Thursday, February 20, 2014

Blown Head Gaskets on Land Rover V8 Engines

I've got a blown head gasket in my Land Rover.  What will it take to fix it?  That's a question Land Rover service managers hear fairly regularly.  We used to do head gasket jobs all the time, but in the last 5-7 years many of those jobs went bad. Many shops won't even do LR valve jobs anymore.  Why?

This is a big issue for people with the original LR-Buick aluminum V8 engine - all the US market Defenders and Range Rovers to 2002 and Discovery I and II models through 2004.




The engines overheated and failed either right after the head gasket job, or within a few months.  At first I thought the failures were comebacks - errors of workmanship.  But I soon realized the problem was not the workmanship - it was the motors themselves.  They were failing internally, in seemingly invisible ways.

Join me now as we look closely at a seemingly simple head gasket failure, and find there is much more to the story . . . 

This 2001 truck came into Robison Service a few days ago (early 2014) with a noise complaint.  Exhaust gases were blowing out between the head and the block.  It seemed like a classic case of fire ring failure in the head gasket.  In years past, I'd have been right on it with new head gaskets and a cleaned up heads.  Today, not so quick . . . As it happens, our caution turned out to be well founded.

Here are both heads off the motor.  As clearly shown, only one has a blowout (second cylinder from the left, upper head)  So far, it looks like a blown gasket.






OK, now lets look at a closeup of the failed cylinder in that upper head.  The blowout line is in the center.  Look at the holes for the head bolts.  The one on the left is actually facing the front of the engine, as you are seeing the head upside down (we will look at that hole in the block in a minute)   Note how its orange from coolant intrusion.  The one on the right (the rear bolt hole) is clear and that’s what you want to see.

The orange is sedimented Dex-cool.  Sedimentation is a known issue with Dex cool and it's particularly visible here.  It looks very different from freshly spilled coolant, like you get from teardown.  Scroll down the page four more images and you'll see freshly spilled coolant in a head bolt hole.  There's no mistaking the difference.

There are some motors where the head bolt holes go into the coolant galleries.  This isn't one of them.  These holes should be dry.  The fact that they are not suggests a crack or leak somewhere.

The blowout has damaged the head slightly but it's nothing we can't fix with a weld and resurfacing before the valve job.  Right now, that is the least of our worries.



Like I said, that dried coolant is a bad sign.  There are no coolant passages from block to head in the middle cylinders of a Rover-Buick V8.  There is no connection between the head bolt holes and the coolant galleries in the block, either. That area should never show coolant.  It's a bad sign, suggestive of internal failure.  But there's more . . . 

Now lets look at the block on the right side.




The blowout is clearly visible in the bottom of the second cylinder back.  Note the rectangular marks at the left and right edges of the block.  Those are the passages through which coolant flows into and out of the head.  There is no coolant flow through the middle.

The reddish spotting between 1-2 and 3-4 cylinders and the respective lower head bolt holes is even more suggestive of slow coolant leakage.  We will look at that in closeups.  When you look at the second and fourth head bolt holes from the left, those are dark.  The other holes are light, meaning no long term coolant intrusion.

This is a closeup of the lower head bolt hole and the edges of cylinders 1-2 on that side.  Note the pattern of leakage from the liners in toward the head bolt hole. See the “burnt” coolant in the bolt hole.  Those are sings of long term seepage.  This usually indicates a crack emanating from the bolt hole out to the liner faces on 1-2.  Also note the pattern of leakage continues onto the surface of the front piston, which is washed clean of carbon in the same area.  This is a sign of coolant intrusion into the cylinder and its location corresponds to the seepage on the block



For comparison here are the front and rear head bolt holes with a view up to the coolant passages from block to head.  In the rear one you clearly see some coolant leaked in disassembly and the difference between that and what’s dried and hardened in those inner bolt holes in the earlier images  is clear




Here is a top view of the blown head gasket.   There's no damage (other than the blowout) between this gasket and the head.  It's very clean and normal looking.



Closeup of the bottom reveals a different story on the engine block side.  That face makes the leakage pattern strikingly clear, and shows it’s all coming out of the block.  That's not surprising, because it can't come from the head.  There's no coolant in that area!  However, the area between the head bolt bore and the cylinder edges is hollowed out in the casting, for coolant flow.  That is the area that is vulnerable to cracking, and that's what's cracked in this motor.



The leakage traces on the bottom side of the head gasket really tell the story.  The way the gasket is discolored we know the leak has been developing over a long period of time.  What happens is that the crack grows, and as it does, the crack opening relieves the tension on the head bolt. That's probably one of the reasons the fire ring seal blew out.



Here’s a broader view of the bottom of the gasket, where two leaks are visible, to the left and right of the blown fire ring..  

Like I said, five years ago I'd have put head gaskets on this motor and sent it down the road.  And looking at today's evidence - it would have failed because the block was already damaged.  Head gasket leakage was a symptom of that failure, not the actual problem.  It may have lasted a week; it may have lasted six months.  Either way, it would have blown.

Here's a photo of a stripped engine block, after we cut it in half and circled the crack that made it fail.  The coolant passages are clearly visible


What do we do now?

The correct fix will involve removal of the liners, welding the cracks, and then fitting liners with flanges at the top, so that the liners will seal agains the head basket and coolant will not be able to go between block and liner and cause a blowout.

Read more about that in this article from 2012.

As a footnote to this story, when told about these issues, the owner of the truck told me he'd been adding coolant for a while for a period of months.  Now we know where it was going.  The coolant was getting burned in the cylinders as it leaked from the cracks behind the liners.  The fire ring blew out because the block cracks caused the head bolts to lose their clamping force.  We have an explanation for the whole thing, disagreeable as it may be to the one who has to pay the bill.  Still, I contend it's better to know the bad news up front than to discover it after a $3,000 valve job and head repair goes bad.


John Elder Robison is the manager of J E Robison Service, independent Land Rover specialists in Springfield, MA.  Find him online at www.robisonservice.com or on the phone at 413-785-1665

Friday, February 14, 2014

Restoring seats in collector cars from Europe

One of the issues to be aware of when buying "restored" cars is that the examples offered for sale were often restored with the resale market in mind.  To that end, those restorers tend to focus on the things you'll be able to see at an auction inspection.  More substantive things - such as would be revealed in a two-hour cruise - are often ignored or even deliberately glossed over.

Sellers will often take exception to my characterization, but the facts speak for themselves.  If it takes $50,000 or $100,000 to do a cosmetic restoration on a car, you can assume that a similarly thorough mechanical restoration (almost all of which will be invisible on superficial inspection) will cost at least as much again, maybe more.  Doing both will price the car well above the auction averages, which are based on superficial restorations.

You see that in "show winning" cars that have to be pushed off the field because they barely run.  You feel it in a "concours" car when you sit in the seat and it feels like you ass is on the floorboards.  That is the subject of today's essay.

Cosmetically restored seats look good but feel awful


You can't really tell if a seat is restored by looking at it.  You can see wear, obviously, and you can tell if the seat is crooked or mangled in some way.  But a seat can look perfect and still be totally worn out and uncomfortable.  How can this be?  It's simple.  People put new covers over worn out old seats all the time.  It takes far more time to redo the innards of a seat than to recover it, and the cover is what a buyer sees . . . 

Prior to the widespread adoption of foam seat pads in the late 1970s car seats were often made with a metal frame that held a steel box spring like you'd find in an old bed.  Those of you who remember vintage summer camp beds know that box springs wear out, and when they do, they just go flat when you sit on them.  

The steel box spring is often capped with burlap, which tends to crumble but is otherwise trouble free.  Above the burlap you will often find a Spanish moss or horsehair pad, and above that a felt pad and then the seat covering.  Those things provide the "look and feel" of the seat but the comfort will never be there if the box spring is collapsed.  When they get old, the burlap, horsehair, and moss also provide the brown furry dust that tends to rain down underneath vintage car seats.

Here's an example of the junk that falls out of old seats.  In this seat the box spring is so loose that the seat cover has fallen right out of the grooves in the base.



It's possible to remake metal box springs but it's getting harder and harder to find the materials.  Today, most restorers fortify those old steel springs with robust molded foam.  In these photos you can see us doing that very job on a seat from a 1964 Porsche 356.



We start by removing the seats, which is pretty easy on an old Porsche - they just slide out.  The top and bottom are separated, and the cover is removed from the base.  The frame and spring and "everything else" are in two piles on the bench.



This particular seat has good leather, and the felt and padding are pretty decent too.  We're going to tighten up the rod that forms the pleat across the middle of the seat, as shown below:

The rod

The pleat
Now we're going to trim the original padding where we'll be replacing it with foam.  We're going to install a two-inch thick dense foam pad which will largely take the place of the collapsed spring.  The spring will be compressed by the pad, which will sandwich it tightly, and the whole structure will be a lot firmer.



The pad makes the cover fit a lot more tightly, which reduces the chance it will fall apart on the car. Here's the assembled lower cushion.  On close examination, you can see it looks a bit more "full" than before we took it apart but to the average person it would look unchanged


In this car we are also installing period headrests.  Some of you will say "Porsche 356 didn't have headrests" but I offer this page from the 1965 workshop manual - which we still have in original print - which says otherwise and shows how to fit them.



Here they are, and here's the finished seat.  With the exception of the headrest I'm the first to admit it hardly looks any different.  But the difference when you sit on it is striking.






Doing over a pair of seats like this is a full day's work, maybe more.  But if you want to drive the car - as opposed to just look at it - it's time or money well spent.

The seats in 1950s to 1970s BMW, Rolls Royce, Mercedes or Jaguar are all made in a fashion similar to what's shown here, and can be taken apart and restored using similar techniques.  Sometimes you can buy precut foam.  Other times you have to cut your own with a hot knife.  Some times the inside of the cover will require repair, and that can be complex if the seat has a pattern.  The worst is when the frames have rusted because it's tricky welding sheet metal seat frames and breaks can be tough to repair.

John Elder Robison is the general manager of J E Robison Service Company, independent restoration and repair specialists in Springfield, Massachusetts.  John is a longtime technical consultant to the Porsche and Rolls Royce Owner's Clubs, and he’s owned and restored many of these fine vehicles.  Find him online at www.robisonservice.com or in the real world at 413-785-1665