[mfuni]

I just wanted to check in and say that after over 500 miles, my repairs have appeared to hold up. I hope I didn't just jinx myself...

 

[Fleetw00d]

I'll try to summarize my daughter's P2646 saga. After I did a partial rebuild (honed cylinders, new Honda piston rings, new Honda rod bearings, new Honda gaskets everywhere I disturbed) at 206,000 miles, she started having very intermittent P2646 issues, typically while on longer highway drives. She could usually pull over, shut it off, restart and go on her way without having the code immediately repeat. She's put about 10,000 miles (fresh Mobil 1 5W20 high mileage at rebuild and again at 211,000) on it since the rebuild and started keeping a better mental record of when it happened. It usually took a good 45 minutes of highway driving before the code (might) set. Several trips to a Honda dealer resulted in new VTEC oil control valve, rocker arm solenoid and pressure switch (including gasket/screen), VTEC screen (behind the PS pump), and replaced the switch connector because of corrosion. The problem didn't go away. Honda finally pulled out all the appropriate diagnostic equipment (adapter to read oil pressure at the rocker arm switch) and determined that on the highway after driving for 45 minutes to get the engine good and warmed up, if you backed out of throttle lightly from about 60 mph, the transmission would stay in (or shift to) 5th and the engine rpm would drop to maybe 1500, resulting in enough of a drop in oil pressure to about 25 psi to close the switch (high pressure opens the switch) and trip the fault. Oil pressure at idle is only 5 psi (supposed to be at least 10). They want to drop the pan and check the oil pump (discount price of just $600). If the pump is bad (how likely?), Honda doesn't sell just the pump, only the entire oil pump and balance shaft assembly (list price $810). If it is worn crank bearings, that's an even more expensive repair. For now, we plan to try running 5W30 and see if that raises the pressure enough to prevent the problem from recurring as often as it has.

For those who want more info, I have the original engine (another long story). I cut apart the rocker arm solenoid valve to see how it works. The valve spool is spring loaded closed; based on the spring load and valve area, it takes about 35 psi to open it. The switch for this particular assembly also opened at 35 psi. The valve is supplied with oil pressure to the lower port (with the screen). There is a passage from this lower port up to the solenoid. From this passage, there is an orifice that feeds the middle port to the rocker arms. With the valve closed, this little bit of oil is probably there for initially filling the rocker arm feed passages and making up for leakage through the rocker arm pistons. When the solenoid is opened, oil pressure is applied to the top of the valve, pushing it down to connect the lower port to the middle port. The top port is vented to the under right side of the valve cover. There is a passage from the center port to the switch. When the solenoid shuts off, the valve moves up to isolate the lower port from the center port and opens a vent from the center port to the upper port to drop the pressure to the rocker arms.

 

[Xangryth]

Thank you for giving me the numbers on the oil pressure setting that caused the faults. The oil pump pressure relief spring can be shimmed to increase the Honda pump pressure. You can google to read about it. I remember reading how someone cut the end off a ring tongue electrical crimp lug and used it increase the pump pressure. I don't know why the Honda tech didn't just do an oil change and put in 10-30w oil and repeat the test to see if that would raise the oil pressure enough to cure the fault. Oh- it's because there is no 10-30W oil to be found in the Honda dealership, haha!
I have been saying all along that low oil pressure is the main culprit. Using a heavier weight oil has always been done to keep oil pressure up on aging motors and back in the day when cars had oil pressure gauges a person could see the immediate beneficial results with a heavier weight oil. There is a good reason why 10-40w and 20-50w oils are available- they often work to keep high mileage cars running longer. Here is a gauge for a mid 1970s Camaro- as you can see 40psi is the normal running pressure- 30psi was also a common mid point pressure for cars. If only Honda had put in a real pressure gauge for us owners. Honda knows what the normal operating oil pressure is for a Honda Element- it's just too secret to share with us.
The common solution is to replace parts- no tech ever asks the question-maybe the oil pressure is too low and that is why the switch has faulted out?

Honda Oil Pressure Switch Issues

Resolve oil pressure switch problems in Honda vehicles by replacing the rocker arm oil pressure switch or the engine oil pressure switch.

www.underhoodservice.com

 

[07lmnt]

Good info on the workings of the system.
I added an led to the vtech oil sensor circuit. This way I can monitor when the system turns on or off. What I found is the vtec engages at 2200 rpms and turns off at 2000. Does not matter if it's hot or cold, always at those rpm's.

I have no problems, but only have 82,000 and I have used 10w 30 mobile 1 since new.

 

[Xangryth]

oil pressure too low?

My 06 element started sputtering when I push down on the accelerator and the check engine light came on. So I took it to the mechanic and it was reading an oil pressure code. My mechanic changed out the sensor in the V-tec and cleaned the screen. Long story short he has changed out two sensors...

www.elementownersclub.com

Clueless mechanics- back in my day, if the oil pressure warning light was coming on, the mechanic wouldn't say "lets change the pressure switch for the light and see if that fixes it"

 

[Xangryth]

Good info on the workings of the system.
I added an led to the vtech oil sensor circuit. This way I can monitor when the system turns on or off. What I found is the vtec engages at 2200 rpms and turns off at 2000. Does not matter if it's hot or cold, always at those rpm's.

I have no problems, but only have 82,000 and I have used 10w 30 mobile 1 since new.

Thank you so much for that info-- and your Element never blew up using "non-approved" 10W-30 oil, as some people think will happen?
Who is it that always says here, only use HONDA APPROVED 5w-20 oil!
I would have thought the rpm would be over 2500-

 

[07lmnt]

OPPSIE I meant 5w 30.
Some of the K engines in Acura's actually spec 5w30. I use it because it's free.
Yea I was surprised the vtec rpm is that low. Driving around town, that thing is on off a thousand times! Thought it was more like 3600 or more.

 

[Xangryth]

I recently posted a picture of the oil cap for an Acura car that says 5w-30- I use that weight in mine and I just turned over 237K miles

 

[Fleetw00d]

If there is too much leakage through worn bearings and such, increasing the preload on the pump pressure spring won't help - it is the resistance to flow downstream of the pump that determines the pressure throughout the engine. The more leakage there is, the less resistance to flow. Increasing the viscosity of the oil increases the resistance to flow, thus the pressure. Know which Acura application specifies 5W30? Maybe I'll buy her a new oil cap.

 

[Xangryth]

Lookie, what you can buy at the Acura and Honda parts store-- somebody tell them that 30w oil is not approved for Honda engines!
The service writer would lose his job, if he told the mechanics to first do an oil change with heavier weight oil, before he started changing out $$$ V-TEC parts.

 

[07lmnt]

Agree 100% If the pressure is lower than normal due to internal wear more volume is needed , not a higher pressure relief spring.
Way back in the 80's Buick V6 were well known for low oil pressure at idle. The band aid was to install a high volume oil pump. It was easy to iinstall because the pump is external. Longer spur gears and a spacer installed with the same pressure spring. I did many of these, and never had any more oil light on at idle.

 

[Rolla Wayne]

Vtec oil sensor problem

I didnt have this problem till I hit a deer and had the vehicle fixed and they said the Vtec sensor had a recall and they replaced it for nothing since its a recalled part. Now year and half later getting the same problem and they replaced the sensor again saying the oil needed to be changed which I had changed 3 weeks prior to the problem showing up. The Honda dealer said the screen was cloged and they cleaned it and said to change the oil again which I did. IT STILL HESITATES AT 3000 RPM after all this!! But I noticed it doesnt do the hesitation until after it is heated up from driving for about 5 minutes. Wish there was some way to bypass that sensor totally and get rid of it!!

I have the exact same problem on my 2005 element (110k miles ) I've replaced the VVT asy (sw and solenoid ) twice ,first time fixed it for 6k miles ,second time only a few days . The other screen they talk about on the front of the engine is new too. My oil is new and it's been changed 2 times to fix the problem to no avail.
Did you get your problem fixed and if so , what fixed it ?

 

[409]

Well my dear EOC friends I have had my first problem with my Element after 96,000 miles. While driving down the road the other day my malfunction indicator lamp came on. I instantly thought it was my gas cap so I removed and reinstalled it and took off again but I started cutting out and losing power. I instantly thought it was an 02 sensor, plugs, wires, or cat. convert. so I took it in to Autozone this morning to have the code pulled and got [PO2646 Rocker Arm off or stuck] code. I have never heard of such a code so I took it in to my dealership for them to hook it up and see what they find. I was wondering if anyone else has heard of this code or has any clue what could be the possible cause. Ill keep you updated.

 

[godfather_perez]

I have the exact same problem on my 2005 element (110k miles ) I've replaced the VVT asy (sw and solenoid ) twice ,first time fixed it for 6k miles ,second time only a few days . The other screen they talk about on the front of the engine is new too. My oil is new and it's been changed 2 times to fix the problem to no avail.
Did you get your problem fixed and if so , what fixed it ?

Dreaded VTEC P2646 code - Tried everything

Well, my 2005 Element (212K miles) became infected with the P2646 VTEC malfunction code last week. CEL comes on and goes into limp mode (won't go over 2800 RPM). Turn the ignition off and back on, and it goes out of limp mode but usually comes back on soon. My 2003 CR-V (199K) has had the same...

www.elementownersclub.com

 

[mdelux]

After reading many jpages regarding this issue, it appears to me that replacing all of the common parts associated with P2646 and P2647 will more often than not fail to solve the problem, myself included. Is it possible to bypass the vtec sensor wiring to prevent the warning light activation, avoid the engine shutdown at 3000 rpms, and continue to drive without the infuriating on/off P2646. I realize that is risky if other major engine problems are present but no one in the forum seems to find any common issues; and Honda is certainly no help. At 225,000 miles, I am willing to take my chances if I can rid myself of this plague.

 

[Fleetw00d]

Unfortunately, the switch is used both to indicate that the system turned on (high pressure opens the switch to verify that the solenoid opened) and low pressure closes the switch (to verify the solenoid closed), so if you disconnect the switch (circuit open), it will always indicate high pressure and if you short the switch it will always indicate low pressure; either way you get a fault and the system goes into limp mode. If you were clever, you could set up a circuit that opens the switch signal when the solenoid is powered, then closes when the solenoid is depowered.

Do you know what your oil pressure is, either at the pump or at the rocker arm switch?

 

[07lmnt]

Try 10w30 oil. Only if you are sure the screens are clean.

 

[te1966]

After reading many jpages regarding this issue, it appears to me that replacing all of the common parts associated with P2646 and P2647 will more often than not fail to solve the problem, myself included. Is it possible to bypass the vtec sensor wiring to prevent the warning light activation, avoid the engine shutdown at 3000 rpms, and continue to drive without the infuriating on/off P2646. I realize that is risky if other major engine problems are present but no one in the forum seems to find any common issues; and Honda is certainly no help. At 225,000 miles, I am willing to take my chances if I can rid myself of this plague.

You are correct. There is no "Just Do This" answer that is guaranteed to solve this problem. Even if you do Everything that we discuss at length on this site!

I'm one of the lucky ones, I suppose. My '03 EX went into limp mode twice about a year ago. But both times it went away all by itself! And it hasn't come back since! (It's been almost a full year and about 8K miles now.) I did put about 5 ounces of Seafoam into the oil, and ran it about 500 miles before changing it; just to see what would happen. Also I'm using Valvoline 5W20 Synthetic Blend and upgraded to a Wix 57356 oil filter. But, I still have no idea what would have made it go away! I mean, cars just don't go and fix themselves! So I understand your frustration.

Unfortunately, I bought my '03 EX with 111K miles and zero maintenance records! : - ( I took it to my highly trusted mechanic (who also owns an Element) to check it out. The very first thing he did was remove the oil filler cap, looked down into the valve cover, and stated "Well, it has had oil changes, but not often enough". So I started using my old standby oil additive, the thin viscosity Rislone that I have been using since the '70's; thinking that it would "clean it out". I used it with my first 3 oil changes. Was this a mistake? Did the Rislone cause the P2646 problem? I really don't and won't know! However I did stop using the Rislone on the next 2 oil changes after the problem started. The only other thing that is worth mentioning here is my VTEC solenoid assembly looks newer than the rest of the back side of the engine; like it has been replaced at least once.

Is my problem solved or will it come back? I don't know.

In the end, this problem is initially caused by not changing the oil often enough, using cheap oil filters, and perhaps by using oil additives that shouldn't be used. If since their Element was new, one changes their oil often at maybe every 3-4K miles (or whenever it just starts to turn black if you use it in dusty conditions), uses a very high grade oil in either 5W20 or 0W20, and a very high grade oil filter, you will most likely never have an issue with the VTEC system.

With all that said, I think you are on the right track by using Seafoam (I read your other post) which hopefully will in time clean out any gunk and grime in your engine; which is bound to be still in there if it went for very long intervals between oil changes. And if that was indeed the case, you may need to change your screens a second time.

And I also agree with 07lmnt's suggestion of trying 10W30 oil. At 225K miles your bearings are going to be worn somewhat, lowering your oil pressure. It's a simple fix to try and at this point you have nothing to lose by trying it.

I wish you well!

 

[snake_eyes]

Well my dear EOC friends I have had my first problem with my Element after 96,000 miles. While driving down the road the other day my malfunction indicator lamp came on. I instantly thought it was my gas cap so I removed and reinstalled it and took off again but I started cutting out and losing power. I instantly thought it was an 02 sensor, plugs, wires, or cat. convert. so I took it in to Autozone this morning to have the code pulled and got [PO2646 Rocker Arm off or stuck] code. I have never heard of such a code so I took it in to my dealership for them to hook it up and see what they find. I was wondering if anyone else has heard of this code or has any clue what could be the possible cause. Ill keep you updated.

so what i did because this plagued me for a long time was i took off the vvt cleaned out the crud from the screen got a new OEM this is key!!! oil pressure switch... honda hates a/m engine parts and so far its good...

 

[bork]

so what i did because this plagued me for a long time was i took off the vvt cleaned out the crud from the screen got a new OEM this is key!!! oil pressure switch... Honda hates a/m engine parts and so far its good...

I had tried 2 new OEM sensors & they still tripped the code. I know a mechanic that used a Dorman and he said it's lifetime guarantee, and he said it worked. I wonder now after months have passed if this is still true. I'm back to my old original which is still "dare I say" fine. I did replace the new OE ones because they set the code & I could see a bit of oil starting to leak inside where the wire connector plugs in. Took out sensor to bench, and I hooked up ohm meter to check contacts when adding air to the hole, I found the original, more sensitive & could actually hear a slight click when appling air on and offf. I dont know what the sensor looks like inside, diaphragm or a sealed plunger? But maybe they just need extreme exercising by applying air, on & off, I don't know? IIRC the air pressure was dialed down to approx 5 lbs, which my cheap little regulator could not keep constant, so ended up with 20#. Probably the right way would be making a small manifold and screwing in all the sensors with a manometer water column "U" tube for acuracy, while tesing sensors with air pressure. The mechanic I know, used to work at Dealer & still has connections, told me there was a bad batch of OE sensors. This mechanic also had a problematic element that kept returning with the code & they were leaning towards a PCM/ECU to correct the issue, but the last I heard the client has not been heard from since the last sensor change. So after a few months the mechanic doesn't know if issue has been resolved, no return client so far. I'll try calling him to see if any update, been almost a year now. My mechanic also drives/owns an Element.