3.0 main and rod bearings

[AintGotNoGasInI]

There seems to be a lot of differing opinions on bearings for air-cooled cases. There is information supporting using Glyco bearings and information against using them due to quality issues, using GT3 bearings ETC.

What seems to be the best practice and products regarding bearings for these motors?

[john walker's workshop]

Glyco mains and Clevite rods seem to work. Seen some ugly looking glyco rod bearings.

[Jesse911]

Glyco quality issues is mainly for the 3.2 is what I heard?

I went with Glyco mainbearings for my carrera 3.0 and ACL rod bearings (with coating). Still in the process of building the engine so can't comment yet on the quality.

[PeteKz]

If you are rebuilding an existing engine, then measure the clearances in the old bearings with Plastigage. If they are within tolerances, and do not look scored, then keep running them. They have already proved themselves in service, so why take a risk on new ones not being as good?

I know that's controversial, but I practice what I preach. If you are a pro builder, building engines for customers, then you want to say you put new bearings in it, but there is no good engineering reason to replace good journal bearing shells.

[mikedsilva]

Originally Posted by PeteKz

If you are rebuilding an existing engine, then measure the clearances in the old bearings with Plastigage. If they are within tolerances, and do not look scored, then keep running them. They have already proved themselves in service, so why take a risk on new ones not being as good?

I know that's controversial, but I practice what I preach. If you are a pro builder, building engines for customers, then you want to say you put new bearings in it, but there is no good engineering reason to replace good journal bearing shells.

hi Pete, genuine question here and not meaning to be challenging: how many engines have you assembled now and re-used the original bearings?
Cheers
Mike

[Henry Schmidt]

To answer the original question, measurement and prep are paramount.
We like the GT3 mains in the engines that use that size. We also use both GT3 rods and Clevite bearing depending on the application. When I say prep, we always spec the bearings, do any modifications we need, chamfer, drill or groove then we dry film coat all the bearing and most fiction surfaces in the engine.
Reuse bearings? not so much.
The issue with reusing old bearings is consistent bearing crush. A new set of bearings will sit proud in a journal and when properly torqued, the bearing conforms to the journal. This conformity remains constant until the bearing is damaged or the torque/crush force is released. Keep in mind, not every journal will be perfectly round or even be aligned constantly as the case is torqued together. Even when the case is in spec there is a + or - tolerance. If the clamping force is insufficient, the bearing will shuffle. This shuffling can damage the case but more importantly, cause a loss of oil pressure. There is also the issue of cooling. The mating surface between the back of the bearing shell and the case journal is used to transfer heat. A poorly fitting bearing will not cool properly. When the clamping force is removed the bearing will act like a spring to release pressure exerted during the initial torquing process. There is no way to predict how a used bearing will conform to the journal (even the same journal) during the new torquing process.
Bearings are cheap compared to a complete overhaul so new bearings are your best shot at success.
Don't fall into the trap....
"Not enough money to do it right but plenty of money to do it over?"

[PeteKz]

Originally Posted by mikedsilva

hi Pete, genuine question here and not meaning to be challenging: how many engines have you assembled now and re-used the original bearings?
Cheers
Mike

Mike, fair question. I've done about a dozen engines and reused bearings, but, full disclosure, only done that twice on a 911 engine, both times an SC engine. The others were American iron. I'm also a mechanical engineer, so have some detailed understanding of how bearings work, torquing assemblies, and the like.

Journal bearing shells don't "wear out" unless there was an oil problem of some sort--low pressure, loss of oil ,contamination, etc..

A serious question for others: At what point do you think it advisable to replace bearings? Some number of miles, or visual condition, outside of specs, or every time you split the cases?

I think everyone would agree that there is some point at which to replace a bearing shell. What point is that?
How about chains?

[PeteKz]

Late edit: I forgot about my years with VW air-cooled engines, so another handful of those under my belt. However, I did not reuse bearings in those engines. At that time, I just replaced them because they were so cheap. And, IIRC, they showed wear in most cases.

FYI, Gene Berg, the legendary VW racer and builder, was not a fan of reusing VW cases, even if correctly remachined. He understood that the magnesium VW cases had a limited life, especially if they ran hot. Something about the magnesium alloy heat softening and the cases deforming under heat and load. In 1975 VW went to a different Mg alloy, with more Aluminum in it, to reduce the case deformation (my guess is that as emissions tightened, and they went to fuel injection, the heat load increased, similar to the 1974-77 911's). IIIRC, those cases had "AS5" cast on them to indicate the improved alloy. Berg strongly advised to use those instead of previous cases, and preferred to buy them new, because back then VW cases were still being made new in Mexico or Brazil, and the cost was reasonable, something like $600 in the 1980's.

I have often wondered whether the alloy used in VW cases before 1975 was the same as in the Porsche cases. If so, then his warnings about not reusing/remachining worn out Mg cases applies to Porsche Mg cases. I still wonder why Porsche wanted to use more expensive Magnesium castings instead of aluminum, when they had a superior aluminum case in the 1960's. Just to save a couple pounds? Seems penny wise and pound foolish.

[CNB]

Originally Posted by AintGotNoGasInI

There seems to be a lot of differing opinions on bearings for air-cooled cases. There is information supporting using Glyco bearings and information against using them due to quality issues, using GT3 bearings ETC.

What seems to be the best practice and products regarding bearings for these motors?

Glyco, Clevite and Calico work well. Calico coated bearings are pricier, but have heard good feedback.

[Turbo_pro]

I always find it interesting when amateurs reject the advice of well respected experts by recommending half-assed solutions to save pennies when rebuilding a $100K investment. Pound in a flywheel seal vs proper installation during case assembly and now suggesting bearing crush is something to ignore.
Two 911 builds vs 40 years of Porsche only experience. It's wild but hey, to each his own.

[PeteKz]

Okay Jim: what are your criteria for replaced bearing shells? Every time you split the case, even it if was a zero-time engine that you had to disassemble because you caught a mistake? How about replacing rod bearing shells every time they are torqued to check clearances, and then disassembled before final assembly? Or after XX,000 miles? If so, then what is XX? How about on condition, or if they exceed the specs (my preference)?

Henry makes an important point about bearing crush, but let me also point out that in most cases, bearing shells aren't actually "crushed," that is, compressed into plastic deformation. "Compressed" would be abetter description, because they are typically elastically compressed, and when you release the pressure, they expand back to their original dimensions. Maybe Porsche air cooled 911 engines actually do "crush" the bearings when torqued to full spec, but that doesn't make sense to me, because that would cause the clearances to change, and when a metal part yields, it does not yield evenly across its length or cross-section, so that would cause the clearance to be slightly different at different parts of the bearing.

Henry, do you replace the bearing shells after you have done a test assembly to check clearances on rod bearings? After all, the second time you install and torque them, there will be differeces, as you pointed out. How much difference does it take to exceed your threshold for replacement?

I will also point out that I have seen overhauls of big marine diesel engines ($10,000's of dollars would be laughably cheap for overhauling those), and the machinists mates will typically measure the clearances, and if the bearings are within spec, they reinstall them. That's not just to save money, but also time, as it may take weeks to source replacements, and they might have to be made to order, and shipped from another country.

Furthermore, we have seen documented cases where the new bearings were not within specifications (e.g., Glyco). And, a bearing that has been in use for some period of time has proved its mettle (bad pun) in service.

I'm not just arguing for the fun of it, but because I really want to know and understand. It's an engineer thing.

And, I'm not telling anyone else how to rebuild their engines, especially pros like Henry, Neil, Mike, et al. It's also important to them to tell their customers that they put all new parts inside the engine, in case there's a problem. That's a different market than what most of us are doing as a hobby and for our own edification.

[Henry Schmidt]

The Engine Rebuilding Forum was started by Wayne after some errors were found in his original Engine Rebuild book. He started this forum to allow experts to chime in to help sort out some of the errors and to assist DIY guys with their rebuilds.
That said: I come here to offer my advice to anyone interested.
Having built well over 400 air-cooled 911 engines, we have seen everything you can imagine. My advice is designed to give the DYI guy his/her best chance at success.
Generally speaking, there is no cheap way to properly rebuild any engine but air-cooled 911 engines offer some unique challenges. My advice is free. Use it or don't. Challenging it seems counter productive.

[PeteKz]

Henry, I’m not challenging you personally. Don't take it that way. I want to know why. That’s all.

[Henry Schmidt]

"Why" is answered in post #6. Your attempt to equate non-heat cycled bearings used to pre-assemble an engine with old heat cycled bearings is a straw-man argument with no probative value.
Main journals in 911 cases are rarely perfectly round. A bearing that has run for few/many heat cycles will conform to it's specific journal shape. As soon as the case is unbolted, the journal will rarely achieve the same shape as the case is re-torqued.
Assuming an old/used bearing will achieve the proper bearing crush after previously being compressed is a foolish assumption.
Heat causes the deformation so cold to cold re-torque will result in similar results.
Your suggestion that a professional will use specific building techniques simply to appease a customer's perception is argumentative and rather insulting.