Mercedes-Benz Forum banner

Cracked Cylinder Head Repair/Replace

37K views 63 replies 10 participants last post by  winmutt 
#1 ·
Today I found out I have two cracks in my 602 aluminum cylinder head, from my '91 300D 2.5TD. The car is in good shape, ran great until the cracks, and I want to continue with the repair.

I'm looking for experience and suggestions on either repairing the cracks or replacing the head. I would prefer to weld the cracks, at a recommended shop. I am in central Illinois.

Thanks! Wade.
 
#28 ·
There is no reason to put anything into the cavity around the injector, you're only making work for yourself.

Also, what casting revision is the head? The 602 heads had the same cracking issues between the combustion chamber and the water jacket as the 603 heads, right where yours cracked, and I'm interested in which head casting yours is.
 
#30 ·
Amost there

Getting close. She'll crank tonight after a block/heater cooling passages flush. Don't want flotsam in new radiator.

Now if I could only find where the brown vacuum line hooks. I have pictures of everything except that. I looked online for vacuum systems in the FSM, but found none. Is it hidden somewhere?
 

Attachments

#33 · (Edited)
The engine is running, and running well, tight and dry. Cranks smoothly, and fires smoothly. Took 7 or 8 cranking periods over a couple of hours (battery charging) to fire. When it first fired, it was normal noise, now it seems to idle and run loud, but smooth, which I read is not abnormal until some carbon deposits build up.

I've got two problems I know of to troubleshoot:

1) #1 and # 5 fuel injectors are leaking. I will tighten the union nuts, but I don't think that's the leak source. It looks like it's coming from the injector body, at a joint. I had the injectors professionally cleaned and adjusted though.
- The Oil Leak return nipples; are they identical? Does the tube pattern matter? I couldn't see a difference.

2) I have no brake boost, and no key-vacuum shutoff. I've got good vacuum, thumb test, from the new pump. I doubled checked the routing, all looks right. No problems before the head repair.

Any ideas on those items?

Thank you.

I'll keep updates on the cylinder head repair, so far so good, look like Ram's Cylinder Head did a nice job.
 
#34 ·
There seems to be a rash of issues regarding leaking injectors lately, and always associated with recently having cracked them open. I suspect the two halves need to have their contact surfaces inspected and possibly filed or hand-lapped to ensure a tight seal.

You might have cracked your vacuum source pipe for the brake booster. I think that's also the source for the ignition switch shut-off valve. They're all getting pretty brittle with age nowadays.
 
#35 · (Edited)
I found the brake booster vacuum problem, one of the fittings on the large hoses was just loose, male into female fitting. Some blue tape made a firm junction, will replace. That brought back the engine shut-off as well. Now to track down the vacuum problem preventing the center vent from blowing.

Those injectors don't require any kind of break-in / seating period? Just a thought. I will talk to the shop that did the cleaning and adjustment tomorrow about inspecting and correcting the injector-halves mating surfaces.

Once I have those leaks fixed, I'm on to the really important things, like finding a new floor mat that doesn't bunch up near the peddle. And new headliner and interior where the adhesive is failing.
 
#37 ·
It only opens a crack when AC is on, I get just a trickle of air flowing from it. I had hoped that this new vacuum pump would solve the problem, as it worked a few years ago, but it does not. I really want to get it back in order, as I can use the additional cabin airflow.

The rear headrests will not retract with the panel switch, I wonder if they are related?

I suspect that a check valve in the glove box area (I think) may be faulty. I will pick up the MityVac and start troubleshooting, along with the ALDA pressure check. Finding the vacuum system schematic is the first step, I think Performance publishes one in their catalog.
 
#38 · (Edited)
back with problems - oil leak

I found an oil leak tonight from the passenger side rear corner of the engine, #5 cylinder, at the head/block/gasket mating surface. A small area but leaking oil nonetheless. This is exactly the spot I had a slow leak of coolant prior to cracking the head, but this is a different story. With higher RPM, the area builds up oil, breaks surface tension, then drips down the engine block; a fair stream.

What comes to mind is another crack, or a new crack. I followed the service manual to the letter in putting it back together, and was quite particular in surface preparation. I was very gentle in breaking the engine in after the head work, and noticed very small oil traces on the engine after about 100 miles, but could not find source. Now about 250 miles.

Some thoughts:

- pull the cam cover, cam shaft, slacken head bolts in reverse order, re-torque in order. They were new bolts, into clean holes, with lightly oiled threads. I followed the manual explicitly in torque procedure though.

- pull the head and inspect; the head crack repair is under warranty, but I wonder if this is a head problem.

- Replace head if necessary, find one a shop can rebuild for me.

Any ideas of a better approach? Could I have a block problem I did not see?

I just found and repaired all of my vacuum system problems, Turbo, Climate Control, EGR, Throttle valve, all staging very nicely, engine is running extremely well, and shifting quite smoothly. Hopefully I can track this down.

Thanks!
 

Attachments

#39 ·
+1 on checking head bolts, but don't slacken them.

Just check the bolts the way they are, and adjust as necessary. Whatever you do, don't overtighten them. That won't fix the leak and you'll be entering a world of pain.

You may have a head that has microfissures, which could be attributed to your prior use of the coolant sealer additive.

If the retightening of the bolts does not stop the leak, plan on another head removal and possible head replacement.

Sorry to be the bearer of bad news. This really blows (not just oil).
 
#40 ·
OK on #1, thanks sbaert. I won't slacken them first.

These are stretch bolts, only my second experience with them. First was Ford 302 V-8 two years ago, success, leak free.

I followed the tightening pattern with the 4 step process- two specified torque values, 90 degrees angle torque, wait 10 minutes, final 90 degree angle torque. My local shop has a mech w/ much head cylinder stretch bolt experience, and he said sometimes a bolt won't stretch properly, that my leak may be at the oil return port with oil building up to higher levels at higher RPM.

He recommended getting an average torque value with old style wrench, at the head center, read just at point of turning, then taking the bolts near that leak up to that average value. If any bolts turn a significant amount before reaching that value, it had not reached it's correct stretch. Sounds like a reasonable procedure.

I'm not comfortable just checking the suspect bolts without a reference value, because of the angle torquing procedure. I have no experience, no reference tactile feel for stretch bolts.

Any better away to approach checking these torques?

I looked for, and will again, anything obvious, like getting the corner of an electrical connector or wire routed under the gasket, no sign of that. Both surfaces were clean, well prepared, with clean gasket laying nicely. Head met the matching alignment protrusions (little alignment pins) well.

I'll take your warning on over-tightening sbaert, I want to avoid that like the plague.

Wouldn't a shop pressure test have revealed a possible additional microfissure sbaert?
 
#41 ·
If the head was suffering from microfissures, then yes, the machine shop should have spotted it. But human error does tend happen more frequently than mechanical error.

As far as tightening the bolts go, I highly recommend using the clicking type of torque wrench on stretch bolts. I've never had the best of luck with the old style torque wrench on these type of bolts.

Just use a good wrench, preferably a Hazet or nothing less than a Snap-On. Don't step over dollars to pick up pennies on something this critical by going to Harbor Freight.
 
#64 ·
Just use a good wrench, preferably a Hazet or nothing less than a Snap-On. Don't step over dollars to pick up pennies on something this critical by going to Harbor Freight.
Except the HF clickers performs just as well. I used the digital and HF clicker when I did my head.
 
#43 · (Edited)
I spoke to the cylinder head shop, they said a crack from an oil gallery is possible, but unlikely, as he has seen an extremely small amount of them. He said they did test for them, but as you mentioned, human error is possible, and they will warranty the work.

He said most likely a new head gasket will be in order, as the pressure passage oil feed into head, most likely, is in this area, (not return as I guessed earlier) and that pressure passage seal area on the gasket is not sealing properly. He said it may be caused by the gasket sitting on a shelf for years, though this one looked fresh to me. (Benz head gasket kit with hologram). There was a big orange round circle of material in this area on the new gasket, I though it was for reinforcing against a coolant leak weak area on early design, but perhaps that is for the pressure oil feed?

He also said to check the block surface with straight-edge and feeler gauge, which I did not do. Since I previously had a coolant leak (not after the rebuild) from the same point, maybe I do have a low spot in the block mating surface at the area.

Sbaert, since these are angle torqued, how do I judge a torque to take them too on re-tightening? That was the idea on testing the center bolts with old style torque wrench, to get a reference value to apply to the other angle-torqued bolts. I'm scared to tighten them further without an idea of how far to take it.

I bought two brand new Craftsman clicking torque wrenches for the job, they are nice, I love working with them. Every bolt/nut/fastener for the whole job was torqued to spec if I could find it, I'm a big believer in spec torques and sequences for all jobs, if listed.

Inspector, the shop said they pressure tested the head underwater, similar to finding a hole in a tire. Submersed, passages fill with high pressure air, look for bubbles.

I'll take any hints, clues, suggestions you have on the best steps forward, especially with respect to determining how to best re-torque these angle-tightened head bolts.

If the head comes off, I'll use new bolts, new gasket, and have my local machine shop check for warping on the block mating surface and of course the cylinder head.

On a good note, if I do have to pull the cylinder head, I know how to do the job, the engine and compartment are clean, and it's all stand up work.

Thanks!!
 
#44 ·
A pressure test will not find microfissures, crack testing should. You need to make sure the mating surfaces are flat, you wont need a feeler gage as much as you will need your eye to spot the light from under the straight edge.

Follow the MB procedure for the bolts, torqueing is a huge subject and one that is often not carried out correctly...........
 
#45 · (Edited)
+1 on having the head crack tested.

Always use new head bolts & washers when doing a headgasket using torque-to-yield bolts. No exceptions. Using old bolts will not give you accurate torque readings.

Do test the engine block for flatness, there should no be no gap anywhere where a feeler gauge can pass through, although the head is far more likely to warp than the block.

IF you do have to pull the head, make sure the machine shop checks the head for flatness/straightness and crack tested. Since they're doing this under warranty, have them toss every single test at that head.

Did you have the head skimmed on the first go-around??? These long heads usually always need a skim, especially the gassers like the M104, but diesels are not exempt either even the 5 cylinders.
 
#46 ·
I studied that torquing procedure in the manual extensively, it matched the gasket kit instructions, and I followed them explicitly. I know that's a critical procedure. Detailed surface and bolt hole preparation for cleanliness. I did not take the time to check the block for flatness, that's a step I should have taken.

How does a crack (microfissure) test differ from a pressure test?
 
#47 ·
Yes, head was skimmed sbaert. Shop owner said it went out flat.

Roger on new bolts always on the torque to yield (no washers on my old or new Benz headbolts, just the bolt flange), and all tests from the shop for warranty work.

For my first step, re-tightening without slacking, since these are torque to yield, is there a better way than the idea of getting an average torque reading from several center bolts, then applying that to the bolts in the leaking area?

Thanks for the input sbaert, Inspector.
 
#48 ·
Pressure tests looks for leaks under pressure. Crack tests look for cracks and other flaws that may lead to leaks, a pressure test will not find surface imperfections at all unless they leak, a crack test or Dye Pen test on the Aluminium based head?? and Mag Particle test on the Iron block should, if it is performed correctly.

Don't get hung up on the bolts, if you do as MB recommend that will be fine.

A note on the use of feeler gages, they are wide, a flaw only has to be very fine to cause a leak. This is where an eye-crometer is great. Look for the light under the straight edge, very fine pin points of light too.....
 
#49 · (Edited)
Pressure test is just that, a test to make sure pressure holds.

A crack test is usually done using ultrasound or other imaging technology to check for the smallest cracks in the metal and/or metal fatigue.

When you tightened the bolts, did you start from the center of the head outwards using a criss-cross pattern???
 
#51 · (Edited)
A machine shop can bring back flatness to the block mating surface?
Yes, they can but in the very rare case this should be needed, engine removal will probably be required.

Never known them to be able to do this with the engine installed as they need the engine to go onto a rotisserie for complete unrestricted access. You don't want to go there if you can help it.

A warped block would be unheard of in the MB world, unless we're talking Chrysler-MB junk especially in the early 2000s with the V12s.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top