GregBBRD

I hate to bring it up, but there are different hardness crush washer, confounding the required torque, in a huge way.

If your new crush ring came out of an "Elring" bag, or you ordered it from the factory. it will be a harder material than most of the other "aftermarket" crush rings, out there.
26 ft. lbs will work just fine.

26 ft.lbs on a "soft" aftermarket crush ring will result in a crush washer that looks like Ed's on post #48.

Rob Edwards

We are not interested in your fact-based agenda. This stuff is supposed to be easy.

GregBBRD

Don't confuse me with reality, right?
I find very little on a 928 that is "easy", at this point in time. Maybe that is because we tackle the "problem children" that others can not fix. It does get easier, when you have the correct parts, however.
Unfortunately, not all parts are created equal.

hacker-pschorr

Just a few days ago I was on another forum discussing a similar topic: quality of copper crush rings.
The gist of the discussion was crush rings that are not properly annealed that do not crush properly, or at least take too much torque to do so. A few of the posters agreed and said years ago they started simply annealing every crush washer themselves just in case.
General consensus was lower grade washers out of China, many of them are not annealed at all, just plane copper washers.

khalloudy

Another reason I use and insist that my selected shop ONLY use genuine parts. Not even OE. Always found subtle differences that made all the difference.

I only veer away from factory parts is when there is a significant design improvement from a trusted source (such as Greg’s parts) and it is needed.

bureau13

This would be the opposite though, wouldn't it? The correct ones are the ones that take more torque here, or am I reading it wrong?

hacker-pschorr

The whole point of using copper or aluminum as a seal is the washer needs to crush and take the form of both surfaces, that's how they seal. This is why re-using such washers is a bad idea, they are already "crushed" and when you work metals like copper, they get harder and more brittle.

A local friend who's a metallurgist by trade showed me this years ago. No idea how the topic came up, but he demonstrated that taking a used crush washer, anneal it, you can re-use it. Not ideal, but it will seal better than just re-using the used washer. I think this came up at the track when someone had to change their oil and nobody had a new crush ring in the proper size. We did have a torch, problem solved.

From this site:
https://copperalliance.eu/about-copp...ys/properties/
The scale is 40HV - 110HV for copper. For the sake of this discussion a used crush washer we'll say is 110HV (I have no idea, just trying to work to a point).
What hardness was used for the factory torque spec? Arbitrarily we could say 60HV, and using a washer that is annealed to 40HV would deform too much at the factory torque spec.

The problem is generic washers from your local auto parts store (or Harbor Freight) may not be annealed at all or not very much and could be a hardness of 100HV.....

I'm taking what Greg said to be the highest quality crush washers will be annealed within the same range the factory used to make the torque specs. This is also why I keep a supply of WURTH copper sealing washers in all the common sizes needed for Porsche's and VW's. Higher quality and consistency I'm used to.

Bottom line though. I'd rather have a too soft of a washer than a rock hard one when trying to seal something up.

Lot's of threads / videos on this:

dr bob

Bit of a tangent on the washers --

The picture that Ed shares shows the late bolt, the one with the tapered or coned section between the bottom of threads and the face of the bolt head. This takes a specific washer with the correct ID, and that ID is larger than the OD of the threads. Buying a washer with 12mm ID will net the funnel shape of the washer his picture. With that washer, the bolt will be a dangerous b!tch to get out. Rob's picture of the early bolt shows how it will take a wider washer face, plus it has a larger hex so larger OD too, demanding more torque to get the same amount of pressure on the washer.

These washers should be aluminum rather than copper. Fineas Galvanni, who coined that bimetallic battery theory with flowing stream water across the change in a frog's front pocket, indirectly reminds us to manage materials carefully in this service. Putting a copper washer between the steel bolt and the aluminum block, coolant flowing across the steel and aluminum junction, with a copper layer sandwiched at one end, just can't be a Good Idea. High-pressure connections demand the softer copper sealing washers. Low pressure applications like the sump drains, gearbox drains and fills, plus these coolant plugs, all deserve aluminum washers.

The industrial mechanics i get to observe once in a while use teflon thread sealing paste on the sealing faces of crush washers in hydraulic banjo connections. I've learned to do the same, plus I can reduce the banjo or drain bolt torque by ~~20% or so and maintain a reliable seal. The turbine guys are using big steel banjo bolts into bigger steel valve bodies, so they feel no desire to reduce the torque. Those are also many-thousand PSI, so the area of the bolt and banjo is included in their bolt-stretch calculation for torque. Our steel drain plugs and banjos tend to go into aluminum with aluminum washers, where the actual pressure on the sealing faces depends a lot more on the washer compression than it does on bolt stretch. The aluminum housing threads tend to yield well before there's bolt stretch, especially in a drain bolt that has effectively zero length for "stretching".

Interesting side -- The studs that hold the turbine shells together are stretched by a hydraulic tool that holds and pulls on the end of the stud with a precise tension. The nut is moved down to the 'correct' clearance from the flange face before the tension on the stud is relaxed, and it ends up with the right amount of stretch to hold the shells together. Thread and face friction play no part in the process. A little non-metallic nuclear-grade anti-seize is applied, but really doesn't do a lot once the unit has been heat cycled once.