Oil pump (crankshaft front) oil seal failures, my answer.

In my opinion after diligent research and inspection of several pumps that exhibited failed seals, the reason the seals fails is due to pump wear through fuel contamination of the engine oil. In most cases this is due to either incorrectly mapped modded engines or failure of a second turbo to come on line on none AFM engines. In some cases this is exacerbated by, or alternatively caused by oil contamination by detritus due to poor engine assembly cleanliness. A pump can be "worn out" very very quickly if the oil is diluted by excess fuel. I am over 90% certain of my findings...

I'll post up more detail when I get the time.

I had FMS failure about 4 times!

Seemed to go on me when i gave the car a bit of a hard drive (not thrashing though)

Got so fed up in the end. I canged oils, oil weights etc etc. Cured mine with a double spring mod and using some sealer on the outside of the seal.

Frown upon it some may but it worked

Thanks for all your research Chris, it's good to see proper examination work being done and not just best guesses.

Glad to see we've ruled out the recent failures of seals being down to pumps being faulty because of a change of design by Toyota

If the front rotor to front face of the casting isn't too badly worn, and nor is the recess the rotor locates in, then a new seal, and / or tighter seal spring *MAY* fix things, depends how much volume of oil the rotor bleeds off behind the seal. the very small drain hole Toyota used is easily overwhelmed and a high PSI will build behind the seal. On Skylines with near identical pump design, this drain back has about 10 times the area, and these never give a problem, but also the pump design on a Skyline GTR / GTS-t engine has more area on the rotor front face to housing, so there's more material to be worn away, and the rotor is better stabilised.

After checking my pump, i would agree with that diagnosis, in my case i think it was down to something abrasive getting in somehow, but i can certainly see the point about oil dilution if the second turbo has failed, as the std fuelling is very rich anyway, but it does broach the question, if the second turbo has failed, the ECU map will only provide fuel for the MAP sensors load map, however if anything else like a SAFC or piggyback is fitted and not mapped right, that will certainly accelerate wear.

Chris, do you feel there is scope for making the oil pump, feed and drain better for modified power levels?

Or do you think it should be fine assuming the oil isn't contaminated with rubbish or fuel?

Thanks for all your research Chris, it's good to see proper examination work being done and not just best guesses.

 

Glad to see we've ruled out the recent failures of seals being down to pumps being faulty because of a change of design by Toyota

As i'm sure Greg and Steve Manly will confirm, there is an issue with a recent batch Toyota pumps, where they have actually cracked!

I'll just say a big thanks to Chris for investigating, in his own time, the sudden rash of FMS failures that seemed to have plagued us.

I'm unsurprised by the results but at least it's now proven that it's not the age of the engine or a dodgy part from Toyota, so hopefully the reputation of the 2JZ in the UK will be picking back up again.

I'll be watching this thread carefully

-Ian

As i'm sure Greg and Steve Manly will confirm, there is an issue with a recent batch Toyota pumps, where they have actually cracked!

Yes, but I don't know if either of those people are qualified to comment. To me the cracking of the pump is also a different matter.

As i'm sure Greg and Steve Manly will confirm, there is an issue with a recent batch Toyota pumps, where they have actually cracked!

If they cracked due to a casting issue hopefully it was a small batch! If they cracked due to installation issues it may be that Steve wouldn't be able to tell the difference when the part was returned. Who knows. How big a batch was it and how many cracked and was it in the same place every time?

I know you've had FMS probs, did you do the work yourself?

-Ian

I haven't been able to examine (cut up?) a cracked pump, nor have i personally experienced this, so I can't make any worthwhile comment on these failures.

It will be *VERY* difficult to significantly increase the area of the drain back hole, due to the design of the casting. It may be possible, but would be time consuming and would need a jig making to hold a casting in a fixture on a milling machine. As the pumps work fine on stock engines in good condition for up to 300,000 miles I think the fix should be approached from a different direction.

If they cracked due to a casting issue hopefully it was a small batch! If they cracked due to installation issues it may be that Steve wouldn't be able to tell the difference when the part was returned. Who knows. How big a batch was it and how many cracked and was it in the same place every time?

 

I know you've had FMS probs, did you do the work yourself?

 

-Ian

I am inclined to think that Greg at Turbo fit is qualified to comment;) especially after talking directly to him, i can confirm he does know his 2JZ.

Yes i do all my work myself, except this one time as i don't have an engine lift or ramp, so i got someone else to change the pump, with me keeping an eye on things;)

Nice work Chris

Can I ask why you were specifically interested in any failures on the VVTi engine?

subscribed.

Sounds like good work Chris.

Have you created / are you creating the failure mode on a 'good' engine to prove the diagnosis?

Have you created / are you creating the failure mode on a 'good' engine to prove the diagnosis?

Are you ?

You're implying the build of an engine with some deliberate particle contamination - only to have to take it apart and meticulously clean up all oil ways / water ways etc. Have you any idea how much work (and expense) that would be?

Are you ?

You're implying the build of an engine with some deliberate particle contamination - only to have to take it apart and meticulously clean up all oil ways / water ways etc. Have you any idea how much work (and expense) that would be?

Of course I'm:blink: ...but that's not the point.

Having been involved in this type of work before (nearly 20 years with various OEMs) I'm well aware of the cost and time involved in building and stripping engines.

The ability to recreate a failure helps to prove root cause.

However I've no idea of how much time and money Chris has or is planning to spend on this.

I think it's a fair question given my not knowing Chris' situation. Again IMHO.

I'm also happy to apologise to Chris if the question is offensive.

It could be tested cheaply on a NA engine...........

There's only one thing i'm curious about and thats, if oil is contaminated to the extent that these oil pump surfaces are wearing why are the bearings/bores ok?

Sounds like good work Chris.

Have you created / are you creating the failure mode on a 'good' engine to prove the diagnosis?

Am I @&% Not unless someone pays me to, anyway:D

The reason i was aking about VVti failures was because myself and a friend who is a true ECU expert believe that the VVti may not suffer as much oil dilution with a failed turbo as it used AF metering, so is intelligent enough to know that depite the throttle angle and speed the airflow itself isn't greta enough to warrant more fuel. the MAP sensored models, unless they have pretty trick and failsafe 3D mapping, will add more fuel, without the airflow, hence go mad rich once off the Lamda. This also falls into place with seal failure being unknown (???) on the N/a's.

Slick

How much over fueling are we talking about here?

Bore wash has been mentioned here a bit in recent times, can the sump oil be contaminated without bore wash occuring?

In a recent thread about someones smokey rebuild Chris stated the following

"Sounds like running it crazy rich on new rings and bores has *uggered it"

&

"Yes, my money would be on bore wash. You ^MUST^ get an engine mapped as soon as it is started and you ^MUST^ start mapping from idle. It's madness to build some wild engine (wild meaning bigger injectors, N/A to turbo, different fuel pressure, blah blah) and then drive it to a mappers or furtle around in it for hours beforehand. You should be *AT* the mappers for when it *FIRST* fires up!!"

Chris was suggesting the owners oil burning problem was down to bore wash from overfueling.

Now though, overfueling is contaminating the oil which is causing premature pump failure?

Is it that seriously rich conditions cause bore wash / piston ring sealing problems very quickly, whereas prolonged slightly rich conditions cause the eventual highly contaminated oil leading to pump wear?

Serious overfuelling is bad, period, it can borewash and cause rapid ring wear, and bore wear and glazing, as well as cause general engine wear. If fuel contamination of the oil is really bad a backfire can blow the sump and rocker covers off engines. This was a fairly common scenario in the days of the Ford V6 Zephyr / Zodiac engines, where the mechanical fuel pumps would weep neat fuel into the sump due to diagphram puncture, and a spectacular explosion was occassionally the result. I am not saying that this is likely here, but mentioning it just to show that various levels of contamination can cause different levels of damage.

This also falls into place with seal failure being unknown (???) on the N/a's.

So VVTi and UK Spec which, all run MAF are included in the above ?

Ok, loads of scary sh1t then.....

Is a 'well monitored' AFR gauge is a reasonable way of recognising the symtoms?