[maxud]

Did you clean the valves before you took picture of that head? They look awfully clean, which could mean coolant was getting into the combustion chamber.
How was coolant level before you tore it apart?
Leakdown test is a percentage of blow by, how much air pressure are you putting into the cylinder, while you get your numbers?

 

[maxud]

thinking about it more, you did the most basic test you needed to do already, compression.
IF you did the test correctly, those numbers are not enough to create proper combustion, so that leads to the question, where is it loosing compression.
Did you inspect valve seats? Cylinder bores? Anything jumped at you.
I agree that having all 3 cylinders fail at once is a bit of a stretch. I have seen burned up valves and bent valves due to timing belt breakage. You say you have none of those issues.

If I had to hypothesize, it almost sounds like your head is warped, which is how coolant may have gotten into the cylinders in the first place. Not letting machine shop
look over the head was a major mistake. They will ensure your valves are sealing and head is flat within spec.

At this point in time, see if you can hear air in the coolant passages, which is my guess where its going.

 

[maxud]

Any ideas on root cause that would cause all 3 cylinders to suddenly failed?

If you are not going to reply to questions that were asked, I think you will find yourself talking to an empty room

Stuck rings are common issue with older engines. I don't think its the smoking gun here, but definitely needs to be addressed.
How are you tearing this engine apart? Is it still in the car? Is the front head off?
Debris from a catalytic converter had been known to wear out engines (hyundai), but I imagine you would have p0430 code with
your front cat looking like that.

 

[maxud]

2007 EXL is VCM1 which disables cylinders 1, 2, & 3. Therefore cylinders 4, 5, & 6 are being overloaded when VCM is on.

You do realize that all pistons move at all times, right?
Nothing is getting overloaded in front bank, the only difference is there is always combustion in the front bank and no combustion in the rear bank when vcm is active.
So in fact its exactly the opposite, the rear bank suffers the VCM curse.
The front bank is working like the normal engine, not burning oil, nor stressing the piston rings.

 

[maxud]

No need to check for warped head, I am only looking for primary issue, I don't need optimal performance. As I said previously, the leakdown test showed that air was coming out of the dip stick; meaning majority of the air is passing over the piston cylinders.

hmmm, somehow, I got an idea that you were looking for answers, but apparently you already got them all, good luck.

 

[maxud]

I just found a good example of front head from another thread

Now compare it to the image you posted


Besides the burned exhaust #6 valve, do you see a difference in these pics?
I asked if you cleaned the valves in your pic and you ignored me.

Your pistons and valves are getting steam cleaned.

 

[maxud]

That is interesting that my cylinder head looks very clean compared to this head. I did not clean anything. Oh, I replaced this cylinder head 2/3 years ago, so it is relatively new; I am sure that partly explains the cleanliness difference.

ok, that is a major piece of info that may change things. Obviously the head I posted is what a used head should look like.
Combustion creates carbon deposits.
Yours head looks steam cleaned, it's what happens when coolant gets into combustion chamber.
My hypothesis was that that either your headgasket failed or head warped enough to get coolant into combustion chambers.
It does not take much, and it would certainly explain the issue as it is effecting the entire bank.

To continue, a warped head will allow air to escape to oil and coolant passages, as well as adjacent cylinders.

Was the replaced head a brand new head from Honda or a junk yard head? Why was the head replaced in the first place?

Another possibility is that the event you experienced on highway is somehow related to destruction of the catalyst.

I just don't see how 3 pistons failed at once with ordinary driving while the coolant and oil level were ok.

 

[maxud]

Your understanding is mostly incorrect.
The amount of gasoline injected is measured relatively to the amount of air ingested. Look up stoichiometric combustion.
In naturally aspirated engine, like J series, there is no forced induction, so the amount of gasoline depends on readings of MAF/MAP sensors,
and fuel trim adjustments based on O2 sensor outputs, but not based on whether its working in 3 or 6 cylinder mode.
The VCM activations are not arbitrary, but happen when demand from the power plant is at lowest, like coasting on straight roads or downhill.
The 3 cylinders are NOT doing the work of 6 cylinders, they are doing the work of 3 cylinders, that's why van feels laggy coming out of VCM.
The instant more power is needed, VCM is deactivated and different cylinder configuration is activated, deemed necessary by the ECU.

 

[maxud]

I measured the cylinder head for flatness and there is section with a 4 mil gap. Thoughts?

Here is the spec, 2mil (aka thousands of an inch). You are twice over the limit.

 

[maxud]

Thanks for sharing.

I don't think this is the front banks low compression primary issue. We will see. Thoughts?

Does anyone know of certain head gaskets that are designed to absorb the margin?
The gasket I have, Fel-Pro 26265, is ~30 mil thick with 3 sheets. On RockAuto I see there are other gaskets at 40 mil & 50 mil. With all the different gasket combinations between thickness, number of sheets, & materials some will definitely absorb the margin better than others. Any thoughts?

Look, here is the thing about DIY, and my approach to it. You do it to save money and do a better job than the dealer.
You already saved a ton of money by doing it yourself, why compromise on getting the job done properly?
You are under assumption that you can just get the engine limping along, there is no such thing.
It either works with proper compression or it does not.

Same goes for parts, I would use OE or Mahle gaskets in a heartbeat, FelPro if I could not get anything else.
All else is just asking for trouble, and this extra thick gasket to take up the slack? Really?

Again and again, you are asking, but not listening nor answering.
Talk to machine shop if you don't want to listen to what the club members are telling you. I don't claim to know it all,
but just trying to piece together a puzzle for a complete stranger who is stuck and asking for help.

 

[maxud]

No need to check for warped head, I am only looking for primary issue, I don't need optimal performance.

I have no intentions of trying to limp along.

Do above two quotes look contradictory only to me?

I ran a proper leak down test and cylinders 4, 5, & 6 were ~40 to 45psi. The air was coming out of the dip stick.

Leakdown air pressure was 50psi.

So, let me summarize. You performed a leak down test. You provided 50psi pressure and each of the front 3 cylinders were at ~40-45 psi?
Meaning that at worse you had 20% leakdown, but maybe closer to 10%?
10% leak down is not even considered a problem. No engine is ever perfectly sealed.

Regarding your head gasket recommendation, the Mahle gasket on RockAuto is the one I was referring to as 50 mil thick. Which is ~20 mil thicker than the Fel-Pro.

Your research seems incorrect here. The 50mil head gasket is for regular VTEC aka non VCM engines. For VCM engines, Mahle lists 40mil gasket.
I can not find a spec on FelPro gasket, but its possible that it could be slightly thinner, I cant see the specs of any other makers, nor do I think it matters much.

If you are in this to find the exact cause, then clean the pistons, put new rings on, put new head gasket and new bolts on and let us know how it works out.
If it does, problem solved. If not, then you will be at it again. What else do you want to hear?

 

[maxud]

I erroneous wrote psi in the leak down test rather than %, it should write "the front 3 cylinders were at ~40-45%".

Thanks for sharing about the VTEC vs VCM on head gaskets, I have never considered that. Although when I look at the Mahle gasket description at RockAuto it doesn't mention anything about VTEC or VCM. The other gaskets don't mention it either. So the manufacturers don't make it easy to find this info. I am guessing the reason why is because it is in the noise for the typical customer. How about 3 layer vs 5 layer, do you have a preference?

The reason I posted this was to see if anyone knew what could cause sudden low compression in the front bank. I wasn't necessarily looking for help fixing the compression issue, it is a relatively straight forward fix. I shared earlier there was another post, van-shut-off-on-the-highway that experience something similar.

Again not sure where you are confused with Rock's description. Here is a pic from the web site. Each gasket is either marked VTec (non vcm) or i-Vtec (vcm). Or with engine code J35A6 (non vcm) or J35A7 (VCM). Looks perfectly clear to me. You can also compare factory part numbers, they are different for different engines.

As far as number of layers, I don't get into the depth of it. I use either factory or Mahle, they know their business. I fix the cars, not reverse engineer them.

I think I already wished you good luck fixing the issue, just make sure report the results back here.

 

[maxud]

The difference is 2010 vs 2007. The 2007 head gasket's don't call it out, but the 2010 does. Seems like a mistake/overlook from RockAuto. Are the heads the same between VCM 2007 & 2010?

I see, my bad, indeed the info is less than obvious. Rock just displays what manufacturers provide for their products. They don't distinguish two same size displacement engines in their catalog, like they would if displacement was actually different between them, ie 3.5 vs 3.7
In this case, you just gotta know what you are buying. I have been burned by this stuff before, hyundai is even worse in this respect.