Inspection of a 904 torqueflite

-
48923642651_d9ba6f13a6_m.jpg
-

48923843067_8452294304_m.jpg
-

When pressing that busing in. The lower part of the sun gear is solid and straight. It’s not gonna bend or flex. It worn hurt anything.



48923842987_ec848211ab_m.jpg
by --

If you turn it over. There is nothing supporting that sun gear.....except the thin shell. If I press a bushing into that end of the gear. Without supporting the bottom of the gear with something solid. All that pressure is going to be absorbed by the sun shell. It can bend and warp.

I saw it flex and then return to shape when the pressure came off.

What I should have done....is take the sun gear off and press the bushings in that way.
Or .......just put a large socket under it to support the sun gear
 
glhx said:
I didn’t press the sun gear in. Those went in with snap rings.

What I did was press bushings into the sun gear. Without taking the sun gear off of the shell.

Pretty sure it warped the shell
I’ll post a picture of what I did......and why it warped the shell. I think.

I’m going to get the next set of bushings from a different supplier.
If it doesn’t work. I’ll take it to a trans shop and have them do it. Maybe find out why the reaction shaft bushing is compressing so badly. It’s pretty far down in there so it’s hard to mess with
Click to expand...
If you are going to buy new bushings, I would take the parts and the bushings to a trans shop and ask them to put them in for a reasonable fee. Very easy to get in trouble with a press and inexperience. I have too.
 
I’m thinking about it. I’ve already trashed two of them in the 999. Reaction shaft
The 904 reaction shaft was done at the same time with no issues.

This time I ordered 3 replacements for the reaction shaft. I will try one more time.
Then let a trans shop try and see if they can get it. If they can’t and they see why. I’ll have a third one ready to go.

I was able to get the ring gear support replaced with a new bushing and I think I have a sun shell on the way for the 999.

After all this. I should have all the parts and knowledge to finish both units.

On the sun gears for the 904 and 999..... The bushings in them are both new and rotating like they should....I had to repress the 999 again with another set but the second time they were fine. I did not put pressure on the shell this time. The bushings were pressed with the sun gears removed.

All that’s left is the 999 reaction shaft to finish the bushings.

I’m not sure why that one is compressing the middle. Maybe there is a high spot in the housing that is causing resistance on the way in and making it harder for the bushing to go in smoothly. So it compresses it on the way in. At this point I’m pretty good at getting these pressed in with a press and a hammer manually.
Maybe some dawn dish soap will help......

It’s difficult or impossible to measure that far down the hole to find high spots so I’ll have to look at it more carefully. I wonder if the bushings were originally pressed in and then machined in place. I know it’s common with the seam style bushings at the factory to do that. The original could have had a high spot and just been machined out.
 
Last edited:
glhx said:
I’m thinking about it. I’ve already trashed two of them in the 999. Reaction shaft
The 904 reaction shaft was done at the same time with no issues.

This time I ordered 3 replacements for the reaction shaft. I will try one more time.
Then let a trans shop try and see if they can get it. If they can’t and they see why. I’ll have a third one ready to go.

I was able to get the ring gear support replaced with a new bushing and I think I have a sun shell on the way for the 999.

After all this. I should have all the parts and knowledge to finish both units.

On the sun gears for the 904 and 999..... The bushings in them are both new and rotating like they should....I had to repress the 999 again with another set but the second time they were fine. I did not put pressure on the shell this time. The bushings were pressed with the sun gears removed.

All that’s left is the 999 reaction shaft to finish the bushings.

I’m not sure why that one is compressing the middle. Maybe there is a high spot in the housing that is causing resistance on the way in and making it harder for the bushing to go in smoothly. So it compresses it on the way in. At this point I’m pretty good at getting these pressed in with a press and a hammer manually.
Maybe some dawn dish soap will help......

It’s difficult or impossible to measure that far down the hole to find high spots so I’ll have to look at it more carefully. I wonder if the bushings were originally pressed in and then machined in place. I know it’s common with the seam style bushings at the factory to do that. The original could have had a high spot and just been machined out.
Click to expand...
Well even if you just took the ones in that you are having trouble with to a trans shop. You still have the experience of doing most of them yourself, but you can move your project along.
You have to be careful with a press, big pressures there that can get you in trouble easily. Ask me how I know. I ruined an over drive component when the parts did not go down evenly on the splines.
It shaved metal and I had to buy a new one from wittrans.
 
CudaFactHackJob said:
Education isn't free, but bushings are cheap. If you take ANYTHING to a transmission shop I say you gave up.
Click to expand...
IMHO some jobs where you are probably never going to do it again, it is hard to justify the investment in specialty tools. I always weigh that out between how much it would cost to pay someone to do it versus the cost of the proper tools and how much use I will get out of that tool. I believe the recommendation to bring parts to a trans shop to have the bushing done was in Tom Hands new book.
 
Yeah, that press and what I don’t know has cost quite a bit in replacement parts and shipping. I added it up last night to the tune of $200 in bushings, shipping, and broken parts.

Tom did say that governor play was excessive and told me to replace the output shaft. It was in his book on page 80, about sealing rings cutting into the reaction shaft support housing and clutch retainer due to excessive end play. Apparently they move back and forth so much it acts like a saw over time and creates fine aluminum dust on the back part or cast dust on the front part. I’ve got so much in this I wasn’t leaving anything to chance. The play I had was in the governor housing and heavily wearing the output shaft support. There is no spec on this endplay in the atsg and it would not affect endplay for the whole transmission. Just the governor housing because once it’s locked with the snap ring it’s basically stand alone. The snap ring was checked and correct. A shim could have been put behind the governor but it would have been better to replace the whole shaft.
The ......sealing rings.....output shaft support....and output shaft were replaced.

On the bushing in the reaction shaft
I might try to hammer the next bushing in instead of using the press. Just have that one last bushing to get right. There has to be a reason it’s buckling so bad.

At least they are easy to get out now. I’ve got 4 more sitting here.
This is the third one taken out and the second one I trashed with the same buckling problem.

The first one I thought was a fluke and maybe just bad luck. The second one....it’s going to be a consistent problem so I’ve got to figure out why.

I’ll rebuild a few more of these at different points. The curve was worth it and $200 isn’t much to gain that experience. Plus all the extra servos and shift kits. The frictions and steels in the 999 were good...almost new. I replaced them anyway.
WIT trans has been a great source.. cobra had some hard parts as well. Transpartsusa was ok. Decent prices......slow shipping.
The rest were google searches that landed me on eBay for some of the harder parts. Rat rod al on here had a spare 904 in pieces he sold me a bunch of misc parts from a 67 904. Couldn’t use most of them but in the end. He sent that parking pawl spring and the trans bolts with the filter bolts I needed to finish it up.

After this is done. I just have a rebuilt 904 with with a tfsc kit and no real purpose sitting around. I built it for the experience and it came in very handy for comparison. The 67 parts as well were different on the reverse apply. They have a strut.....my 76 904 doesn’t.

I ended up with the design progression from the
older 904 ......4/3 .......6” shell with reverse strut
Middle 904 .....4/4......6” shell with reaction pin
And the 999 5/4........6 1/8” shell

I think mid 90s had an even wider shell. Like 6 3/8”

So this wasn’t just a learning experience in hard and soft parts and terminology. It had a lot of hands on design stuff as well. I took it slow and worked on it in spare time.

We will just have to see now how it actually functions when it’s done.
 
Last edited:
You said "hammer". I never hammer a bushing to start it in. I GENTLY tap it using a large flat plate. Many many gentle taps to get it started. That way you don't damage it; and you can tap it back out if it cocks slightly. If you HAMMER it to start it and it cocks; then it;s JUNK. Everything has bushings; cams, rods, distributors, transs, rearends, even carbs sometimes. You're not really a mechanic until you can do bushings.
 
Do you keep tapping until it’s where it needs to be? Or once you get it started and straight you hit it harder?
 
CudaFactHackJob said:
So what.
Click to expand...
I guess what I am saying is that asking for help with one small part of the rebuild after several failed attempts that are now resulting in broken or damaged parts does not mean he is not getting the experience of a rebuild. But I guess that's up to him. It's hard to guide or give advice over the Internet.
 
You'd be surprised how far you can get just tapping gently. Ever wonder why staking is necessary? Cuz they can just walk right out of there. Like I said, just use a large steel plate. Get down low so you can see the bushing from the side. Tap it and watch to make sure the plate doesn't **** at all. The large radius of the plate is the key to seeing if the bushing is going in straight.
 
all the advice has helped a lot. I couldn’t have done it otherwise

I’ll get a plate for this last one and the ones in the future.
 
glhx said:
all the advice has helped a lot. I couldn’t have done it otherwise

I’ll get a plate for this last one and the ones in the future.
Click to expand...
Glad to help. But I am not an expert either. I have messed up my share of bushings and broken a few components too. But I also have never had a rebuild that did not work either. But a guy like cuda hack who does it for a living is a big help. He has taught me a few things too from his posts here.
 
If the bushing cocks the slightest bit, it will be amplified by the large diameter of the plate. I have a thicker steel plate for the bottom and it's also a large diameter. Looking between the two plates; it's easy to see if it's starting in straight. First time I needed a bushing it was to install a wider one in a th350. So I bought two bushings and took it over to the local machine shop. I handed the guy the drum and bushing and he said I'll be right back. I said I got an extra bushing if ya need it. He said no. Two minutes later he was back for the second bushing. THAT'S when I decided to learn.
 
Ok......so I think I figured it out.

When I put this bushing in the first time. It made a high spot in the middle. As if it compressed too much and buckled.

So the second time.....I measured where it rides on the race. Pressed it down a little further. It still buckled. So I pressed it all the way down AFTER it buckled. I pressed it down so far that it was almost seated and bottomed out......maybe 1/16” from where it could go no further. . I knew this was too far but I wanted to see if the race was causing it to bottom out. IT WASNT the race and still wouldn’t clear the high spot.

I had said earlier that there was a trick to getting this reaction shaft bushing out. Find the puzzle piece. Grind it down....weaken it......and use a bushing driver to grab the edge and it pops right out. 3 to 5 minutes. There is a special tool for this that threads in there but I wasn’t messing with that.

Well......this time. I had pressed it so far down in the hole that the trick didn’t work. I left no room for the driver to grab onto. I knew that would happen before i did it so it was fine.

...........
So now I can’t get the bushing out. I couldn’t find the puzzle piece. It was too far down the hole and there really wasn’t enough light to see it.

So......I ground 4 areas of the bushing .....yeah.....by the way. I ground the area on the sides of the puzzle piece and missed the joint buy 1/8” on each side. I was grinding with the dremel with the grinding fitting extended all the way out so I could reach down in there.

So now this was more like a blind bushing without the ability to get much behind it how hammer it out.

So......I decided to HAMMER IT IN......That’s the trick here. It was weak enough ground on 4 areas and one really thin on that it cocked it sideways and bent it. It buckled it to the point it came off center. The bushing was weak and so weak that it didn’t damage the housing. I knew this before I hammered it.

Once it was buckled so bad.....I took the driver from the other end and hammered it out with ease. Caught the edge like I did last time.
Plus.....the main problem with this whole thing made it harder to get out because it was tight in there......down below explains that

..............
So what’s the problem here. I found out when shining a narrow beam light after the bushing was out. That there is a very very fine lip right where that hole is inside the housing. I might have pressed that bushing 1/32 past that lip and buckled it. It was so fine I didn’t even see it.....several times. I pressed it to pretty close where the old one was pressed in. It has to be precise.

When the bushing was out. I felt down in there where the hole inside the housing is. It felt that rougher than where the bushing sat. It wasn’t precisely machined.
So I was pushing the bushing in too far. Just not by much.
And that lip was so fine and so small that you could
Not even see it and it allowed the bushing to push right into it.

Moral of the story.....there isn’t one. It’s an easy mistake to make.
So....
Tomorrow I will press the bushing back in a lot more precise in the depth and not go past this fine lip. That’s why it buckled and was so tight.

The rest......
The parking spring, pan bolts, and filter screws came in that rat rod al sent. The new or used output shaft is in so I’ll see if that fixes the governor play.

Waiting on the drive shell replacement that I warped......and the front ring gear support that’s a 4/4. Had to verify measurements on that as........the part number is the same.......
So check if you ever have to replace that 4/4 and 4/3.....same part number.....different heights of splines. Could cause a bad day if the 4th clutch has nothing to grab onto. I saw a thread where that particular situation drove someone nuts for a few years. So I checked it carefully

Then both transmissions should be good to go. Each failure learned just bit more.
 
You know both of them are gonna work.

All clearances were checked.
Specs were checked.
All parts worn out were replaced.
New seals and washers
New bushings and clutch packs.
Torque converter is new

You saw pictures of ever part in detail and no red flags were raised.
The valve body instructions were followed to the letter
Every part was washed
All bushings were rotated and checked.
Any part I messed up or had a question of being messed up was replaced and checked

There is a good chance this is going to be right the first time out.
I guess we will find out
 
Last edited:
Lol......
Soon as I port the 302 heads for the first time.
Shake you head now and another for the road
 
48953993691_4fedf74f7c_m.jpg
-

The annulus support I broke. The part replacement came in today.
The part from 67 and 76 has the same part number.
The teeth in the 76 are taller.

There is a sub number
67 L28
76 L23

I thought I would post pictures of this NOS replacement box. It’s date coded 76 and the teeth measure correctly. I would say I got lucky with this as finding a loose 4/4 gear of only the one part I broke would have been kinda rare. They switched to the 5” gear not too long after this I think and 76 was the first year for the 4/4.
The 5” came out not much later and they retired the 4/4

Cool old box I’ll save

48953993691_4fedf74f7c_m.jpg


I don’t know why it posted the picture twice
 
Last edited:
Th new output shaft.
It gave the same play on the governor housing.
.023

So it has to be the parking gear or the governor housing. The governor housing would have a snap ring wear mark as I’ve seen the later models have. It’s a softer aluminum piece.

The cast iron parking gear that the governor bolts to would not have worn and has no signs of wear either.
I’m wondering if chrysler just made it this way.

The 904 has considerably less play and didn’t wear the output shaft support
 
I put 2 more reaction shaft bushing in. Had 4 on hand.

I still used the bushing driver that was machined down.
I had machine it just a tad too far so instead of putting them in perfect. The went to the side some and sort of mushroomed the bushing.

Like cuda said. I tapped it in instead of pressed it or hammered it.
It was much better than before. The press was what was causing the high spot in the middle. Not the lip as I said before. It was just too much force at one time.
Hammering them in was similar but not as bad.

The bushing driver was fine before. It wouldn’t go past the hole and pass through but that would have been fine....I had machined it just a little too far on the inside. Lesson learned.... The press pressure caused the problems for this particular one.

Had I used that bushing driver before I messed with it and tapped the bushing in with a hammer. it probably would have been fine


Now for the thin bushing. It was hammered in and went in fine. It had a high spot that I carved down with a knife. Then the planetary gear slid in with some tension. Which meant it was also mushroomed.
I should have tapped this one in as well.....but it would have been ok.

So......
I took them both down to the transmission shop. For $20 he pressed in new bushings and gave me some education. For him they went in With no problems.....I said here’s two....
He told me he would only need one.

I asked him how he did it with no issues.

He told me that when he first started he was using bushing drivers and having the same problems I’m having. He said he bought an arbor press with adjustable expanders that were similar to a cam bearing set up.....40 years ago and had saved about $75-$100,000 in bushings since.

My other bushings were put in with a cam bearing installer which is a set up like this so it made sense as they had 0 issues.
They have the steel expander rubber band that presses inside the bushing. The expander is adjustable so it becomes the perfect size every time....and the sticky rubber keeps it from moving. You can pound the heck out of it and it won’t ruin the bushing just like a cam bearing installer doesn’t trash cam bearings and they are hammered in with a sledge hammer. There is a learning curve to installing cam bearing as well.

......There is no downward force on the top of the bushing like a bushing driver has that can mushroom the top side of the bushing. All the pressure is spread out evenly.

With bushing driver.....perfect sizing inside and out ....and tapping on some of them can make all the difference.

But the arbor press is more straight.....more precise and has better and slower.....more precise control....and can still use the bushing drivers with it or expandable pieces.
I’ll be looking into getting one of these for the next build which will be a 700r4.

So now all the bushings are in and turn freely and some lessons were learned.....again.
 
I’m finally now finishing both of these up.

All the clearances were checked and everything torqued and the clutch packs were air checked.

The bands were adjusted. The 904 is back together and Now that it’s done. I have a couple of comments
It feels tight on the input shaft. Of course that’s relative. It had trans goo paste instead of fluid with makes for a lot more resistance. Nothing was put in dry. The shaft turns fine. I have to use a short pair of channel locks to turn it. I can’t by hand very well. Output shaft is the same hard to turn by hand.

So here’s how I adjusted the bands. At first I torqued down the reverse band to 72” pounds and backed off 3 1/2. This kept the reverse locked down and wouldn’t turn at all. I turned back out another half and it was free but seemed to ride a little tight. This might have even been correct

When I got to the front band. It had a square head on it.
So.....what I ended up doing, was to back both the reverse band and the kickdown band back out to where they were both loose.

I tightened the reverse band down to lock it. Then slowly backed it out where it would barely turn freely and smoothly. Then backed it out 1/2 turn....it would turn faster with less resistance.

Once the reverse band was set. I went to the kickdown band and did the same thing for that one. Rotated the transmission the other way. Locked the kickdown to full lock and slowly backed out until it would barely Turn. Then went another 1/2 turn out for that one as well. Hopefully these are right......I know they are at least not too tight.

The 999 is still mostly apart but will be finished today. I guess I’ll adjust those the same way.
 
Last edited:
You must log in or register to reply here.
-

Back
Top