The Future Holds:

The Details

I had originally planned on using an NP435 wide-ratio Ford truck 4-speed transmission, unless I could get a great deal on a heavy-duty, wide-ratio five-speed. Since I'm planning on using an engine that came paired with the NP435 transmission from the factory, it would allow me to use all stock parts to put it together. (I personally find this concept very attractive.)

I liked the NV4500 transmission, but I'm not into paying those kinds of bucks for a transmission, and then still have to purchase another $500 or so worth of adapters to mate it to the Ford engine. I also looked for a five-speed New Process box (the NP 540, 541, or 542), which is a version of the NP435 with an overdrive fifth gear added. They're kinda rare, so I really didn't get my hopes up of finding one of those. One other possibility is the Ford/ZF-5 truck 5-speed. This is the transmission that I have in my F-250. This transmission, which has an integral bellhousing, would bolt right up to my Ford small-block engine, as this is the engine that the transmission was designed for. Also, the ZF-5 has a set of gear ratios similar to the newer version of the NV4500. (The newer NV4500's ratios are 5.61, 3.04, 1.67, 1.00, and 0.75, while the ZF-5's ratios are 5.72, 2.94, 1.61, 1.00, and 0.76.) However, a used ZF-5 is $1100, plus a $200 core charge.

Before I go on here about the transmission let me spout some more of my personal opinion, this time on manual transmissions.

I've had the Ford toploader 3-speed in my first two cars (1960's Mustangs), and the 5-speed Borg Warner in a 1990 Mustang GT. The 3-speed had a pretty wide gear spread, which was just fine for street driving behind a 200 CID six-cylinder. For street driving in the GT, I found that the mileage went up a few points if I only used the odd gears. Since I'm not a drag racer, wide gears are just fine for me.

With this Jeep project, I've tried to balance out practicality and reality against being idealistic and just plain doing fun stuff. Where practicality comes in as far as the transmission is concerned, deals, of course, with gear ratios.

Okay, so I am from the Pacific Northwest, where rock crawling gears rule. However, I also lived in the Dallas/Fort Worth "Metroplex" for four-and-a-half years. What this means is that I've also found the necessity for an overdrive, since in Texas, it's always a long highway drive from one metropolitan area to the next. If I want lower gears, well, there's additional gear boxes available for that (see further down on this page).

So, after a lot of thinking and discussing with my dad and some friends, I picked up a Ford TOD transmission (and got a great price on it, to boot). This unit is of the classic Ford toploader design, with wide gear ratios and an overdrive 4th gear. (Think of this as being similar to an NV4500 without the granny gear first.) Not only has this transmission seen a very recent rebuild, the Ford toploaders are the only transmission that I've personally taken apart and put back together, so I've got a familiarity with the workings of it.

A disassembled Ford TOD 4-speed.

* Note: I used to think overdrives were some gimmicky thing for imported econoboxes. Then, I found that my V-8 Thunderbird with an overdrive transmission can get low- to mid-30's mileage on long freeway trips. Overdrives definitely have their place.

After a lot of thought and a good deal, a case iron Ford 3.03 3-speed appeared in my already overfilled garage. This transmission is very similar to the 4-speed TOD. The case is a bit shorter, but the general shape is very similar. The tail housings are 100% interchangeable; even the rear bearings are the same. (The front bearing is larger on the 4-speed, so the front housings aren't quite interchangeable.) As a matter of fact, even though the 3-speed is from a 2wd something-or-other (it was on a pallet of various manual transmissions at the junkyard, so I have no idea what it came out of), the splining and shaft length is just right for the adapter housing that came with the Dana 20 transfer case.

Ford 3.03 in a home-built assembly fixture.

One thing about this 3-speed is that even though it's a top-loader, the shift forks are controlled by two levers on the left side of the case. Because of this, it's very easy to make up a shift linkage that puts the shifter as far forward or backward as I want it to be.

The next trick is putting the two transmissions together. With the way that the cases are shaped, there's no way to do this without a plate in between.

The shafts I need to join.

The narrow part, halfway between my hands, is where the shafts need to be joined.

Now, I can think of several ways to join the workings of the two transmissions:

1. Machine the output shaft of the front transmission and the input shaft of the rear transmission, so that they're shortened and splined, then use a connecting "collar", slid over both shafts.
2. Machine the input shaft of the rear transmission in such a way that it also functions as the output shaft of the front transmission.
3. Machine the gear off of the input shaft of the rear transmission, spline the inside of it, and then machine the output shaft of the front transmission to accept this gear and to act as the input shaft of the rear transmission.
4. Similar to #3, but find a production gear from some other application, instead of taking the one from the rear transmission's input shaft.
5. Shorten the output shaft of the front transmission, hollow it out, and spline the inside. Spline the input shaft of the second transmission to fit inside.
6. Like #1, except get some bearings with a bigger hole, so that the collar fits inside the bearings.

All are valid possibilities.

#1 is a very common way that guys join two transmissions together, but to me, this seems kinda weak (lots of failure points), and would require more adapter thickness than the other three methods to get enough spline length on both shafts.

In my particular situation, we could rule out #2 immediately; the input shaft of the second transmission is about 3" too short, and that's measuring with the cases flush together (no adapter plate).

Because of the very large hole inside the gear/synchro part of the input shaft, #3 and #4 won't work.

#5 is looking pretty good, but I'm trying to find someone to talk to about if this would be strong enough, and how the heck internal splines get cut (if they can even be cut in this scenario), and how you case harden internal splines.

I'm starting to like #6, as it has the advantages of #1 without the extra adapter length, but where in the world do I find the bearings and the collar?

Why not just weld together the output shaft from the front transmission to the input shaft of the rear one, you ask? Because the bearings won't fit over the fat ends of the resulting shaft.

After a lot of looking at the two stripped down transmissions, and getting input from my dad, joining these two transmissions just is not looking very good. The problem is that even though the output from the front transmission (it could be either one) would be easy to machine for a sleeve, and an adapter block/plate to go between the two would be very easy to make myself, the input shaft of either transmission is too thin to do much with. The only area of the input shafts that comes even close to being thick enough to spline for a sleeve is the area where the bearing sits. However, due to a couple of grooves cut into that surface, plus the fact that it's not even an inch long (let alone that using that area for anything would mean not using a front bearing), make that unusable, too.

Then, on a trip down to the beach, I stopped at a junkyard and picked up an NP435 transmission. Quite literally. It was sitting below a 1984 F-250 2wd truck. The yard had pulled the engine, the bellhousing, the rear axle, and the driveshaft, but had just let the transmission drop. All I had to do was pull it out from under the truck to check it out and pay for it. A guy at the yard carried it out and loaded it while I was paying for it.

* Note: Something interesting I found about the junkyards down at the beach; greasy parts are cheap, but forget about getting any body parts, unless you want to buy rust held together with a bit of paint. One yard I visited had stuff that was so rusty that I couldn't tell if I was looking at transmissions or small engine blocks, as well as a pile of big-rig axles that the housings had rusted thru.

After months and months of gearbox talk and before I purchased any transmissions, my thought processes went on what else I could do that would fit in a CJ-7 and yet be totally off-the-wall. Even though it looks like I'm going a different route, I still thought I'd share some of my ideas here.

The Dana 20 transfer case that I have had been bolted to a Ford 3.03 3-speed transmission. This is a very strong, yet short, transmission. As I mentioned before, this transmission was in both my first and second cars (old Mustangs), and thus I'm real familiar with it. This transmission has the shift levers on the outside, and a case length of only 9-1/4". I had originally given some serious thought to putting this transmission into my drivetrain along with the NP435 that I was planning to use. The adapters would have be similar to what Chris came up with, and the external lever-and-rod shift mechanism is pretty easy to extend/shorten to put the shifter wherever I want it. Sure, I could use a second 4-speed, but those are more expensive (who ever wants a 3-speed?) and physically longer.

Let me spout some of my theory here. Every situation that I've seen that's had an extra gear box involved, whether it be a transmission, transfer case, or other custom gear box, has always put that extra box after the transmission. I can see a good reason for this; there's less frictional rotating mass that the engine has to change the speed of when you shift the regular transmission. As the driver of the vehicle, you'd feel this in the engine taking extra time to change speed (accellerate) when the clutch pedal is pushed in. So, even though in some respects, it would be really nice to put the 3-speed up front (it's easy to place it's shifter, and that arrangement would move the top-mounted shifter of something like the NP435 out from under the dash), that's probably not the way to go.

After all this thinking of using the 9-1/4" long 3-speed, Chris reminded me that it's only 1-3/4" shorter than an NP435. Hmm... The original plan, the NP435, Dana 20, and 4.10 axle gears would give me a crawl ratio of about 68:1. Adding the 3-speed into the mix makes the crawl ratio about 236:1. A second NP435 instead of the 3-speed takes the crawl ratio all the way to about 454:1. This is wayyyyy out there. If I was building a full-sized truck (read: lots of wheelbase), this would be a viable idea. However, I'm working with a CJ-7. There's just not much wheelbase to work with and I've got to keep the overall length of the gearboxes real short if I want to have any rear driveshaft length left. That, and I personally feel that 454:1 is overkill on the gear reduction. Also, something to remember is that the more gear reduction you get, the more stress (torque multiplication) your rig will be capable of putting on the driveshafts, U-joints, and axles.

Here's another idea I worked thru. If between all the gearing, I can get a slow enough crawl ratio, I could potentially use faster (lower numerically) gears in the differentials. This would give me an effect on the highway similar to an overdrive. The mileage in a CJ will always stink, so any gains won't be too big. As Rodney (Jeepskate) has so aptly phrased it, Jeep CJ's "have the aerodynamics of a cinder block". Still, a little extra mileage and less engine wear still seem to me to be worth the thought.

I don't have to have an extra gear box in my Jeep. It's just too fun of an idea to not think about.

With my front axle, I needed a transfer case with a left-hand drop front output and a centered rear output. An Atlas II would've been great, but once again, I'm not into spending those kinds of bucks. Besides, the Atlas (okay, I know they're tough) uses non-OEM pieces, which makes parts for 'em kinda hard to find when Advanced Adapters is closed for the evening. I found a Ford Dana 20, from an early Bronco of uknown exact vintage that I'm going to use. (Based upon what I've been able to find about the shifter, it's a very early 1966.) Another possibility that I looked at was a Ford-version NP205 transfer case. The downside to the NP205 is that the low range isn't as low as the Dana 20 (1.98:1 as compared to the Dana's 2.46:1.) I know the Dana 20 is strong. The NP205 came in one-ton trucks, so it would be very bulletproof in a CJ-7. Also, both transfer cases came stock behind the NP435, so once again, I could use factory parts to put it all together. (In all reality, I was looking for either.) I just happened to have come across the Dana 20 fairly accidentally, and it was in way too nice of shape and way too great of a price to pass by. Since this transfer case looks like it was recently rebuilt (great looking seals, no slop in the bearings, no metal particles floating around), I did a quick clean-up of the outside of the case with a wire brush, and painted it with some light Ford blue engine paint I had left over from the Bronco. The only real snag I've run into with this transfer case is trying to figure out how to mount a twin shifter, since the current shifter is not of the 1967-and-up variety.

My friend, Chris Waterman, is using the NP435 transmission and Ford Dana 20 transfer case that I was originally planning on using. He's added one more gearbox into the drivetrain, however. He's taken the gear reduction unit off of a Dodge NP203 and has dropped it in between the transmission and the transfer case. He mounted the reduction unit to the back of the transmission with a custom steel adapter, and to the front of the transfer case with a custom aluminum adapter. I am now going this route, too. I'll go with steel instead of aluminum for all of the adapters, since I'm notorious for hosing up aluminum.

* Note: Aluminum and I are a bad match. I have nothing against aluminum alloys; I love the light weight and the corrosion resistant properties. Aluminum makes a nice pop can. I just have problems working with it. In my university's metal shop, they had "the Norwegian test" to determine metal types. The instructor would take a piece of the metal in question, and give it to this Norwegian (me). He'd then ask me to put a threaded hole in it. If I royally screwed it up on the first try, then the metal was aluminum. If everything worked out just fine, the metal was something other than aluminum. (Mind you, it's only aluminum I have problems with. Even tricky metals, like stainless steel, cast iron, zinc, and magnesium, weren't a problem for me.)

On another trip to the beach (okay, my grandparents live down there, so I get double-duty out of my trips down there), I stopped in the same yard I got the transmission from. It was a nasty, cold, windy, rainy day. As I walked out thru the yard, I found this nice NP-203 transfer case with a long rear driveshaft just sitting there in the aisle on an old rim to keep it out of the mud. I asked about it, got a great price, and was told to go get a guy working in the yard to haul it out to my truck. Well, between the facts that an NP-203 is so horrendously heavy, the weather was so lousy, and I was driving a four-wheel-drive truck, the guy asked if I would mind driving back thru the mud and water to pick it up. That was actually kind of fun. Together, we picked up the transfer case and slid it into my truck. I hopped back in the cab and then drove on out of there (while Shelby laid down in front of one of the heat vents in the truck and showed no desire to get out).

I'm liking this yard.

I made a second trip to the junkyard to retrieve the driveshafts from the same early Bronco that I got the transfer case from. The rear is a CV-joint type, and while the front has standard single-Cardan cross joints at both ends, it has the novelty of still having the original aluminum Dana Corporation sticker attached. Unless some unforseen problem crops up, I plan on using these shafts as my starting point.

I got a kind of freebie bonus in the way of driveshafts. When I got the NP-203 transfer case, the junkyard guys didn't want to go to the effort of removing the nice, long rear driveshaft that was still attached, so they just gave it to me. This driveshaft has a slip yoke in it that's in great condition, and the shaft has nice, big yokes on it that will accept a big U-joint. It's also really long, so it can be cut down to work for the Jeep.

This is getting pretty involved.

At some point, Ford started using hydralic and cable-actuated clutches. When this happened, they quit putting a pivot point on the rear of the left side of the engine block. This pivot was used for one end of the "equalizer" bar of the mechanical clutch linkage (the other pivot point is mounted on the frame). On the newer engine blocks (like what I have), there isn't even a boss to drill out to put the block side pivot point in. However, there are a few aftermarket devices that consist of a bracket that bolts onto using two of the bellhousing bolts that takes the place of this on-block pivot.

(for those of you that aren't familiar with the mechanical linkage/equalizer bar, it's a tube that has a bearing in each end that slides over the pivot points and sits cross-ways near the rear of the engine. Welded to this tube are two bars; one that goes up, and one that goes down. The upper bar has a hole in the top end, and a rod runs from there thru a hole in the firewall to the clutch pedal. The lower bar has a hole in the end of it that has a rod that runs from there back to the clutch release lever. The whole equalizer acts like a see-saw, changing the forward push from the pedal into a backwards push to the clutch lever sticking out of the bellhousing.)

My concerns with using a mechanical linkage are fourfold:

So, this leaves me with two other possibilties:

There aren't many cable actuated clutches out there. The Mustang has one, but adapting it to the Jeep would be pretty hard, if not impossible, due to the fact that the Mustang manual transmission bellhousing is radically different than anything that will work with a non-T5 transmission. (Chris Waterman went thru this mental exercise.) Also, the Jeep's flexing comes to mind; I can see it tugging on the clutch cable when the vehicle flexes.

A hydralic clutch solves the above problems, but brings one new one into the works. It places a hydralic master cylinder up high, just to the outside of the brake booster. On some vehicles, there's some interference between the two.

To me, the hydralic clutch is still the best bet. One day, while I was waiting at the junkyard for the forklift to pull down an axle from its location twelve feet up on a rack, I spied a bellhousing that was on the top of a nice, neat stack. It had the Ford hydralic cylinder bracket mount on it, and the part number E4TZ-7505C, and even had the rubber boot and throwout lever (parts E4TZ-7513A and E4TZ-7515C respectively).

The Rest of the Jeep's Gearboxes and Driveshafts

We'll just have to see what happens. I'll update these Jeep pages as I figure stuff out/get stuff done.

(To go back a page, click here.)