Type 4 Head Leak After 300 Miles?

Moderators: Sluggo, Amskeptic

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Post by Amskeptic » Fri Feb 16, 2007 9:13 am

satchmo wrote:
Well, of course no copper spray-a-goop will fill the gap when you have out-of-parallel surfaces. I don't think that's what we are talking about here, though, Colin.
Then I am confused. What is the copper goop here to do? Unlike a water-cooled engine, if you have proper contact area and proper clamping force, you are good to go with an air-cooled engine. The goop has real work to do in a water-cooled application because it really does have a chance to keep coolant (14 psi) and oil passages (60-80 psi) sealed. But combustion pressures? Heck no way Jose.
satchmo wrote:
The first is the amount of carbon collecting in the combustion chamber. I am concerned that this happens as the result of our desire to have a cool running engine. I tend to keep the timing on the low end of acceptable and the fuel mix on the rich side to avoid overheating.
Good thinking. But a bus on the road is like your oven on self-clean.
Set it up to run cool, yes, then drive the hell out of it. I found that carbon was only a problem for me because of oil entry into the combustion chambers from a loose valve guide in my retired original heads. Your rings are doing a fine job in my engine currently.
satchmo wrote:
Second issue is the head/cylinder seal. When I rebuilt my engine 30,000 miles ago, I used valve grinding compound on the cylinder and lapped them on the flattest surface I could find. Then I lapped each cylinder to the heads. Finally, I used the Copper spray gasket on the mating surface of both the cylinder and the head. No aluminum head gasket. This worked for me
The surface area of the seal between the barrels and the head is the key. If the entire surface area is evenly clamped, no leaks. Leaks will only occur at initial assembly if there is imperfection in the machining, lack of parallel, or incorrect head torquing (usually brought on by crud in the threads skewing your clamping torque values). Later leaks can almost always be attributed to an overheating event that causes the head/case surfaces to collapse thus reducing clamping forces.
Colin
(goop will help hold those sealing rings in place during assembly!)

User avatar
germansupplyscott
Trusted Air-Cooled Parts Vendor
Contact:
Status: Offline

Post by germansupplyscott » Mon Feb 19, 2007 6:51 am

colin,

there are a few reasons why the cylinder to head surface needs sealing help. look at the history of the engine's development. to use an extreme example, compare the surface area of a 40HP cylinder top surface to that of a 2.0l type 4 - i'll bet the ratio of selaing surface to bore size is something close to 2x on the 40HP cylinder over the 2.0 t4.

as the bore of the engine got larger and larger, it became more and more difficult to seal the cylinders. the 1.7 seals fine, but my guess is that is about the limit. the 1.8 has very thick cylinders, one reason why it is a good engine combo. it seals well also. then the bore grew a lot more with the 2.0 and VW introduced the head sealing rings. the sealing rings were there for a reason, the bore was getting too large to be able to seal well without other intervention, but the seal rings didn't work as intended, they introduced other issues. VW knew back then that the 94mm bore was not sealing properly and tried to fix the issue with a steel ring between the head and cylinder.
If the entire surface area is evenly clamped, no leaks.
this is the other issue particular to the type 4 engine. the cylinder head studs are not symmetrical about the centre of the bore. once the bore gets as big as 94mm, it is very difficult to get the heads to seal due to this fact. again it is geometry. as the bore increased, the amount of off-centre loading due to the uneven head stud placement also increased. you see the results of this all the time when you tear down type 4 engines, especially 2 litre engines. massive head leaks, and many of these engine were assembled at the factory. the fact is that as the bore gets to 94mm, the cylinder head to cylinder surface needs all the help it can get to seal off properly. prep work on the machined surfaces is the key, yes. and anything else you can do to give the joint a chance to seal, do it.

the reason the coppercoat works is because it is not "goop" as you call it, it is a fine copper paste in suspension. much of the carrier fluid evaporates after you apply the stuff, leaving a fine copper powder on the sealing surface. this acts like a very thin copper sealing ring. it works. it isn't snake oil. just because it isn't in the vw repair manual doesn't mean it doesn't work. just because it is used on a waterpumper doesn't mean it doesn't belong on an aircooled engine.
_________________
scott lyons
http://www.germansupply.com

VW Bus-Centric Online Parts Source

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Post by Amskeptic » Mon Feb 19, 2007 7:00 pm

germansupplyscott wrote: there are a few reasons why the cylinder to head surface needs sealing help. look at the history of the engine's development. to use an extreme example, compare the surface area of a 40HP cylinder top surface to that of a 2.0l type 4 - i'll bet the ratio of sealing surface to bore size is something close to 2x on the 40HP cylinder over the 2.0 t4.
Air pressure uses hydraulic principles. Sealing surface to bore size is irrelevant. Sealing surface to psi is relevant.
germansupplyscott wrote: as the bore of the engine got larger and larger, it became more and more difficult to seal the cylinders. the 1.7 seals fine, but my guess is that is about the limit. the 1.8 has very thick cylinders, one reason why it is a good engine combo. it seals well also. then the bore grew a lot more with the 2.0 and VW introduced the head sealing rings.
VW introduced the sealing rings with the 1.7 in the Porsche and 411's on up through the 2.0's.
germansupplyscott wrote: the sealing rings were there for a reason, the bore was getting too large to be able to seal well without other intervention, but the seal rings didn't work as intended, they introduced other issues. VW knew back then that the 94mm bore was not sealing properly and tried to fix the issue with a steel ring between the head and cylinder.
The issue never was about sealing. The issue was about changes in engine architecture after serious heat events. Those sealing rings work flawlessly until the head *loses* clamping force due to extreme expansion. In the old magnesium engines, the head studs simply tore out. In the later engines, case spigots would cave or the heads would warp. The sealing rings, being the softest link, would blow only as evidence of loose heads.
germansupplyscott wrote: this is the other issue particular to the type 4 engine. the cylinder head studs are not symmetrical about the centre of the bore. once the bore gets as big as 94mm, it is very difficult to get the heads to seal due to this fact. again it is geometry. as the bore increased, the amount of off-centre loading due to the uneven head stud placement also increased. you see the results of this all the time when you tear down type 4 engines, especially 2 litre engines. massive head leaks, and many of these engine were assembled at the factory.
If you look at your average 2.0 with 7.3:1 compression, and compare it to any of the performance engines that have been built using those head stud placements, I guarantee you that the stock engines were not flushing out that particular issue.
germansupplyscott wrote: the fact is that as the bore gets to 94mm, the cylinder head to cylinder surface needs all the help it can get to seal off properly. prep work on the machined surfaces is the key, yes. and anything else you can do to give the joint a chance to seal, do it.
The technical service bulletin that recommended deleting the sealing ring did not have to do with intrinsic weakness in the sealing ring. . .

(the sealing ring is but a canary in the coal mine. Its failure says, "low head torque." Why? crap in the threads during assembly? overheat that pounded in the aluminum spigots or distorted the heads? chronic detonation? lousy state of tune? leaves in the head fins -Porsche 914 owners take note-?)

. . . it had to do with instantaneous heat transfer to the cylinder walls. The 94mm pistons had a rash of seizures due to their greater expansion because of their greater mass. This would occur during transient overheats. The factory bulletin addressed this issue with a three-pronged attack.
1.) demand greater piston-to-cylinder clearance.
2.) cut notches in the rods to cool the pistons more readily
3.) delete the sealing ring to get the heavy load heat in the heads conducting down the cylinder barrels as quickly as possible to keep the barrels ahead of the piston expansion curve.
germansupplyscott wrote: the reason the coppercoat works is because it is a fine copper paste leaving a fine copper powder on the sealing surface. this acts like a very thin copper sealing ring. it works. it isn't snake oil. just because it isn't in the vw repair manual doesn't mean it doesn't work.
Scott. I did not spout off with anything like "it is not in the manual" nor did I say "because it is used on a waterpumper, it doesn't belong on an aircooled engine". I gave my reasons. I said there is damn close to nothing under the sun that is going to survive an air-cooled engine's combustion temperatures and pressures if that air-cooled engine has any sort of structural issues between the barrels and the head. And that's the truth. If you want to copper spray adhere your sealing rings to the heads to make them stay in place during assembly, knock yourself out. I am not saying the stuff will get in the way, I am saying that it has nothing to do with ensuring a seal, nothing. Nothing can stop hot combustion gases and pressures that have found a defect in the sealing surface or low clamping force.
Colin

User avatar
satchmo
Old School!
Location: Crosby, MN
Status: Offline

Post by satchmo » Wed Feb 21, 2007 11:52 pm

I worry that there could be a slight leak at the head sealing surface when I do that first engine start up after a rebuild, no matter how careful I was in making sure the surfaces were parallel or how accurate I torqued the heads. It seems to me that this is 'THE' crucial time to prevent leaks because the clamping force is at it's lowest at this time (clamping force increases as the engine expands, correct?), and any little leak now will expand to a full blown one later due to the constant force of hot expanding combustion gases.

I look at using the copper spray gasket material at the head sealing surface the same way I look at taking vitamins: I shouldn't need to take vitamins if I am eating correctly, but I take them anyway, just in case I don't. The head/cylinder sealing surface shouldn't need any help other than careful machining and the correct torque, but I use the copper stuff just for a little insurance. Sure, the stuff is probably gone after a few heat cycles, but maybe it helped a little as the heads and cylinders settled into each other and made the perfect seal.

This is all theory, of course, unless someone has actually done the experiment. I respect both Colin and Scott's point of view, both borne out of experience.

Tim
By three methods we may learn wisdom:
First, by reflection, which is noblest;
second, by immitation, which is easiest;
and third, by experience, which is bitterest. -Confucius

Gnasha
I'm New!
Status: Offline

Re: Type 4 Head Leak After 300 Miles?

Post by Gnasha » Sat Feb 29, 2020 10:30 am

I’m rebuilding a type for 2.0 engine. I have ground all cylinders to a machined deck. I have ground the top of each cylinders into their respective head. I have put a straight edge across no’s 3&4 cylinders and I can get a 0.002” feeler gauge under the straight edge at one end.

Is there a tolerance for this dimension?

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: Type 4 Head Leak After 300 Miles?

Post by Amskeptic » Thu Mar 12, 2020 9:40 am

Gnasha wrote:
Sat Feb 29, 2020 10:30 am
I’m rebuilding a type for 2.0 engine. I have ground all cylinders to a machined deck. I have ground the top of each cylinders into their respective head. I have put a straight edge across no’s 3&4 cylinders and I can get a 0.002” feeler gauge under the straight edge at one end.

Is there a tolerance for this dimension?

I have ground all cylinders to a machined deck.

Could you explain this? Has the case been machined at the cylinder barrel seating surfaces?
We generally do not under any circumstance ever grind the cast iron cylinder contact surfaces ...
Update?
Colin
BobD - 78 Bus . . . 112,730 miles
Chloe - 70 bus . . . 217,593 miles
Naranja - 77 Westy . . . 142,970 miles
Pluck - 1973 Squareback . . . . . . 55,600 miles
Alexus - 91 Lexus LS400 . . . 96,675 miles

rwk
I'm New!
Status: Offline

Re: Type 4 Head Leak After 300 Miles?

Post by rwk » Sat Mar 14, 2020 6:52 am

It is possible IF the case and heads have been cut, that the cuts are not square, in other words the cylinders are tilted, just because it's been "machined" doesn't mean its correct, even a brand new, say Bridgeport milling machine has to be squared up( by machinist) so that it cuts square, the mill has 2 tilting features they have to be perfectly square with table, the straight edge trick works ok but is not close enough, (doesn't spot tilt very good) and you have to check the heads also, if the case cut is just slightly out one way, (mill out of square) and heads are cut the opposite way, slightly the other direction, you have a compound error, I have checked and recut lots of heads and case's most that have been cut previously are out one way or another. Sorry no T4 picks, don't do many of them! Lots of garage machinist out there, some are perfect others not so much! I'm anal about it, you MUST know its correct before you build, causes all kinds of problems, deck height problems,head and case leaking, and piston scuffing in extreme case's.
IMG_3803.JPG
IMG_3803.JPG (433.52 KiB) Viewed 7172 times
IMG_3804.JPG
IMG_3804.JPG (587.38 KiB) Viewed 7172 times

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: Type 4 Head Leak After 300 Miles?

Post by Amskeptic » Sat Mar 14, 2020 5:14 pm

rwk wrote:
Sat Mar 14, 2020 6:52 am
It is possible IF the case and heads have been cut, that the cuts are not square, in other words the cylinders are tilted, just because it's been "machined" doesn't mean its correct, even a brand new, say Bridgeport milling machine has to be squared up( by machinist) so that it cuts square, the mill has 2 tilting features they have to be perfectly square with table, the straight edge trick works ok but is not close enough, (doesn't spot tilt very good) and you have to check the heads also, if the case cut is just slightly out one way, (mill out of square) and heads are cut the opposite way, slightly the other direction, you have a compound error, I have checked and recut lots of heads and case's most that have been cut previously are out one way or another. Sorry no T4 picks, don't do many of them! Lots of garage machinist out there, some are perfect others not so much! I'm anal about it, you MUST know its correct before you build, causes all kinds of problems, deck height problems,head and case leaking, and piston scuffing in extreme case's.

I noted with my latest deck height imbroglio, that the machinist's edge was square with the adjacent cylinder, the mismatch had perfect parallel and the case looked untouched. So, I blamed the crank grinding process for giving connecting rod #2 a .025mm smaller stroke than #1!
Colin I dunno really

Image
BobD - 78 Bus . . . 112,730 miles
Chloe - 70 bus . . . 217,593 miles
Naranja - 77 Westy . . . 142,970 miles
Pluck - 1973 Squareback . . . . . . 55,600 miles
Alexus - 91 Lexus LS400 . . . 96,675 miles

rwk
I'm New!
Status: Offline

Re: Type 4 Head Leak After 300 Miles?

Post by rwk » Sat Mar 14, 2020 9:38 pm

So barrel 1 is .01 lower then 2?

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: Type 4 Head Leak After 300 Miles?

Post by Amskeptic » Tue Mar 17, 2020 2:10 pm

rwk wrote:
Sat Mar 14, 2020 9:38 pm
So barrel 1 is .01 lower then 2?

. . . used to be. Added a .010 base shim. Said to heck with deck height. Ended up with exactly the same cranking compression as #2 at 135 miles. Go figya ...
Colin
BobD - 78 Bus . . . 112,730 miles
Chloe - 70 bus . . . 217,593 miles
Naranja - 77 Westy . . . 142,970 miles
Pluck - 1973 Squareback . . . . . . 55,600 miles
Alexus - 91 Lexus LS400 . . . 96,675 miles

Post Reply