1943 Willys MB Rebuild
The MB doing its thing
The Willys MB project was a feat in that it was completed in about 2 months. Having been in my friend’s family for many years, they were looking to do a restoration/upgrade on it.
It had undergone so many repairs and body modifications over the years. That fact, combined with the cost and relative impracticality of a full authentic restoration led to the decision to rebuild and modify. If I were in charge of the finances, I would have at least upgraded the powertrain/drivetrain to that of a YJ, at the minimum. A fuel injected 5-speed MB would be awesome. However, I wasn’t in charge of the finances or major decisions.
When we started on the MB, it was in terrible shape. Huge rust holes, many pounds of oil/dirt/cow crap on everything, and a terrible engine knock. A complete engine/body rebuild was necessary without question.
This is a relatively brief summary of the project. In the interest of time and space, I have left out a lot of the boring details. As I have more free time, I may add to it. I have so many pictures that I took during the job, that I have a hard time picking them. The XJ page was a lot easier to put together because it had specific sections.
I learned a lot of the ins and outs of the MB/GPW during the process, so feel free to email me if you have any questions. If I can’t answer them, I may be able to refer you to someone who can.
Jump to: Engine Drivetrain Body
Engine
For those unfamiliar with the “Go Devil,” here’s a little bit of general background: the Willys MB and nearly identical Ford GPW were built specifically for use in WWII. Reliability and simplicity were obviously important concerns, and in the case of the MB/GPW, it needed to fit under an obstacle about 4 feet off the ground (with the windshield folded down). The L-head, which had proven itself in the Willys Americar, had more horsepower, torque, and greater displacement than the other vehicles competing for the government contract. The L-head designation signifies the flathead construction of the engine, with both the intake and exhaust valves located in the block. This construction is still fairly common in some single-cylinder yard equipment engines. Compared to a modern OHV or OHC engine, the L-head is very inefficient in terms of airflow. By the mid 1960’s, the L-head design was gone from nearly all automobile production.
As is the case with many early bypass-filtered engines, the filter had been…well, bypassed on this engine, and was missing entirely. The Go Devil didn’t have a specific path for the oil filter between the oil pump and the engine. Rather, the oil that couldn’t squeeze through the oil passages was bypassed through a filter. This method only filters about 10% of the oil, and attempts to increase the filter flow will cause the #1 rod and main bearings to be starved for oil since the oil filter feed is located next to the #1 oil gallery.
The engine had a 2A serial number, and had been rebuilt in the past. Things were much worse than was expected, and the severity of the problems with the engine weren’t evident until it was fully disassembled. There were large ridges on the already 0.040 over cylinders, the crank was deeply scored and pitted, the cam was scored, and there were cracks in the top of the block on both sides of one of the stud holes, both leading into coolant passages.
The nasty block crack
Given the condition of the internals and the cost of the block repair, we decided to find a new L-head. We found a Jeep junkyard about 2 hours away from where we were in Los Angeles that said they had a “guaranteed rebuildable” L-head that they would sell for about $230 without the head. After a walk around the shelves of hundreds of old Jeep engines, we get to the one that was supposed to be it, and it was an F-head, not an L-head. We finally found an L-head that was supposedly in decent condition. The guy at the junkyard insisted that this wasn’t the engine they indented to give us at the quoted price and reminded us repeatedly this was a “fu**’in good engine.” He pulled the dipstick and showed us some oil that was black as coal, apparently to prove how good of an engine this was?? After 30 minutes of hammering to get the head off, the engine was ours. They picked it up with a forklift by one of the head studs (ouch!).
Upon disassembly, I found that the engine was a timing gear model instead of a chain like our old one. It was also already bored 0.060 over. Fortunately the crank and cam were in significantly better shape than the old engine. When it came time to remove the 15 head studs, I made a point to be extremely careful so as to not break any studs. I had used the double-nut approach on the first engine, and broke one or two studs. I began a long ritual of wire brushing, PB blaster, tapping with a hammer, heating, etc. I continued this process for a couple of days. This time, I used a stud remover to prevent as much inappropriate loading on the stud as possible. When the dust settled, there were three snapped head studs and two exhaust studs (which were like that when we got the engine). Anyone who has done this kind of work knows the sickening feeling of a bolt letting go (shudder).
Since EZ-outs were out of the question, I found an EDM shop to remove all five studs. After a lengthy 10+ day jerking around and $100 later, we finally had the block ready for machining.
The machine work was done by Valley Head Service. The engine was bored 0.080 over, the mains were machined 0.010 under, new valves, new valve guides, hardened valve seats, new oil pump, new timing gears, etc. When all was said and done, the machine shop bill was around $1000, about double what we had originally been quoted. The quality of the machine job was definitely overkill for the type of engine and what it was going to be used for. However, it’s not easy to find a machine shop that will deal with an engine like this one. At least in the greater LA area.
I rebuilt the Carter W-O that’s in the picture below, but it was eventually replaced with a new Solex carburetor that starts easier, runs smoother, and doesn’t leak.
The engine compartment after and before
We finally got around to installing the oil filter in the MB. This job was greatly delayed by part sourcing problems and life in general. Yes, it has a Fram PH-16 on it, and no, I wasn’t the one who bought that terrible filter.
I knew that PermaCool sold an oil filter relocation kit, so I figured that we should be able to get just the filter manifold. It was less than $20, but it wasn’t extremely easy to find a parts dealer, let alone one that could actually get it.
The manifold was set up for 1/2" NPT fittings, and 1/2" to 1/8” NPT adapters with 1/8” NPT to 1/4" inverted flare brake line adapters got to the line type i wanted. The large 1/2-1/8 adapters made a great place to install the necessary restrictor. The restrictor size was the size of a #52 drill. I made the restrictor by turning a 0.001”-over press fit cylinder to go inside if the reamed-out adapter. A #52 hole finished it off. I have no doubts that the press fit will easily stand up to the ~50 psi of oil pressure, and even if it didn’t, it would run into the smaller adapter before it went anywhere.
The bypass filter, it's better than nothing
Back To Top
Drivetrain
Ideally the entire drivetrain would have been scrapped. We were going to be running a set of 31x10.5 mud tires and the Dana 25/23-2 combination is far from tough. However, since the engine wasn’t going to be putting out massive amounts of power, and time and money were issues, the axles got cleaned, painted, and received new seals and bushings where applicable. The axles had been through some nasty stuff, and had a ton of caked on junk. The really nice looking stuff came out of the inside of the case though.
Based on the advice of old Jeep experts (and the Jeep Bible, I believe), we filled the knuckles with about 1lb of grease each, and topped off with gear oil. Rather than kill our wrists with the grease gun or find an electric one, I made a large grease syringe to fill the knuckles.
Our front axle looked like the original, and had the early Tracta u-joints in it. The rear end had been replaced with a “script F” GPW axle.
The transmission didn’t need much. A new synchro and a spring washer for the shift gate, and a paint job took care of it.
The transfer case should have been simple too…but it wasn’t. One of the mounting tabs on the front bearing cap had cracked off.
Instead of possibly having to re-machine the surface after welding it back, I set out to find a replacement. It wasn’t an easy part to find, and it came complete with a nice coat of heavy rust on it. After that was taken care of, the transfer case was a simple job. I left the interlock pin out of the levers so that 2-low was available.
Back To Top
Body
The body tub became the last priority of the project. All of the delays with the engine left us very little time to fix the body. Most of the gusseting on the underside of the tub, along with a lot of the old tub floor, had to be removed. This involved far too much time with the cold chisel/small sledge, reciprocating saw, and angle grinder.
Initial mess, and post stripping/hacking.
One of the priorities with the tub was to replace the “hat rail” channels that were under the floor. The original ones still had some of the wood left in them. I found appropriately sized, pre-bent hat rail at Eastwood.com. The underside of the tub consisted of a large amount of prep work (see above), and patching/section replacement. A number of the welds aren’t beautiful, but for my first welding experience, they’re not so bad.
Back To Top