passenger cars used Bakelite, reinforced with a fibrous material of unknown composition. It could well have been asbestos, since the material was used for Bakelite gears in other applications.

My engine was fitted with a Bakelite gear. For reliability, I replaced the part with an aluminum gear manufactured by Coyle. The fiber gear was sawed off, the hub drilled at the keyway and split. Repeated attempts were made to persuade the aluminum gear to slip over the camshaft nose. Instructions that came with the part said to heat it for five minutes in boiling water. That merely burned my fingers.

I took the parts to an automotive machinist, who heated the gear with a rosebud-tipped acetylene torch. It still refused to go on. At that point I should have called a halt to the attempt and purchased a fiber gear. But once begun, projects like this take on momentum. We were not going to be defeated. A ten-ton hydraulic press was called into action and, with the gauge red-lined, the gear went home. No doubt the cam was bent, but I did not have the heart to measure it.

The camshaft was installed with the necessary .001"-.003" end clearance, established by the proximity of the timing gear to the thrust plate. (The more the camshaft floats, the greater the variation in ignition timing.) Camshaft lobes and lifters were coated with molybdenum disulfide grease, the journals oiled, rockers received preliminary adjustment, and the oil pump primed with a drill motor. Upon startup, the engine was run at 1200 for 20 minutes to work-harden the wearing surfaces.

The engine was then shut down and the hot clearances set, with intake valves at .006" and exhausts at .016." These clearances refused to hold. By the end of the 1300-mile trip, several of the adjustment screws had been tightened to the end of thread travel and the valves still clattered.

I had to leave the truck in Veracruz without investigating the cause of the problem. Perhaps it was nothing more than loose rocker-shaft holddown bolts, which are secured with flat (not lock) washers and may not have been torqued properly. Does the torque spec include an allowance for valve-spring tension? Or must the valve-adjustment screws be backed out before the rocker shaft is tightened?

It's also quite possible that the cam and lifters failed, although the engine continued to run well and seemed to develop full power when the proper lash was restored.

Ignition

The distributor was disassembled and a somewhat straighter shaft installed, together with new points, condenser, rotor and cap. A Mallory ignition and high-output coil had been incorporated into the system some years before. All electrical connections were cleaned and lubricated with dialectic silicone grease.

When warm, the engine would come to life almost instantly and, except for a couple of hickups during heavy fog, the ignition functioned flawlessly. However, at the end of the day, examination of the cap showed extreme erosion under the terminals. The plastic had been burned away. I attribute that to coil, which generates enough potential to spark early, before the rotor tip comes adjacent to the metal terminals. Ford flathead owners report the same problem with high-voltage coils.

Fuel System

The mechanical fuel pump was rebuilt with a new diaphragm. And, because the parts were on the shelf, an electric pump and AC fuel regulator were mounted and the necessary plumbing, including a return line to the tank, fabricated. If the mechanical pump failed, the electric pump could be put into service in minutes.

A week or more was spent on the carburetor, trying to tune the "universal" Rochester BC for the application. What an arcane instrument! Eventually the circuitry became explicable and various little tweaks were made to smooth response. With some trepidation, the #57 main jet was opened an rch with a numbered drill bit. The resulting mixture was slightly rich, but still gave more than 16 mpg at engine speeds in excess of 4000 rpm.

The gas tank was sent out for chemical cleaning and the mother of all filters, intended for large diesel engines, was installed on the firewall. Ironically the filter did not prevent a speck of dirt from getting between the inlet needle and seat somewhere south of Tampico.

In conclusion, a 235 can take you anywhere you want to go, although it would be wise to carry a few tools. The example discussed here had belonged to my father, who passed on almost 30 years ago. He would be pleased to know that the old girl survives and has found a new home.

 

 

Home