Oooh boy. This is why you need to take what I say with a grain of salt. So I was looking at some cross sections of bearings and I was wondering “gee, how do they assemble those?” I mean what holds the inner and outer race together?

Damn. The cage does in a way. If the balls are held equally spaced in the cage there is no way (without a lot of force) to separate the races. The balls hold the bearing together…till the cage breaks apart and all the balls can migrate to one side.

So my technique of wiggling the bearing to look for race and ball wear is only half good. You need to make sure the cages (usually made of flimsy sheet metal or plastic) are in good shape and not falling apart. I had no idea that these flimsy things are what kind of holds it all together.

Wow. The internet. An old dog like me can still learn. We can just look this up. BTW the How it’s made ball bearing video glosses over the fitting of the balls and races…probably to protect trade secrets for ultra precision annulars that don’t have much slop to slip things together.

ETA - ok you got me worrying about this. My bearings looked and felt good when I last looked at them but I didn’t realize that the cages can fall apart and the bearings can drop down. Upon visual inspection the cages looked intact but the cars are 40 years old and that bearing spins around several times per revolution of the engine. Probably a good idea to preemptively replace them. I’ve got about 300k and 175k on my two turbodiesels. So I bought four SKF 626-2RSL (double rubber shielded) annulars for $18 on eBay. Not bad. I’ll get a couple of Reinz vacuum pump gaskets from Pelican and I’ll be good to go.

__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD)
82 240D stick shift 335k miles (SOLD)
82 300SD 300k miles
85 300D Turbodiesel 170k miles
97 C280 147k miles

Now we’re talking!

It always felt like minimal effort to avert something catastrophic.
Was always surprised servicing wasn’t common practice.

Let me know when your bearings arrive. I’d like to follow suit.

__________________
1981 240D 4sp manual. Ivory White.

Got the bearings yesterday. SKF 626-2RSH. They are properly marked and feel great. They’re 8 ball but I really doubt this being an issue. The whole six ball eight ball debate seemed kinda nit picky and not well informed. Just my opinion. I mean if six ball is so much better why am I changing bearings?

Will get them in today.

Amazing.

Are you removing the cam follower arm or replacing bearings in situ?
The hydraulic press rigamarole for that step left me a tad spooked.
Not sure if I misunderstood, but seemed removing the piston was necessary to remove the arm.

__________________
1981 240D 4sp manual. Ivory White.

Hey,

I finished the job on my 300SD. To answer your questions, yes I pulled the cam follower. I don't think there is a way to do it without pulling it. The interference fit is really close. Everything needs to be pressed apart and together. Tight German precision. I used a vise, some sockets and a drift punch. Removing the piston was 100% necessary. After fussing around trying to reassemble then stopping for lunch and thinking, I found a big C clamp and pressed it all back together. I really cobbled it together but the clamp worked well.

Here is the lowdown in pictures...because we are visual creatures.

The new bearings were indeed SKF 626 rubber shielded.
Eight ball. I got the four bearing special because I have two cars. Worked out to ten bucks a car. A couple of bucks for the gaskets.



On My SD there's plenty of room to get it out without removing the fan and shroud. Just loosen the power steering belt and pull the pump.



Yes the pump piston must come out. The screws are in really tight. I used this T handled hex bit screwdriver to pull the screws. It really helped.





Press the pins. Press the bearings. The snap rings are really small. None of my pliers fit them. This job was actually pretty difficult. Things just didn't want to come apart. Pins were worn in and stuck, everything needed pressing, etc. It just was a lot harder than Beagle made it sound in his thread about "just change the bearings it's easy." It was kind of a pain is what I'm saying. I think it took me four hours with cleanup.





Old bearings. SKF six ball made in germany. Unshielded. Plastic cages. In great shape for 40 year old bearings. No reason to take them out. I think the last time I saw them was five years ago. They look the same as they did back then.







Replaced with these. SKF Italy. Double shielded and sealed. I pulled the seals off. I believe (I reserve the right to be wrong) that they need to be open like the OEM ones so they get splashed with oil. I think this keeps them nice and cool. The failures we see may be cars where the oil was never changed or it ran low. Most of our cars are second hand, third hand, rescues...etc. A few of us have these ones passed down from our dads or bought when we were big shot dudes in the 80s but for the vast majority, our cars have these questionable histories. I think if you don't change oil or run it low it may overheat these plastic cages and the ball bearings fall apart.

That's my theory. So I cannot really tell anyone to do this or not do it.

I'm not sure if this is an improvement. 8 balls not 6. Steel cages not plastic. Italian bearings not German. All SKF. I'll say it again, my originals were in great shape. I'm wondering if it was a good idea changing them out. Now I'm all paranoid that going in there 1) introduced dirt, 2) added a bearing that may be lower quality, 3) messed up the gasket and created a leak, 4) openend me up to a misassembly mistake - like did I seat those circlips in the grooves properly? Did I get the valves in the right way? Of course I did but everytime we go in there we run the risk of mucking things up worse than when we found it.

My fancy press. Took a bunch of failures before I came to this. I tried with my hands, blocks of wood. Eventually I just slipped a socket over one of the screws on the front side and pressed directly on the idler wheel.





All done. It would have been really nice to have a bag of seals and check valves to rebuild it with while I was in there. But getting new bearings in is a big leg up.

Sorry to sound so negative but this was a really figity job and I am wondering if it was unnecessary. I would not call it easy just go in and change a couple of bearings like on a bike wheel and get out. It was a pain and my old bearings look pretty dang good. I think it really depends on how the car was treated in the past. Uhm...I just said that. Time for an advil.

__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD)
82 240D stick shift 335k miles (SOLD)
82 300SD 300k miles
85 300D Turbodiesel 170k miles
97 C280 147k miles

Holy smokes.

Well first of all, hats off to you for the Herculean (or Sisyphean?) effort.
I’m sure this will help many a happy diy-er down the road.

It sounds to me like your original bearing cages (plastic) were in working order?
Based on this experience, do you feel like you’d know what to look for in determining bearing health in situ?

Think I’ll still pull mine to have a look.

__________________
1981 240D 4sp manual. Ivory White.

I don’t know Shern. After reading the stories of bearings in the timing gallery I have to say, this is an issue. So I cannot say it is meaningless to change the bearings. It’s just that on my cars they’re fine.

One problem is I’m not sure what the failure mode is. It is likely the plastic cages. But do they get brittle with age? Do they destruct with low oil or old oil? Does leaving a leaking vacuum system for years overwork the pump? I don’t really know.

Looking at my cages they’re in great shape. I poked the cage with a pick and the plastic was still strong. The balls are still mirror smooth with no play in the cages.

If you just want to inspect I’d say first look for play. Then spray the bearings out so you can see them. Run the cages around so you can see them and see if they’re cracking up or loose. Check for embrittlement of the plastic. Don’t forget to oil them up before reassembling.

I got the feeling this is made of some really good plastic that lasts forever under very harsh conditions. But it’s good to check. Who knows what the POs did to our cars. For example maybe the poured in some harsh leak seal or engine flush in the oil and it compromises the plastic in the cages. Anything is possible after forty years.

The good news is it’s really easy to check. I had the pump out on my bench in half an hour working slow. And my feeling is it important to at least check.

I’ll be honest, I’m not a mechanical engineer and have no idea how to spec bearings. Reading about all the different variants of ball bearings and when and why to use each one makes me feel that I’m just a hack throwing some eBay bearings at a well thought out system. Seriously 6 ball 8 ball, shielded/sealed/open, plastic or metal cages, CN or C3 tolerance, ABEC1 or 3, I know what but I don’t know why if you know what I mean. Like I wanted the tightest tolerance ones at first because they were the most expensive then I read the loose ones are better over wide temperature ranges like engines. So we are kind of making engineering decisions on our cars.

Also reinstalling them is fraught with opportunities to damage the bearings. The order you press and support the inner and outer races gives rise to multiple opportunities to indent the bearings. I think I did mine properly but it is really hard to support both the outer race and inner race while pushing the pin through at the end. It left me wondering if I peened up my races in the end. I used sockets but they are not the same as precision mandrels. At some point you’ll have to press on the balls depending on the order you assemble because there is an interference fit on the outer race and the inner race at the same time. Oh well, so much for perfection. Just saying I have this nagging paranoia I made things worse.

Speaking of which, we are going to have to do this more and more as we deal with NLA. I am seriously thinking of a) going down to PNP and getting a pump to put on the shelf, b) researching electric conversions. This is one of those parts that you need to operate the car that will get harder and harder to get as time goes on.

__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD)
82 240D stick shift 335k miles (SOLD)
82 300SD 300k miles
85 300D Turbodiesel 170k miles
97 C280 147k miles

I decided to replace the bearings on my blue 300d last night. It was an identical procedure. I did have to take out the fan and shroud though. It’s tighter on a w123.

One big thing I noticed after I was cleaning up was that the bearings had a lot of play. They were quite worn. I didn’t notice when they were pressed in the mechanism with the axel pressed through them. I could feel it more when they were individually held in my fingers and rolled. Somehow when assembled I couldn’t feel the play. Probably the axel and second bearing kind of supported each other. But alone there was noticeable side to side looseness.

I dug out the bearings I’d pulled from my SD the night before and they too had the same play. I’d missed it. It’s subtle wear but it is there and had the new bearings wiggled this much I’d have tossed them in the dumpster. The new ones have no detectable side to side wobble.

The wear is most likely in the races. The balls looked good.

So…I’m actually leaning in the direction that the bearings were not that good and they were on their way out eventually. I know I sound flip floppy but now I actually think it was time to change bearings. I don’t think they shipped like this. It’s wear. Note my cars are mid life turbodiesels with less than 300k miles.

I wouldn’t say I was in imminent danger of a bearing bust up but I’m gonna take back what I said about the bearings being fine. They clearly were well worn. I just didn’t test them outside the assembly. So this makes my entire old inspection procedure I’ve described here suspect. Sorry…just posting what I’m learning as I go.

The set out of my 85 were not marked either. They looked a bit different. Six ball, plastic cages of a different design and unmarked.

Just thought I’d add that extra info. And as usual, take my opinions with a grain of salt. I get things wrong all the time and I rush out and write it down.

__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD)
82 240D stick shift 335k miles (SOLD)
82 300SD 300k miles
85 300D Turbodiesel 170k miles
97 C280 147k miles

Now that's tenacity I can get behind.
I'm floored you went in for round two!

I've spent a fair amount of time studying both Stretch's and your tear down photos.
Now that I know what to look for, it seems possible to inspect the bearing play provided I'm able to spray enough oil from the bearing cages (I'm assuming brake cleaner won't be caustic to the plastic??)

You mention here noticeable "side to side looseness." If the bearing was standing on it's side (like a bike wheel) Are you referring to the X axis (or technically Z)? That seems difficult to inspect with the bearings still installed. Do you recall whether the Y axis (up down were also worn). I'm leaping a bit here but I suppose bearings wear spherically rather than radially?

Hats off to the intrepid voyage!

__________________
1981 240D 4sp manual. Ivory White.

Just a guess. Tight vacuum system and perhaps less chance of wear in the bearings. Pump may not have been designed with a constant duty cycle intended. Dead giveaway is a car that will not hold vacuum overnight perhaps.

Lack of oil changes will put more soot etc through them than manufacture intended as well. I always thought that keeping an old car up goes a long ways to keeping it reliable.

A road breakdown can prove far more expensive today than keeping the car up. Plus in my experience a road breakdown. Not always but often involves a serious loss of time factor. Especially on a trip. A service function was available at almost every gas station at one time. That is ancient history now.

Every area in the 616 and 617 cars that can cause reliability problems is very well known.

These bearings operate in a different way than usual. A constant back and forth movement of the balls. Has perhaps an unusual loading effect of the bearing cages. My best guess as I know next to nothing. Six ball bearings where used to allow stronger cages. So keeping the vacuum pump in good condition perhaps means keeping the vacuum system tight. The only item in the system I can think of designed to lose vacuum. There is probably a better term for that. Is the power brake canister.

Easy test I think is letting the car sit for a few days. Doors locks should still work. Most preventative maintainance on these old cars is more time intensive than money intensive.

Yeah I’m really bored right now. I think I’m doing this to avoid other things.

I guess I’d say it’s wobble in the x direction. To better describe the kind of play I’m looking for is like when you Jack up your wheel and grab the tire and try to wiggle it back and forth to see if your bearings are shot. I’m just doing it in my finger tips. I’d say the grooves in the races are gradually getting deeper and wider.

Apparently having the bearings inside the assembly hides some of the wiggle. I’m pretty sure I could drive another 100k on the old bearings without issues. But what’s 100k on one of these cars?in my opinion if you get it apart you’ve done half of the work. Might as well swap in new bearings and don’t look back. At $5 a bearing I can’t complain.

@barry yeah the application really puts the bearings through some crazy cyclic loading. I’m sure it isn’t easy on them.

__________________
79 300TD “Old Smokey” AKA “The Mistake” (SOLD)
82 240D stick shift 335k miles (SOLD)
82 300SD 300k miles
85 300D Turbodiesel 170k miles
97 C280 147k miles

Kudos to you for thinking this thing through and all the details .

__________________
-Nate
1982 240D 408,XXX miles
Ignorance is the mother of suspicion and fear is the father

I did then what I knew how to do ~ now that I know better I do better