Note: Our experience did not include replacement of the valves and rebuilding of the heads. Bob Hoover has prepared a couple of excellent treatises on these subjects, ”New Heads” and “Head Job” (unfortunately these articles are no longer available on the Internet).
Here we will primarily discuss torquing of the head bolts.
Dave wrote to Rob -
There is an editorial in the latest VW Trends about the retorquing of cylinder heads. About six "experts" are quoted; half of them say that re torquing of the heads on a stock engine is not necessary, the others say it is. The editorial summarizes their survey with the following:
The common denominator here seems to be that, if the opportunity presents itself, it is a good idea to at least check the bottom, easy-to-get-to head nuts, and if you are running anything other than a stock 1600cc engine, then it should be done with more regularity. I'll go out on a limb here and suggest that maybe every other valve adjustment or two, take the rocker arms off and check the lower four. You were going to set all the valve clearances anyway, so what's four more nuts? I will also recommend that if you are running more horsepower through a cam, headwork, bigger cylinders and pistons, or more compression, it becomes a mandatory routine. If it stops one minor problem from becoming a bigger one, it would be well worth the extra time and effort.
We will definitely be re-torquing all of the head nuts as part of our work on Friday. I need to go out and get a new torque wrench; the one I have is very old, about two feet long, and only has a long metal pointer to roughly indicate the torque. There must be better, smaller wrenches on the market now.
Rob wrote sending the torquing sequence -
Make sure you do this carefully -- there are two tightening patterns, first to 7ft-lbs, and the second pattern to the full torque of 18 or 22 ft lbs (18 for 8mm studs, and 22 if your engine has 10mm studs).
First to 7 lbs -
top of engine
7 5 6 8
4 2 1 3
then to 18 lbs for the 8mm studs and 22 lbs for the 10mm studs -
top of engine
8 2 4 6
5 3 1 7
Dave related his experience -
I re-torqued all of the cylinder head nuts; all were a little loose.
Rob responded -
THAT was a profitable job then -- the usual result of prolonged loose head bolts is leaking combustion gases at the head/cylinder joint, which burns the heads out and usually the top of the cylinder too bin jobs for both. In this case, an once of prevention....
New heads in the US are about $120-140 each if I remember right, and even cheap cylinder/piston sets are about $100.
Dave wrote a year later -
When I get to the head bolts, I'm assuming that they are all already tighter than 7 ft-lbs, so I should tighten the bolts as follows in according with the sequence you sent me (see above).
Rob responded -
Yes, I agree -- go straight to the second tightening. The bolts are quite visible with the tinware off, so checking the size is easy. My guess is that they will be 8mm -- I think I saw in one of your earlier pics that you have case savers (threaded inserts), and these are used with all 8mm head bolts -- to increase the thread area in the soft case. VW changed to the thinner bolts because they expand more evenly with the engine as it heats, so the "hot" head torque doesn't get so high it pulls the bolts out of the case. Even so, Bob Hoover says the "hot" head torque is about 70 ft-lbs -- a lot more than the 18 cold torque! My guess is that the increasing horsepower of the more recent engines (1600tp mainly) meant more heat, and this became a problem with the 10mm bolts overtorquing at running temps.
Dave wrote regarding torquing of the head bolts -
No way out of removing the tinware I'm afraid. And the rocker arms must be removed, too.
I went directly to the second tightening, as we agreed, and I found that they were ALL loose. I tightened them all, in the proper sequence, to 22-23 ft-lbs (they were all the same size).
Rob responded -
The 8mm bolts are tightened to 18 ft-lbs. You need to measure the bolt diameter to see which torque to use. Since I think I remember your car has case savers, it's likely it will have the 8mm bolts.
Dave wrote -
I'm sure you're right. I'll check; if I need to drop back to 18 ft-lbs, should I loosen them all and then retighten them in sequence? Do I need to loosen them in sequence?
Rob responded -
Loosen them all a little (hopefully still more than 7ftlbs so you don't need to do the double tightening, and redo them to 18 (check they are 8mm first!)
I'd loosen and tighten in the second sequence, just to be sure.
Dave wrote -
I torqued the head bolts while I had the tin and intake manifold off. And I did it right -- they are all 10mm, so 23 ft-lbs is correct. They were ALL loose -- something like 18 ft lbs.
Rob responded -
Maybe your ham-fisted mechanic just knew the 18lb figure, and uses it on all bugs! Good to fix them anyway -- Bob Hoover and a few others talk about retorquing the heads every time you need the engine out, just as insurance. Since there is no head gasket to allow for any slack, I guess it's more important in a VW engine than most others.
Dave wrote -
There were little screws/bolts missing EVERYWHERE in the engine tin ...
Rob responded -
Isn't it amazing how they can run well with bits missing. But it feels really good to get it all buttoned up right doesn't it!
Someone wrote -
The vehicle was still pinging like crazy and now felt like it was running on three cylinders, so I took it to another VW specialist. They found the timing to be at 26 degrees advanced and the head gasket to of blown.
Rob responded -
The VW engine does not use a head gasket. The cylinders sit inside a recess cut into the cylinder head (this is called fly cutting the head) and a seal is made between the recess in the head and the top rim of the cylinder itself. If this area becomes damaged or overheated in any way, high temperature combustion gases leak out, causing burning of the head and cylinder top, requiring expensive repairs. The "running on three cylinders" you describe is typical of an engine with at least one damaged cylinder/head joint (assuming all cylinders have working spark plugs and the correct fuel/air mixture is being supplied).
Someone wrote -
After torquing down the heads on my '67 bug (cylinders are new and the heads new/rebuilt) I noticed something that didn't look right. The heads didn't seat all the way- I could still see space between the heads and the cylinders …
Rob responded -
Eh??? The heads are fly cut (recess for the heads to fit in), I don't understand how you could see a space between head and cylinder inside that recess.
Since there is no head gasket on VW engine, the mating between the heads and the cylinders is vital -- it's what seals the combustion chamber. ANY leaks there, even tiny ones, will result in burned cylinder tops and/or burned heads around the sealing lip. So don't try to start the engine until you are sure the heads are correctly seated.
The only way the head could be "cocked" so only one cylinder wasn't sealing properly is if the cylinder is not sealed in the case correctly (resulting in one cylinder being high), or the head is damaged.
Before the heads go on, you should be able to place a straight edge across the cylinder tops and see it touch the two cylinder rims in four places -- then you know the cylinders are fully seated in the case and are identical heights.
The person wrote again -
I pulled the heads and cylinders to see what was going on. The cylinders do indeed sit flush on that lip in the heads, although there is slight movement (I can wiggle the cylinder back and forth ever so slightly).
Rob responded -
That's normal. The fly-cut in the heads is fractionally larger than the cylinder rim, so there is room for the cylinder to move just a fraction inside the head.
The gap that I was referring to, (and the part that has me confused), is between the flat part of the head and lip on the cylinder.
Rob responded -
I still don't understand (sorry). "flat part of head" - do you mean the underside of the bottom fin (nearest the case) or are you talking about the inside of the fly cut recess?
The person wrote -
One of the cylinders sits flush on the raised 1/2 moon portion of the head.
Rob responded -
You mean the combustion chamber where the valves are?
The guy wrote -
But the other three don't quite touch the moon - they are all different sizes, the largest creating a gap .082 (measured from cylinder lip to flat part on head, in between moons).
Rob responded -
Hmmm - do you mean between the outermost cylinder fin and the innermost head fin? The distance between those might vary a little - the critical measurement is the length of the cylinder between the case and the outer rim. The case needs a completely flat area where the cylinders slot into the case, and them with even height cylinders the rims of adjacent cylinders will be a perfectly straight line. The recesses (fly-cut) in the heads are cut to an exact depth too, so when the heads go on both cylinder rims will sit straight inside their recess and seal properly when the head studs are torqued.
It's a complicated part of the VW engine design, an unusual design to have twined heads but separate cylinders (horizontally opposed aircraft engines use separate heads for each cylinder for example). That makes it essential that both cylinder have exactly the same height, or the head won't seal properly on both cylinders.
Someone wrote -
The motor is a 1600 dual port. It has 106,000 miles on it. The seller told me the top end was rebuilt. I don't know this for sure.
Rob responded -
Ok - so if the heads have been done, then suspicion for low compression lies with the rings/cylinders (especially if they have never been repalced before - 106,000 miles is a good life for cylinders/pistons), but DON'T discount the heads - maybe they need retorquing, or maybe the valve need adjusting, or maybe it was a poor head replacement - still plenty of possibilities.
Head retorqueing is important too - new heads "settle" somtimes and the studs need checking after a few thousand miles. When I bought my last beetle (only the second one I've bought in 33 years! - Bradley was bought to keep me sane after Bertie got squashed by a nasty Ford - he's being very slowly rebuilt), it had "new" heads and when I checked the studs, 5 of the 8 on one side were almost loose (5-10 ftlbs).
You can check the lower bolts with the engine still in the car (remove the rocker covers and remove the rocker arms to get at them) and if ANY are loose you then know you need to drop the engine to check the rest.
If your engine has 10mm studs use 22 ftlbs, if it has 8mm studs, use 18 ftlbs. You can measure the dia of the studs at the nuts. I like a "bending beam and pointer" torque wrench myself, as you can hold it at the required torque and see if the studs turn a fraction - with the clicker type, once it's clicked you can't be sure you are still holding the right torque, so you have to redo it on each stud 2-3 times to make sure. But if you have the clicker type, use it - retorquing VW heads should be done any time the engine is out of the car - just good insurance. And always after any head work at all (valve replacement or whatever).
Please see our procedure regarding pushrod tubes.
Rob explained how the pushrod tubes work -
The VW engine has four separate cylinders and and two cylinder heads, and therefore does not have internal openings for the pushrods to sit in, as happens in an "inline" watercooled engine with a full engine block. Therefore the pushrods have to have tubes around them. These tubes provide the return path for the oil flowing back from the valves to the sump, and even provide some cooling for the returning oil which has been heated by contact with the very hot heads, and then flowing past the cooling airstream for the cooler running cylinders. A very clever piece of engineering by Mr Porsche.
The tricky bit is that since metal expands when it gets hot, the VW engine grows and tiny bit wider each time it runs (and a fraction longer too, but the engine case is not the problem here). So the pushrod tubes have to be able to expand and contract a mm or two, and at the same time must seal against both the heads and the case to provide a closed oil path back to the sump. This is why they have a concertina section at each end to provide the "spring" needed to seal the neoprene washers at each end (to the head and case), and to allow for the engine expansion with heat.
BUT -- the neoprene washers do gradually harden with age (takes a long time), and the concertina sections do loose a little of there "spring". So leaks can develop here. And damage/bends to the tubes can make this worse. And they can rust from the outside, resulting in pin-holes.
To check for leaks, get under the car after it has been on a run, and look very closely at the pushrod tubes themselves just under each cylinder. If they look fresh wet with oil, or there is a mess of oily deposits on the cooling tinware under the pushrods, this may be (partly at least) your problem. Be careful -- don't burn your nose on the hot exhaust/engine when doing this!!!
If leaky/damaged pushrod tubes were your only problem, there is a replacement tube which can be fitted without removing the engine I haven't ever used these so I can't describe the process exactly. But since you may have to install new cylinders/pistons anyway, the original equipment type of tubes can be used. The old tubes will be fine if they are not damaged/rusted (as Dave says, they must be straight), but you MUST replace all the neoprene washers (two for each tube). The "old boys" trick is to gently stretch the tubes a little (measure the length of each tube, then pull outwards gently on the concertina sections till it's a few mm longer) before reinstalling. This provides renewed "spring" to the tubes and ensures a good seal on the new washers.
Someone wrote -
The valve covers ARE leaking …
Rob responded -
You have to look closely at the engine/head area itself -- not just the puddle on the ground. The pushrod tubes and the valves covers are very close to each other and the puddle would be in the same area.
Bob Hoover has written an interesting “sermon” on pushrod tubes (unfortunately this article is no longer available on the Internet). Just a note or two from the article -
The pushrod tubes form an oil-tight conduit between the valve gallery and the crankcase, allowing multiple return paths for the oil pumped out to the rockers through the push-rods. At overhaul time it's best to use new push-rod tubes since the bellows portion on either makes them especially difficult to clean. They cost about a buck each, cheap insurance
The stock tubes are wizards at getting rid of heat! Given a thin coat of flat black paint to preserve them and lower their thermal resistance, your eight push-rod tubes serve as eight auxiliary oil coolers
Since the push-rod tubes are bathed in the flow of air that has just passed over the cylinders, the air is hot. But not as hot as the oil coming from the valve gallery. The cylinder head is the second-hottest part of your engine (the exhaust valves and stacks are first) and the oil in the valve galleries is typically a hundred degrees or more above the oil temperature in the sump. But that short trip down the push- rod tubes is sufficient to suck a lot of the heat away, thanks to the slightly cooler air coming off the cylinders and the generous surface area of the push-rod tubes.
If you replace your stock push-rod tubes with those trick two- part anodized aluminum jobbies so beloved of show-car freaks, you've just thrown away one of the more subtle gems of the VW engine design.
Rob wrote -
The concertina ends of the pushrod tubes very slowly go soft with age, and the seal at each end can leak eventually. Just something to keep an eye on.
Dave wrote regarding someone’s new pushrod tubes -
I'm intrigued with the design of the new ones he installed two piece, with compressible spring in the middle.
Rob responded -
Yes, I was intrigued too -- obviously designed so you remove the rocker arm, pull out the pushrod, yank out the old tubes sideways (probably where most of the gross damage occurred as they are supposed to be removed after the heads are off, put the long section of new tube in place, then compress the spring section on the short piece so it can be inserted and the spring and thin seal seals the two pieces together, whilst pushing the whole assembly in both directions to seal the ends against head and case. Then put the pushrods back in and replace the rocker arm.
Bob Hoover warns (in his article on push-rod tubes and cooling -- no longer available on the Internet, unfortunately) -) -
Whatever you do, don't even think of using two-part push-rod tubes. They will make your engine run hot.
Someone wrote -
The engine will be pulled out by another person, I suggested to change all the pushrod tubes to stop the leaks. This mechanic said something I've been thinking about and I don't understand. He said since carbon builds up in the heads, If I pull them out to change these tubes, this carbon will fall away, so that "natural" seal will disappear and my engine will start passing oil, or burning it.
Rob responded -
The seal is actually the top of the cylinder against the recessed "fly cut" in the cylinder head. Provided the heads and cylinders are accurately cut, and the head studs are torqued correctly, this makes a good seal. It's possible (but not very likely) that removing the heads might disturb that seal.
So long as the cylinders are not removed, you would not disturb the piston ring/cylinder bedding though.
If you are worried about it, you can get pushrod tubes that don't need the heads off to replace - they telescope into position. But there are at least 2 sorts of these, and at least one sort leaks badly (I've never had to use them myself). Ask John at www.aircooled.net about the best ones - if you decide to used them. If you want to use the original type, you just stretch the old tubes a few mm and use new end-seals, and place the seams at the top when re-assembling.
Another question -
I think I might pull the heads to install those original pushrods. So if I put attention in the installation, I mean, clean well the head-cylinder interface before re-assembly to improve the seal action of these parts, I shouldn't have trouble...?
Rob responded -
That's correct. You need new end-seals for the pushrod tubes, and you MUST stretch them a little before you put them back in, so that they push the new seals into the head and case firmly. Measure each one and stretch them only a few mm. If you do it too much they might crack. And make sure that the weld is pointed upwards - just in case it ever starts to leak (unlikely, but....)
Someone wrote -
The rod sleeves seem to buckle a little when I replaced the heads. I followed the instructions word for word ("How to rebuild your Volkswagon by Tom Wilson"), so I'm not sure what I did wrong-or how to fix it.
Rob responded -
That book is quite good so I'm puzzled by your description too.
The pushrod tubes should be stretched a few mm at the concertina sections so that when the heads are tightened, they compress back down just a little to make a good seal with the rubber gaskets. Do you have the gaskets on the right way round, with the tapered side pointing to the heads and case at each end?
These normally fit very easily, and the tubes themselves won't buckle, just the concertina ends compress a few mm.
The person wrote again -
As for as the rod sleeves buckling, well, somehow I got wrong rubber gaskets in my set- they were a full 1/8th in. taller then the ones I should've put on.........go figure??
Rob responded -
Very strange - a different sized gasket? Type 4 engine gasket perhaps - I wasn't aware that there was more than one size of pushrod gasket.
Disclaimer stuff: Rob and Dave have prepared this information from their own experiences. We have not assumed any specialised mechanical knowledge, but we DO assume that anyone using this information has at least some basic mechanical ability.
We hope you find this information useful, but we don't take any responsibility for anything which happens to you, other people, your VW or any other property or goods resulting from your use of this material.
Feel free to print off any of this information for your own use. If you intend to link this material to another site, reprint it, or in any other way redistribute it, please leave the information complete, including this disclaimer section, and provide a link to this Web site.
Last revised 5 May 2004.
