Welcome all 928 forum refugees!
  • User avatar
  • User avatar
  • User avatar
User avatar
By Crumpler
#46631
I’m wondering if you guys have seen this before.
The big o-ring at the base of the water bridge going into block: failed after about six years. Pics show old and new o-rings with quite a bit of flattening. It was happening on the rear side. No cooling issues but it would leak about 10 minutes after at hot shut down.
Kudos to Roger, he has a nice kit pre-packaged, with all parts including new hex bolts.
Image
Image
Image
User avatar
By worf
#46665
I don't see anything out of the ordinary. As far as an old water bridge o-ring is concerned?

Maybe I missed something.

In my experience that big o-ring always comes out 'flat.' It takes 10-ish years - depending upon use of course - before it starts to allow a bit of coolant to seep during temperature transitions. Then after a couple of years it leaks enough to fill up the 'wells' on either side regularly.

I see it as just another example of '70s-era engineering like all the other big 'rubber' pieces that seem to have about a 10 year lifetime (e.g. cam cover gaskets, etc.)

I've found that for the cam covers, at least, Permatex outlives the gaskets. It's a time sink to do it cleanly but IMO worth it.

Perhaps today there's a material that won't outlive its plasticizers for 20 or 30 years that could be used for the o-ring. Possibly the new ones are a different material. Or maybe not. It's not like PAG would intentionally make a replacement part better than the original. In fact, as gaskets go, some of the new ones are worse.

In any case, it's an o-ring with measurable dimensions. Experiments could be conducted. But, given the 10-year(-ish) time frame for results, materials selection should be done by someone skilled in the art.
User avatar
By Crumpler
#46675
Ok, in that context it makes more sense to me.

Also, it sounds like that is an area where you do expect to see a coolant leak, given time/materials/design.

I guess I was disappointed in the timeframe. Maybe the track days have accelerated the changes.
User avatar
By worf
#46686
Crumpler wrote: Fri Jan 01, 2021 4:48 pm Also, it sounds like that is an area where you do expect to see a coolant leak, given time/materials/design.
Unfortunately yes. I expect them to last 10-15 years with some 'seepage' towards the end. I recall seeing 1 or 2 that were beginning to seep 'round 8-ish years and 30k+ miles.

I'd have to go back in my records and see the 'oldest' intake refreshes I've done to get good data. And off the top of my head I don't know when I'm 'expecting' them to be in need of a second round. But, I'd guess that there are a lot of 2005-2010 intake refreshes that have reached the 10-15 year threshold. On the other hand, some of those 928s have changed hands, been totaled, or just vanished etc. The at-hand data point might be my '91 which got it in 2007. But, it's on a storage lift. So, I can't just run and check it.

Crumpler wrote: Fri Jan 01, 2021 4:48 pm I guess I was disappointed in the timeframe. Maybe the track days have accelerated the changes.
I think that's a 'tough spot' for a big rubber gasket of unknown-but-presumably-70s-materials selection to live. I expect that the expansion rates of the bridge and block are quite different and obviously there are wide temperature transitions. So, it's possibly getting heat shocked and "crushed" every time the block transitions to operating temperature.

AND it has to be resistant to coolant. Come to think of it, I wonder if the coolant age (depletion of the additive package) effects the lifetime of the gasket. The coolant may also limit choices of materials. Less so today than 40 years ago. But, coolants and their additive packages have to take into account every material touched by it.

In any case, while I see them seep at various rates, I've never had one 'fail' a cooling system pressure test cold. So, this supports my theory that the seepage occurs during warmup due to loss of plasticity and differential expansion rates. Of course, it may leak a bit when hot, but the coolant will evaporate.

Lot's of speculation their.
User avatar
By worf
#46721
Shifted wrote: Fri Jan 01, 2021 6:55 pm Wouldn't a wax paper gasket be a good upgrade?
Try it. Let us know how goes. :popcorn:
User avatar
By worf
#46722
Wait. R U Serious?
User avatar
By maddog2020
#47655
worf wrote: Fri Jan 01, 2021 6:29 pm
Crumpler wrote: Fri Jan 01, 2021 4:48 pm Also, it sounds like that is an area where you do expect to see a coolant leak, given time/materials/design.
Unfortunately yes. I expect them to last 10-15 years with some 'seepage' towards the end. I recall seeing 1 or 2 that were beginning to seep 'round 8-ish years and 30k+ miles.

I'd have to go back in my records and see the 'oldest' intake refreshes I've done to get good data. And off the top of my head I don't know when I'm 'expecting' them to be in need of a second round. But, I'd guess that there are a lot of 2005-2010 intake refreshes that have reached the 10-15 year threshold. On the other hand, some of those 928s have changed hands, been totaled, or just vanished etc. The at-hand data point might be my '91 which got it in 2007. But, it's on a storage lift. So, I can't just run and check it.

Crumpler wrote: Fri Jan 01, 2021 4:48 pm I guess I was disappointed in the timeframe. Maybe the track days have accelerated the changes.
I think that's a 'tough spot' for a big rubber gasket of unknown-but-presumably-70s-materials selection to live. I expect that the expansion rates of the bridge and block are quite different and obviously there are wide temperature transitions. So, it's possibly getting heat shocked and "crushed" every time the block transitions to operating temperature.

AND it has to be resistant to coolant. Come to think of it, I wonder if the coolant age (depletion of the additive package) effects the lifetime of the gasket. The coolant may also limit choices of materials. Less so today than 40 years ago. But, coolants and their additive packages have to take into account every material touched by it.

In any case, while I see them seep at various rates, I've never had one 'fail' a cooling system pressure test cold. So, this supports my theory that the seepage occurs during warmup due to loss of plasticity and differential expansion rates. Of course, it may leak a bit when hot, but the coolant will evaporate.

Lot's of speculation their.
this is where the failure rate depends on the installation and the process as much as the part used. I've seen some 928's that a PO or shop ran stop leak through or something similar, and there is gunk buildup on the aluminum surfaces. If the original seal leaked and coolant was baked into the area above the seal (as in a very long term leak) then to install a new one, you really need to do this with the water pump out so you can clean out the sealing surface and flush it well, and have good lighting. If you have the water pump installed and the coolant wasn't drained, you are not able to see down the hole with enough light to really see the surface that the ring seals on. Also the bridge needs to be clean enough to eat off of. Scrub the ring sealing surface with a stiff plastic brush to make sure you get any scaling or buildup off the bridge. most owners (& some shops) will just replace the ring and assume the surface is good to go. This area is notoriously dirty due to the location for oil leaks and just general grit flying around ( especially if there is no belly pan)
User avatar
By worf
#47719
maddog2020 wrote: Tue Jan 05, 2021 6:25 pm this is where the failure rate depends on the installation and the process as much as the part used...
This is one of those things where I have to think about it. Kinda like when Greg Brown told me he frequently finds teeth marks on crank pulleys from folks using big channel locks. I'm like ... why? OH, because they don't know about fly wheel locks.

Why wouldn't you drain the coolant when doing the water bridge? Why wouldn't you clean gasket sealing surfaces?

The last couple of water bridges I've installed looked like this one before putting them back on:
Image

Might need solution for this depending how testing[…]

The 928 Photo Thread

I think that the chip in my instrument panel has g[…]

Yeah - I am putting everything back to stock ... t[…]

I hate insurance companies

UPDATE: I told my insurance agent we were cancell[…]