Second porous valve cover - known issue or bad luck? [Archive] (2024)

Lardog888

9th February 2015, 09:12

Last year my son and I bought our first NA, a '93 white with 124k. Looked a long time for an unmolested car, and finally found one with all original body panels and paint, three owners, and thoughtful modifications. We knew it had some oil leaks, but this was a project car, after all.

During the baselining, we cleaned up the engine compartment, and started diagnosing the oil leaks. After replacing the CAS o-ring and valve cover, we found oil continued to leak from the rear upper area and down the bell housing. Turns out that oil was leaking from a porous spot at the rear of the valve cover on the intake cam side, on the CAS hump right near the coil mounting bolt. While it was running, we could clean the spot, and watch the oil seep out of the surface of the valve cover; it's in the direction of rotation of the intake cam, so the cam is flinging oil in the direction of the seeping spot (if that makes sense). This was three days before we were supposed to drive to MRLS, so I cleaned with acetone and JB welded right over the top of it. That worked for several months, but eventually oils amazing penetrating capability prevailed, and a very slow leak from the same area has developed. We finally found a 1.6 at a Pick your Part, and collected a new valve cover. We cleaned and painted it, but before we could install it the car was hit while parked and totaled.

While arguing with the other parties insurance to value our car, we found a super clean '92 Sunburst car and purchased it (we also bought our '93 back). To make a long story longer, during the baselining of the the '92, we found exactly the same oil leak from the same spot on the valve cover. What are the odds of that? We've installed the valve cover we got from the salvage yard for now since we have an HPDE event in two weeks, but we painted it red for the '93 and it looks funny on the yellow car.

I've searched here as well as Google for porous valve cover, but haven't found any trends. Am I missing something? Has anyone else experienced this?

Needless to say, I'm in search of another valve cover, and see many on ebay, but I'm worried about buying a used valve cover, only to inherit the same spot of porosity.

One other thing that I noticed when changing the valve covers is that both cars are seemingly missing a bolt for the coil bracket at the back of the head. Is it common to leave this out? There's a rubber grommet there, so vibration will likely transfer anyway, but maybe that caused micro cracks in the valve cover? Anyone know what size bolt is supposed to go there? M8 like the two bolts on top?

Sorry for the long post. Just frustrated having purchased two cars with exactly the same leak and it appears that nobody else has this issue, or at least it's not common enough to trend.

Thanks for any further information that anyone can provide.

3MiataFamily

9th February 2015, 09:22

Lardog888

9th February 2015, 09:34

3MF, I know from lurking here on this forum that you have a tremendous amount of experience. What can you tell me about the (seemingly) missing bolt from the bottom of the coil bracket?

Necessary, or left out because it's difficult to remove and replace?

GasolineCologne

9th February 2015, 09:48

3MF, I know from lurking here on this forum that you have a tremendous amount of experience. What can you tell me about the (seemingly) missing bolt from the bottom of the coil bracket?

Necessary, or left out because it's difficult to remove and replace?

Left out because difficult to replace. Designed to reduce vibrations and prevent bracket or coil failure. May or may not be necessary.

Never heard of the leaking valve cover you are describing.

Kevin M.

9th February 2015, 09:51

A lot of people leave that bolt off because, as you said, it's a pain to remove. A better solution is to slot the bracket so the coil pack can be removed by just loosening the bolt.

As for your leaking valve covers, I've never run across it either. It's not a common issue.

huesmann

9th February 2015, 09:52

Schwanz

9th February 2015, 09:58

Lardog888

9th February 2015, 10:25

Huesmann - I only JB welded from the outside. I had just replaced the valve cover and siliconed the 6 spots, so didn;t want to remove again and risk a leak from the gasket itself. I roughed it up with emery cloth, and JB welded the surface.

Schwanz - I should have taken a video. It's pretty interesting to watch; oil just appears on the surface of the valve cover and starts running down the hump. There is no visible defect on either side in the spot that is leaking. I looked with a magnifying glass under strong light, and see nothing unusual.

3MiataFamily

9th February 2015, 10:33

As already stated..that 3rd bolt is a royal PITA..and is frequently left out..

Pic of the leaking valve cover would be interesting....this is a new one for all of us

Kachow

9th February 2015, 10:39

MC70

9th February 2015, 11:50

My guess is that you have a microfracture, probably due to improper torqueing sequence and torque values. I have seen this happen in other parts, and it is nearly impossible to detect when the assembly is un-mounted, but shows up when installed.

The idea that you could very quickly observe this tells me it is a crack and not a porous area, even the worst Voit-made Harley Davidson engines, which were renown for this, would take much longer to well oil, and in much smaller quantities than you are describing.

Lardog888

9th February 2015, 13:19

MC70 - I guess it's bad luck on my part then. I torqued both of these valve covers with a Snap-on Techangle FR240B in three passes (contact, 30 in/lbs, 60 in/lbs) with the sequence shown in the garage section (from the center bolt in an outward spiral CCW).

Who knows how many times the valve cover has been off in 117k on the '92 and 124k on the '93.

I hope the next valve cover I find hasn't been ham-fisted.

Thanks to all for the info.

2trains2

9th February 2015, 13:43

pacomutt

9th February 2015, 13:56

Lardog888

9th February 2015, 14:55

Pacomutt - I did not remove the bearing cap. I simply slid the CAS out and replaced the o-ring. What MC70 said about a microfracture makes sense, although I would think that over time, oil would have stained the crack to the point where we could see it. These valve covers stain so easily anyway.

Unless, of course, the crack just happened, but I doubt it. I was extremely careful about installation and torque.

Not sure what I'm going to do. I suppose trying to seal it with JB weld from the inside is as good a place to start as any. I am going to watch the local Pick your Part for a 1.6 this week. Maybe I'll get lucky.

pacomutt

9th February 2015, 15:06

Pacomutt - I did not remove the bearing cap. I simply slid the CAS out and replaced the o-ring. What MC70 said about a microfracture makes sense, although I would think that over time, oil would have stained the crack to the point where we could see it. These valve covers stain so easily anyway.

Unless, of course, the crack just happened, but I doubt it. I was extremely careful about installation and torque.

Not sure what I'm going to do. I suppose trying to seal it with JB weld from the inside is as good a place to start as any. I am going to watch the local Pick your Part for a 1.6 this week. Maybe I'll get lucky.

Hope so. I won't even remove the valve cover until the engine is at room temperature, and then I loosen the bolts just a little at a time in a backwards sequence. When reinstalling, I probably made 8 or 9 passes in sequence at tightening. I can't imagine two separate covers having the same porosity in the same exact location! Buy a Powerball ticket, and then buy a NEW car with the winnings! (And buy me one too!) :D

GasolineCologne

9th February 2015, 15:26

MC70 - I guess it's bad luck on my part then. I torqued both of these valve covers with a Snap-on Techangle FR240B in three passes (contact, 30 in/lbs, 60 in/lbs) with the sequence shown in the garage section (from the center bolt in an outward spiral CCW).

Hope so. I won't even remove the valve cover until the engine is at room temperature, and then I loosen the bolts just a little at a time in a backwards sequence. When reinstalling, I probably made 8 or 9 passes in sequence at tightening.

:eek:

I have tightened like a thousand valve covers with nothing more than a 1/4" driver handle and socket, without issues. I do use a criss-cross pattern, but have never really done more than two passes.

I suspect your level of precision has angered some higher power and not only caused a crack in the valve cover, but also in the space-time continuum.

gtxhawaii

9th February 2015, 19:40

huesmann

10th February 2015, 10:10

Laz

10th February 2015, 10:24

pacomutt

10th February 2015, 14:18

9 passes? Wow. I have done mine a few times and only two passes with an OEM gasket. You can feel when it's right as well. Good luck.

I don't need luck, and I'm wondering why you've "done yours a few times" in only 105K miles. Perhaps your abilities are such that you shouldn't be questioning others, except for maybe soliciting advice on how to do things right, rather than just repeating the same job several times.

gtxhawaii

10th February 2015, 15:44

pacomutt

10th February 2015, 15:54

gtxhawaii

10th February 2015, 17:25

I know what I've done and what of that worked or didn't. There are no Ultimate Experts here, we are sharing experiences. Torquing a part's bolts down 9 separate times is something I've never had related to me. What isn't being understood??
It directly contradicts the 'best practices'* of people who design and build both mechanical fasteners and set assembly standards for the US automotive industry. YMMV

*ONE smooth pull to final torque on any critical fastener. This implies you Have to see movement during the final torque set. Pattern staging for multi-fastener parts, the designers will specify patterns on most if it matters.
It's specifically recommended any torque set to be backed off and repeated if there is any question. Repeating by 'check' torque's or very small staged additions is said to be a misuse of the common torque wrench and it's relation to 'break free' torques of most threaded fasteners. It requires over torque to move a correctly set bolt in any amount. It can require more torque to break a fastener free than the gap between too small of torque differences in staged torques.
If the bolts aren't moving at set torque on the tool when I get a 'click', I'm through or I back the fastener and set again when doubt exists. If I am moving bolts each 'check', something is wrong, possibly the wrench is set wrong and I'm stretching bolts or metal contact areas aren't clean or free of obstructions. I'm inspecting everything before proceeding.
So far, following the Mazda FSM has been quite productive. Rereading it has helped, memory isn't all that reliable.

pacomutt

10th February 2015, 17:57

gtxhawaii

10th February 2015, 19:02

Laz

10th February 2015, 22:11

I don't need luck, and I'm wondering why you've "done yours a few times" in only 105K miles. Perhaps your abilities are such that you shouldn't be questioning others, except for maybe soliciting advice on how to do things right, rather than just repeating the same job several times.

I had foolishly used the non OEM gasket. It failed after a few hundred miles. The first time was when I first got the car and baselined it. I wanted to see the shape of the gasket plus timing belt. Second time with non OEM gasket that was suggested to me, third time with the OEM gasket that has not failed. This was at around 80k miles. Guess what? It has not failed, and I have not broken anything just yet.

I don't assume how much technical abilities you have, but from experience with this car, and other cars that I have tooled on over the years going around 9 times to tighten it seems excessive. Not to mention that I have not read anywhere to do it 9 times.

But hey, what the hell do I know. You seem to have superior knowledge, but I'll keep stating what I have experienced and know. If you don't like it, don't read it. There is an ignore function on this forum just for those cases.

Btw I take my "good luck" comment back.

GasolineCologne

10th February 2015, 22:21

Laz

10th February 2015, 22:45

Valve cover bolts are serious business on m.net

:rofl::jump::ohno:

My bolts are made of titanium and I torqued them in a special vacuum chamber!

gtxhawaii

11th February 2015, 04:38

:rofl::jump::ohno:

My bolts are made of titanium and I torqued them in a special vacuum chamber!

Wait? Yours aren't Magnafluxed, cryogenic treated, ceramic coated and monogrammed? And you are admitting it??

gtxhawaii

11th February 2015, 04:45

Lardog888

11th February 2015, 09:34

Holy moly! I'd just like to say that I did not intend to start the great valve cover debate of 2015.

Different technicians have different methods and techniques. I suppose it depends on who originally taught you, as well as practical experience. I, for one, agree with gtx in that one smooth pull to final torque is critical. I also think that these particular bolts have a tendency to give false torque readings due to the length of the shank; there is definitely some torsional flex, even at relatively low torque values.

That said, the beauty of a modern digital torque wrench such as the Snap-on FR240B is that it displays the active torque in 0.1 in/lb resolution, as well as gives a visual representation of progress with a series of five LEDs that light in succession as the preset torque value is approached, as well as displays the maximum torque achieved when complete. The maximum torque is almost always slightly higher than the preset torque no matter how careful you are about bringing it up to the target. For example, I preset the torque at 60 in./lbs for these valve cover bolts, but the actual finished value was generally 61-63.

For many applications and materials the acceptable torque range can be quite large, such as the factory spec for these valve cover bolts at 43-78 in./lbs (5-9Nm). For other applications, it's critical. Many modern torque specifications are no longer limited to twisting force, but are reliant on carefully calculated clamping force using specific low scale torque plus angle of rotation. Sometimes in multiple angles of rotation (i.e. 8Nm +60 degrees + 60 degrees) with a terminal value that must fall within a resultant torque range.

I know that my valve covers did not crack with my removal and reassembly; they were already leaking before I took them apart, which is why they were coming apart in the first place. I just thought it was weird that of the two Miata's I've owned, both of them suffered leaks through the surface of the valve cover in exactly the same place. I found it difficult to believe, especially since I could find no evidence of trends to support this, which is why I asked the question in the first place.

Thank you to all that replied. This is a great forum with awesome amounts of information and experiences. The data that's stored in this website, as well as in all your brains is invaluable. Although I am new to Mazda and Miata's, I am no stranger to automotive technology.

Now, can we all just get along, and sometimes agree to disagree?

Cheers!

gtxhawaii

11th February 2015, 16:49

Thanks for letting me know about yet another tool I lust for I'll never afford!
No Torque-to-yield or torque-to-stretch bolts on Miatas, which is good, we can reuse anything for a reasonable number of times. Except SNC crank bolts!

My metal patch episode was on the outside of the fuel tank, about an inch below the waist seam. First try leaked. Second time I took the surface down to clean but rough sub-surface metal with a steel rotary brush in a drill motor. JB Weld, sat a day or more before use or filling the tank, still holding.
I'd repair a valve cover on the inside, same tool to prep the surface, masking the oil vapor catch labyrinths away from any metal bits or dust with multiple layers.

prb

11th February 2015, 16:54

iflypby

12th February 2015, 10:26

nerdycellist

13th February 2015, 11:57

The leaks in the valve train covers could possibly be sand casting flaws. Early Continental "A" series engines had this problem. Oil pressure was great at initial startup and fell as engine temp increased. At operating temp the oil galleries leaked like a sieve. I overhauled this engine three times in 10 hours before a real old timer told be about the problem. Solution? A new case. A six pack of beer and a sledgehammer to the offending case. I still remember the satisfaction :)

Gives a whole new definition to percussive maintenance.

This thread has definitely given me good pointers for when I replace my valve cover gasket and CAS o-ring soon. Thanks all for being so informative! :wave:

gtxhawaii

13th February 2015, 19:14

If you form a small shield of aluminum to the internal radius of the area to be patched and sandwich epoxy it to the leaking area after proper sanding and cleaning, it's likely to stay there. The unprotected glue isn't nearly as strong as a glued saddle.

A good point. Anyone have any real data on hot oil resistance of common epoxy repair products? I'd be inclined to think JB Weld is decent, given it has repaired so many internal and external engine parts for long periods.

Lardog888

13th February 2015, 21:05

I've been thinking about this myself. Even if I roughed up the area and cleaned the snot out of it with acetone, the thought that the chunk of JB weld has the potential to fall into the cylinder head, and consequently ground around in the valve train would always be in the back of my mind.

I don't think I could deal with that in the long term. I'm a little neurotic that way.

I'm either going to make an external repair attempt, or find a replacement. I'm too chicken to try an internal repair.

Dushan

14th February 2015, 14:19

MY90

14th February 2015, 15:31

I have experienced this leak from a crack in the cover, years ago. Very frustrating. After multiple removals, inspections, massive amounts of sealer, Viton "O" rings and the like it became obvious that it was a crack in the cover.
My deduction was that because the 3rd bolt was left off, like yours, the coil pack vibrated enough to fracture the cover over time. I replaced the cover and never experienced the leak again.

I should add that because of what I considered to be the cause of the crack, the coli pack hanging off of the rear tab, I repositioned the coils to the passenger side near the washer bottle because I could not bear to remove the 3rd bolt again. After a time I moved to COPs, which were more elegant and efficient.

Moral to the story if there is one; If you are OEM, use the 3rd bolt, or the cover becomes a consumable.

gtxhawaii

14th February 2015, 16:18