Honda B Series VTEC Gasket Fractures and Oil Leak after track abuse - Any ideas?

Hey guys,

Hoping maybe other fellow B series owners who track their B's can chime in with their experiences, but also looking for any other general advice!

Ever since I moved from D series architecture to the B series in my track day Civic, I've had repeated VTEC solenoid gasket failures after every track weekend since this engine has been in the car.

The failure pattern is fully observed at this point:

• Install new VTEC solenoid gasket to repair the previous broken/leaking one.
• Attend a one or two-day HPDE event, thrash the car as normal. No leaks or issues all weekend.
• Trailer it home. Let it sit on the trailer for another day out of laziness after the event.
• Crank the engine up to back it off the trailer to put it in the garage, observe the gusher of an oil leak from the VTEC solenoid mounting surface between the solenoid body and the cylinder head.
• Remove the solenoid to inspect, find the gasket cracked/fractured along the top area where the metal screen is.

And repeat the cycle.

Here's what the failure looks like:

Things I've done so far:

• Inspect the solenoid body and cylinder head for burrs, sharp protrusions, etc. - none.
• Monitor for high oil pressure - normal, nothing crazy during cold/warm/hot/on track operation (85lbs cold idle, 35lbs warm idle, 25lbs hot idle, 78lbs hot/WOT on track at 8500RPM)
• Try other gasket manufacturers/resellers - 5 different sources tried, everything from eBay cheap, Dorman aftermarket, and dealership Honda.
• Different ranges of fastener torque - OEM service manual fastener torque with a calibrated 1/4" torque wrench, to german "gutentite" torque.

I'm at a loss so far on the exact reason this is happening, although I have some thoughts:

What seems to be happening is the gasket and surrounding aluminum is expanding and forcing the gasket into its metal screen. There is a metal screen "tucked" into the gasket to prevent debris from entering the VTEC system, and it is this area around the gasket precisely where these breaks/fractures are occurring.

I have a feeling the gasket is being forced into the screen metal edging during expansion due to extended heat exposure, and it is this which is cutting/slicing the gasket, causing a hairline crack to begin forming, where it slowly begins to grow over the rest of the weekend.

At which point the crack has propagated enough on cool down, that when I crank the car up once its fully cooled off for a couple days, the oil pressure hits it and finishes ripping the gasket enough to cause a leak.

There is nothing else that I can see mechanically that might cause this. As explained, the only thing I'm thinking of now is heat; expanding the VTEC solenoid, the head material, and the rubber enough, that the metal screen is the hard sharp place it cannot slide past to safely deform away from.

All speculation of course. But my next approach is to try running it without the screen on the next event to see what happens. I'm not worried about running it without the screen just to see what happens. I meticulously build and maintain my engines, and this engine is well broken in by now, so if stuff is making its way into the VTEC system, there are much bigger problems.

Thoughts from the group?

Thanks,

Talon.

This might get more eyeballs in the main forum. Want me to move it? And, FWIW, never had an issue with my B16A-powered CRX. 

Does this gasket sit in a groove or between two flat surfaces? Is it possible you have a strange high pressure issue that is blowing it out? 

David S. Wallens said:

This might get more eyeballs in the main forum. Want me to move it? And, FWIW, never had an issue with my B16A-powered CRX. 

Sure thing, whatever you think would be better! I'm the noob here haha. I'll lean on your recommendation!

Interesting, thanks for your your input from your experience! This is also a B16A, was yours tracked often? Turbocharged by chance?

Realizing I didn't even lead with even my block code in the original post haha, I suppose it might be a good idea to list out some engine mechanical config and mods:

• JDM B16A; second gen
• Block
  • Stock crankshaft; still spec'd perfect, micropolish
  • ACL Race Series bearings; STD clearance, WPC treated
  • Main and Rod oil clearances are all at factory spec, some bearing shell swapping required, but verified clearance with bore gauge
  • Stock caps and cap bolts, no girdle
  • Stock windage tray
  • Ishihara Johnson Teflon Crank Scraper
  • Stock oil pickup and pan (not baffled beyond OEM setup)
  • B18C5 Integra Type R oil pump
  • 81.5mm bore
  • 81.5mm Wiseco pistons, 11.5:1 CR
  • Speedfactory B16 H beam rods w ARP bolts
  • No CSS or block guard
  • InlinePro head studs
  • JE ProSeal head gasket
  • Stock water pump, timing tensioner components
  • Gates Racing Timing Belt
  • Aftermarket timing belt leading edge tensioner/guide
  • ATI Race damper, underdriven alternator
• Head
  • Stock Cams
  • Stock Valves
  • Stock B16A intake manifold
  • Stock VTEC solenoid/spool valve assembly
  • Ported TB intake opening from 60mm to 62mm to accept B18C5 throttle body
  • Stock Rocker Arms/Cam Followers
  • Supertech 70lb springs (modest pressure increase above stock), Ti retainers
  • K series Lost Motion Springs
  • Golden Eagle Adjustable Cam Gears
  • OEM B series valve cover, baffle removed, cam gear cutouts added, AN fittings added
• Engine Fluids/Related Information and Other Consumables
  • At this point, VTEC gaskets lol.
  • Wix XP oil filters
  • Motul 300V 5w40
    • Use at track; 5wt for good cold volume up to bearings, 40wt to maintain film thickness for bearing clearances in track hot temps
    • Oil temps with my oil-water-air cooler setup never exceed 240°F
    • Idle pressure = 25psi @ 1000rpm @ 235°F
    • WOT pressure = 75-80psi @ 8500rpm @ 235°F
  • Mobil1 5w30 full syn
    • Use on the street; 5wt for good cold volume flow up to bearings, 30wt for good film thickness for bearing clearances in cooler non-track hot temps
    • Oil temps on the street with cooler setup off rarely exceed 205°F
    • Idle pressure = 25psi @ 1000rpm @ 195°F
    • WOT pressure = 78-82psi @ 8500rpm @ 195°F
  • VP Stay Frosty antifreeze
    • Cooling system mods + this coolant = happy ECTs
    • Half rad full shroud +turbo downpipe wrap + ducted front end/intercooler + hood louvers + Bumper Scooper 4000 + VP Stay Frosty = coolant temps never exceed 215°F
  • NGK BKR7E plugs, 0.017" gap

Moved the thread, and my CRX had a JDM B16A: stock internals, header, intake, Hondata ECU. Never had a problem with it. 

Crazy...My integra is only NA and the only issue I've had with the vtec gasket was one bolt backing out causing leaks after a track day. 

I did put a light coat of hondabond around the gasket afterwards and it never leaked. I do wonder if its the extra heat from the turbo. Not many people track turbo Hondas vs how many there are in the drag racing world

rlacasse1 said:

Does this gasket sit in a groove or between two flat surfaces? Is it possible you have a strange high pressure issue that is blowing it out? 

Yes sir, the gasket sits in a groove in the solenoid body itself, exposing just under half of the gasket above the groove surface, which then bolts to the head. The head itself is a flat machined surface. Both surfaces, and the solenoid body groove channel itself, are burr and mark free. I've inspected this thoroughly each time I've had the valve off to replace the gasket, even feeling the surfaces with a plastic o-ring removal tool, running down the valve body channel and the head surface with some force to feel for deep catches I might not be able to see/feel otherwise.

I suppose there could be a high pressure issue, and this is something that I could investigate moving forward. On a side note but still related, I'm currently not doing so, but I've been meaning to tap into the linear 0-5V signal from the oil pressure transducer currently installed to the block to run the gauge, run a wire to an open input on my ECU, program the offsets to the input in the software, and then datalog my oil pressure over my sessions. This could be an easy "2 birds with one stone" item to take care of, as datalogging my run oil pressures has been on my list of things to do regardless!

If there is a high pressure issue, it is currently not sustained long enough to make it out on my dash gauges. The gauges I use include a min/max alarm setting. Min alarm is set to 10psi, so this goes off normally when the engine is just Key On Engine Off. The alarm is a sustained beep plus a bright red LED. I've never once seen it hit the max alarm, which is set to 100psi.

I frequently gauge glance as part of my normal track day routine, since early on with this car I battled high coolant temps. But those issues are pretty much sorted out now, yet I still tend to glance at gauges more than I need to lol. I feel I might have noticed this alarm if it did indicate something, but it is possible it is just too fast for the gauge slow overall update rate.

Cool build!

I would try:

1. Re-torque (to oem spec) the bolts to between the solenoid and the head after they have been heat cycled at the track and before you start it up to take it off the trailer.

2. After that I would try swapping out the solenoid itself and (in a perfect world) swapping the solenoid out to another head to see if the issues follows the solenoid or stays with the head on the car.

3. Honda doesn't recommend Hondabond on the gasket itself, but I've heard of it solving leaks as well. 

Good luck with the oil pressure logging! I suspect you'll have to start by swapping out the OE transducer, or adding another sensor to a different area of the engine-from memory there are two different oil sensors, one for the gauge and one for the idiot light on the dash. 

What ECU setup are you running for logging?

I wonder what would happen if you remove the screen from the gasket.  Maybe 80psi hitting that screen at WOT causes oil to try to go between the screen and the gasket.

I dug up some info from a 2000 Integra B18 engine and this is what I found.  The bottom line is what I am looking at.  (Seriously)  At idle, 10 psi, at 3000rpm 50 psi min.  I have not been able to find whether the VTEC oil operating pressure is different from the engine.

CrustyRedXpress said:

Cool build!

I would try:

1. Re-torque (to oem spec) the bolts to between the solenoid and the head after they have been heat cycled at the track and before you start it up to take it off the trailer.

2. After that I would try swapping out the solenoid itself and (in a perfect world) swapping the solenoid out to another head to see if the issues follows the solenoid or stays with the head on the car.

3. Honda doesn't recommend Hondabond on the gasket itself, but I've heard of it solving leaks as well. 

Good luck with the oil pressure logging! I suspect you'll have to start by swapping out the OE transducer, or adding another sensor to a different area of the engine-from memory there are two different oil sensors, one for the gauge and one for the idiot light on the dash. 

What ECU setup are you running for logging?

 

Thanks!

I've torqued the solenoid to the head using an electronic inch-lb wrench to factory spec. One of the first things a few buddies thought of as well. Combined doing this with a brand new gasket from Honda, unfortunately it split the next day the same as all the others following a very successful and fun track day out at CMP.

I have not yet tried another solenoid body itself, but I do have one in the shed I can refurbish and install to see what happens. I have checked both the valves and the side of the head for warpage with a machinists straightedge and feeler blades over multiple different positions, neither are warped. Definitely don't have another B series head laying around lol, but it is on my wish list to acquire over the coming months/years, as I'd love to build and have another long block and trans on standby. But $$$$.

Seriously thought of just adding some Permatex Ultra Black around the outsides of the solenoid body just to see what happens early on with these failures, but having been through so many gaskets and just knowing whatever is causing these gaskets to split like this, silicone on the outsides just wouldn't address root cause and would likely leak again once the gasket failed and oil pressure began to push on the RTV.

I have a 0-5v transducer installed already for my gauge, just going to tap into the 0-5v signal wire and send the signal to my ECU, and configure it. On this engine, Honda never had an oil pressure transducer to monitor actual pressure, only the idiot light, which would come on @ approx. 5lbs. The the idiot light is gone, and the stock cluster doesn't exist in this car anymore lol, only the aftermarket gauge is present. I use my phone as part of the instrument panel for speedometer, tach, and other data items. Here is what the interior looks like:

ECU is a refreshed stock but modified P28, equipped with a Moates Demon2 emulator, HondaRulez Coil On Plug expansion board, and high current PWM output components for electronic boost control, etc. Tuning software is Neptune RTP; all of this purchased way back when Moates and Xenocron were doing very comfortably, and not feeling intimidation from the EPA. This setup is working well, and I have no intentions of changing it out until I get to the point where I can't massage it back to life any longer (still have lots of Honda OBD1 ECU consumables and known failure components on hand, enough to refurbish maybe 5 ECUs completely). If it ever got to that point of replacing the ECU, or if I ever decide to move to a K series in the future, I'll be going with a FuelTech or Haltech full standalone setup of some kind at that point.

Perhaps:   Some combination of head milling, block milling, and gasket height is yielding too small of a gap when assembled and the molded part of that gasket which is a dynamic seal similar to an O-ring is getting compressed beyond spec and blowing up the part,  or deforming the o-ring type seal into a not sealing anymore shape.

In reply to drtalon123 :

Sweet build! Honda Tuning Suite is still under active development for OBD1 Honda ECUs and I think they support K-series as well.

"I've torqued the solenoid to the head using an electronic inch-lb wrench to factory spec. One of the first things a few buddies thought of as well. Combined doing this with a brand new gasket from Honda, unfortunately it split the next day the same as all the others following a very successful and fun track day out at CMP."

Just to be clear, did you re-torque the bolts after the track day and before you remove it from the trailer? The hypothesis would be that heat/vibration from the track loosens the bolts just enough for the gasket to leak after everything has cooled back down. 

drtalon123 said:

CrustyRedXpress said:

 

ECU is a refreshed stock but modified P28, equipped with a Moates Demon2 emulator, HondaRulez Coil On Plug expansion board, and high current PWM output components for electronic boost control, etc. Tuning software is Neptune RTP; all of this purchased way back when Moates and Xenocron were doing very comfortably, and not feeling intimidation from the EPA. This setup is working well, and I have no intentions of changing it out until I get to the point where I can't massage it back to life any longer (still have lots of Honda OBD1 ECU consumables and known failure components on hand, enough to refurbish maybe 5 ECUs completely). If it ever got to that point of replacing the ECU, or if I ever decide to move to a K series in the future, I'll be going with a FuelTech or Haltech full standalone setup of some kind at that point.

Damn.  So old school.  Similar to what I used to use in my D-series autocross days decades ago. 

And here I thought I was so OG with my recent BRX build using a Hondata S300 in a modded P06.  Still, the box is 30+ years old and refurbed or not, scares me from a reliability standpoint.  Of course, the chassis wiring is all original too, so it kinda means you go all in and rewire *everything*...which is quite the project.  So I totally get why you are staying put for now.  "Not broken, don't fix"