Build First Vehicle to work on - FJ40

This site may earn a commission from merchant affiliate
links, including eBay, Amazon, Skimlinks, and others.

It's probably normal with no weight on the springs. I wouldn't do a anything with the suspension or steering geometry until the vehicle finished. You're likely going to add 3000+ pounds to the suspension once it's finished.
 
Good news! I have power steering. Bad news i didnt know :D Haha while dismantling that thing i had a hard time to understand why the components where not what i expected. I cant find any documentation on how to open and replace the top seal of the worm shaft (not sure its the correct terminology)... It seems i need a special tool, but everything is crusty and rusty so i dont want to destroy it. Also im not sure how the situation is about parts...

Then on the other shaft some rust came in and that shaft with the gears looks a bit pitted.

What do you guys think? Its okay to reuse?

I guess the seal there wont be sealing too good. And i hit that bearing out of the housing where that seal sits. It was rusty. I realized there is that c clip and below the metal spacer and then a plastic spacer. So now i dont understand how i can replace that bearing and now the depth which i need to bring it in? Any advice or documentation would be very much appreciated. Its seems that my fj40 from Europe is not produced often with this configuration. All manuals show different components...

Besides this things are going their way. Im currently dismantling the driveshafts to get the u-joints out they sound crunchy and then get it all painted and reassembled. Linkage has been dismantled, steering link etc. Trying to get everything ready to approach the goal of getting it back onto its feet!

photo_7_2026-01-05_14-42-40.webp


photo_4_2026-01-05_14-42-40.webp


photo_3_2026-01-05_14-42-40.webp
 
I'm working with AI a lot these days. Because it takes ih8mud as a source almost all the time.

Wanted to share his here because I was searching for that info since I dismantled the box. Worried for nothing...

I sent in the image and asked why there is this hole.

Answer:
Based on the visual evidence of the casting and the specific location on the housing, that hole is a manufacturing sealing plug (often called a "ball plug" or "gallery plug").
It serves no maintenance or adjustment purpose for you. Here is the detailed breakdown of what it is and why it exists:

1. What is it?
This is a permanent seal for a hydraulic fluid passage. When this steering box was manufactured, the internal channels that carry the power steering fluid (the high-pressure or return lines inside the metal) had to be drilled in straight lines from the outside of the casting.
Since a drill bit cannot drill around corners, the manufacturer drills straight through the casing to create the internal gallery, and then presses a steel ball or cup into the entry hole to permanently seal it.

2. What should you do with it?
Nothing.
Do not try to tighten it: It is not a screw.
Do not try to remove it: It is pressed in under high force. Removing it will cause a massive power steering fluid leak that is difficult to reseal without specialized tools.
Do not use it as a drain: It is not designed to be opened.

3. Context for your FJ40
Stock FJ40s rarely came with integral power steering (depending on the year and market). The box in your photo appears to be a Toyota 60-Series (FJ60/FJ62) or potentially an 80-Series steering box.
These are the most common boxes used for power steering conversions on FJ40s because they are robust and mount relatively easily to the frame rail.

The "rib" running down the center of the housing (where the plug is located) is a characteristic casting mark of these later-model Toyota Aisin steering boxes.
Summary: It is a plug from the factory process used to create the internal hydraulic veins. It is safely ignored unless it is actively leaking fluid (which is very rare).



Crazy right? I mean I hope it's true but it's so quick and easy.

P_20260106_162824_1.webp
 
I also use these AI searches, and they can be very informative, but...
The answers always sound very credible, but quite often are completely inaccurate or just plain wrong.
Use with extreme caution.

When an idiot tells you something stupid, he doesn't usually sound credible.
I am in tech and holy hell the upper mgmt is so horny for AI. It is helpful, yes. It can do some tasks, yes. Buuuuuuuuut it's not AI it's an LLM - an extremely fast guessing machine using the data that already exists and it has mined and trained on.

It is replacing a lot of data entry, system analyst roles, and some PMs, but replacing engineers and generating revenue is still a pipe dream - but you wouldn't think that if you just listened to execs and the media. And using it to replace junior engineers is just so stupid and short sighted but that's another argument.

You can't trust AI summaries. They are often...almost correct. It's like back before the internet when you could bull**** your answer if you didn't know. Now it's quickly google, get AI summary, screenshot, and then show - see told you so!

You gotta dig into the sources still.
 
Sounds a bit salty to me. I own a tech company. We significantly increased the output. We have eliminated the job profile of UX, UI and Front-end Engineers. We don't need them anymore. Our juniors do backend tasks like our seniors and we mainly let the seniors check all code commits. The quality has improved, our product has improved and our profit. Saying this is exec talk, it's "just" an LLM and real usage is far away is complete denial.

Other examples, we produce video material for marketing, we do staged photoshoots, we have marketing strategies based on user data which were not there before. All this without being professionals in this particular field.

We already see that mid size companies shift their focus on building again. All in all this is a new paradigm.

But if course it doesn't hurt to be cautious about the answers and check the sources. Whenever I see ih8mud as source it's a match for me.
 
In my experience. AI works on the old adage that if you say something with enough conviction, you can convince anyone it's true.

I'm an expert in very few areas. But for those areas in which I am an expert I've found AI to be generally inaccurate. This leads me to believe that the many areas in which I am not an expert are not topics on which I should rely on AI.
 
An example:

1768215104886.webp


This is what we see everywhere. We use tailwind in all our web-applications. This is the reality. Being cautious is fine of course but downplaying it is not a smart move. The impact is here.
 
But now to nicer things. Another paint day was happening this weekend. Of course i forgot the drivetrain... So im currently waiting on Shims, Clamps, Seals, Bearings and Pinion Nuts. Started the Diff, needed Parts, Knuckles, needed Parts, Power Steering Box, needed Parts, Tie Rods, needed Parts, Supsension, needed parts... Next i think i will open the Transmission. But i want to read a bit about it first. I have 0 Clue what to look for. The same as with the 2F... But i would like to open it some time. Just to see if its crusty inside...

photo_4_2026-01-12_11-55-23.webp


photo_3_2026-01-12_11-55-23.webp


photo_2_2026-01-12_11-55-23.webp


photo_1_2026-01-12_11-55-23.webp
 
In my experience. AI works on the old adage that if you say something with enough conviction, you can convince anyone it's true.

I'm an expert in very few areas. But for those areas in which I am an expert I've found AI to be generally inaccurate. This leads me to believe that the many areas in which I am not an expert are not topics on which I should rely on AI.
I asked AI some questions about the terminal velocity of a squirrel in mph - don't ask.
The answer looked credible and the working looked fine, but the final answer was actually wrong.
The calculation had been done correctly, but then somehow it was forgotten, and instead presented in m/s.

The devil is always in the detail in engineering - one tiny mistake can hide for decades in code, and then will bring the house crashing down when least expected.
 
Hi again

So i was inspecting the Transmission and Transfer Case and i wanted to ask if any of you guys could give me some tips hints on what i have to expect or check.

I plan on cleaning the unit but before wanted to check leaks and im not quite sure what is supposed to be oily and what not... Attached are some pictures aaand then... some AI notes. I wonder what you guys have to say!

----

AI...

Based on the images you provided, this is indeed a Toyota H42 4-speed transmission mated to a "split-case" transfer case. The condition of the unit in the photos points to several specific leak points and issues common to this setup.

Here is a breakdown of the issues visible in your photos and what they likely indicate:

1. Input Shaft Seal Failure (Critical)​

Visible in Image 1:The most concerning issue is the heavy accumulation of black, wet sludge inside the bell housing area (the front face of the transmission) and around the input shaft bearing retainer (the tube the shaft sticks out of).
  • The Issue: This indicates the Input Shaft Seal has failed.
  • Why it matters: In your photo, the sludge is thick. When this seal leaks, gear oil travels along the input shaft and gets flung onto your clutch friction disc. Oil-contaminated clutch discs cannot be cleaned; they will slip and chatter, requiring replacement.
  • The Fix: You must replace this seal (Toyota Part #90311-38028) and inspect the surface of the shaft where the seal rides. If there is a deep groove worn into the metal, a new seal will not fix it—you may need a "Speedi-Sleeve" to create a smooth surface.

2. Rear Output Seal Leak​

Visible in Image 2:There is significant "wet" grime caked around the rear output flange (where the driveshaft connects).
  • The Issue: The Rear Output Seal (transfer case output) has hardened and failed.
  • The Fix: When you remove the staked nut to take off the flange, check the sealing surface on the flange itself. These often develop grooves from years of rotation. If the flange is grooved, polishing it or sleeving it is necessary before installing the new seal (Part #90311-38047).

3. The "Fluid Migration" Risk (Internal Leak)​

Visible in Images 3 & 4 (The connection point):You can see the transmission (cast iron/black) mated to the transfer case adapter (aluminum/silver).
  • The Issue: A very common failure in these H42 units is the Intermediate Shaft Seal located internally between the two cases.
  • Symptoms: Since you haven't opened it yet, watch for this: if the transmission was low on oil but the transfer case is overfilled (or gushes fluid when you pull the fill plug), the internal seal has failed. The transmission pumps its oil into the transfer case.
  • The Fix: You must separate the transfer case from the transmission to replace the seal located on the back of the transmission output shaft.

4. Shift Tower & Top Cover Leaks​

Visible in Image 3 & 4:The top of the transmission (shift tower) shows layers of dirt mixed with oil.
  • The Issue: This is likely a combination of a failed Shift Lever Boot and the Top Cover Gasket.
  • The Breather: Look closely at the small breather cap on top. If this is clogged with mud or grease (which is common on off-road trucks), pressure builds up inside the case as it gets hot and forces oil out of the weakest seals—usually the ones mentioned above. Cleaning or replacing the breather is mandatory to stop future leaks.

Summary Checklist for Your Teardown​

  1. Input Shaft: Check for a groove worn by the old seal.
  2. Output Flange: Check for a groove worn by the rear seal.
  3. Intermediate Shaft: Plan to separate the cases to replace the internal seal (prevent fluid migration).
  4. Breather: Unscrew the breather cap and ensure it flows air freely.

photo_4_2026-01-19_15-08-36.webp


photo_3_2026-01-19_15-08-36.webp


photo_2_2026-01-19_15-08-36.webp


photo_1_2026-01-19_15-08-36.webp
 
Waiting on parts... And service manuals...

So I thought why not give welding a go. First patch. And I'm very happy how it turned out. I banged on that sheet and that bent was similar to the bent on the bottom of that bib. A good start to my patch career.

I also learned that we have a subscription model for gas bottles in Switzerland :/ so I wanted to get a full bottle but no chance to buy with gas inside... What a world.

1000056548.webp
1000056500.webp
1000056333.webp
 
Just a small update. First diff rebuild. Backlash fine and the shape of the contact is an oval in the center. So I'm confident this is an improvement overall. I was a bit worried that the pinion somehow has a sort of backlash too which seems more than the backlash of the carrier. But after checking around 30 times with the same result I decided to continue the assembly.

One more to go. Hopefully tomorrow. Then I can start to put in the axles.

A question on the knuckle. When I screw on the felt the resistance to turn is way bigger than without. I measure the load on the drag without the felt?

1000058422.webp


P_20260213_132959.webp
 
Just a small update. First diff rebuild. Backlash fine and the shape of the contact is an oval in the center. So I'm confident this is an improvement overall. I was a bit worried that the pinion somehow has a sort of backlash too which seems more than the backlash of the carrier. But after checking around 30 times with the same result I decided to continue the assembly.

One more to go. Hopefully tomorrow. Then I can start to put in the axles.

A question on the knuckle. When I screw on the felt the resistance to turn is way bigger than without. I measure the load on the drag without the felt?

View attachment 4085577

View attachment 4085582
You stated, "I was a bit worried that the pinion somehow has a sort of backlash too which seems more than the backlash of the carrier." Can you explain why you feel the pinion has more backlash than the carrier? Is it backlash or simply free play because your final assembly used the same crush sleeve used for setting the gears?

Assuming your gear set isn't OEM the pinion doesn't have a shelf that the bearing would sit against which is why a crush sleeve is included. During setup you normally have to disassemble several times to get the correct pinion shims correct for the best pattern. That crush sleeve should not be used during final assembly because during the setup it was probably crushed beyond where is needed for final assembly which is why better kits will come with 2 crush sleeves, one for setup and one for final assembly. If it is not long enough during final assembly there will be play in the pinion bearing which you may be thinking is extra backlash.
 
Don't assemble the knuckle seals, wipers and such when centering, and setting preload on the knuckle trunions. I've only setup a crush sleeve diff a couple of times. I usually set pinion depth and get a good pattern 1st. Then i set pinion preload with a solid spacer or crush sleeve. Then install the carrier, backlash, and double check the pattern.
 
You stated, "I was a bit worried that the pinion somehow has a sort of backlash too which seems more than the backlash of the carrier." Can you explain why you feel the pinion has more backlash than the carrier? Is it backlash or simply free play because your final assembly used the same crush sleeve used for setting the gears?

Assuming your gear set isn't OEM the pinion doesn't have a shelf that the bearing would sit against which is why a crush sleeve is included. During setup you normally have to disassemble several times to get the correct pinion shims correct for the best pattern. That crush sleeve should not be used during final assembly because during the setup it was probably crushed beyond where is needed for final assembly which is why better kits will come with 2 crush sleeves, one for setup and one for final assembly. If it is not long enough during final assembly there will be play in the pinion bearing which you may be thinking is extra backlash.
There was no crush sleeve. It's OEM I have a bushing and a shim. There is no vertical backlash, by vertical I mean along the pinion shaft. So I can not push it up and down it's snug and tight. I assembled it like so, putting shim and bushing in snug it up torqued it like a maniac and measured preload till I got manual spec. It's snug and no wobble or such.

But when I hold the pinion flange and rotate it there is a gap or freeplay between the carrier and the pinion gear. So if I rock this back and forth it feels like more play than when I rock back and forth the carrier. But I'm not exactly sure how it's supposed to be I have no assembled one anymore.

I hope you got what I mean.
 
Last edited:
Boomer Alert: Ignore the Finger in the Photos

A tiny update. Rear third also redone and installed. It looks so new. Unbelievable. I just stare at it and feel joy. I went for sealant because there are some groves from the rust... Sealant - Gasket - Sealant. 28nM torque and when it's all dried I torque again to 28nm. After the first drive I go to 30nm I think. At least that's my plan.

I read that in old systems modern gear oils would leak since they have a different viscosity. Is this the case? Any experience for this? I have 5L of this: Motorex PENTA LS SAE 75W/140.

Cheers

P_20260215_153812.webp


P_20260215_153819.webp
 
The popular or common argument is that modern synthetic’s tend to leak more possibly because of not swelling seals as well as conventional products.. and possibly viscosity depending on which synthetic . I think I have observed this at times on older engines, generally have not run in old gear boxes/axles, probably as much because I’m cheap.
 
Back
Top Bottom