Diesel fuel in oil

[MakinDo]

Having injection pump rebuilt now so we will see if that does it

 

[MVDarlin]

Excess Oil

Sounds like fuel is leaking into the crankcase somehow
And lots of it!

 

[twistedtree]

Having injection pump rebuilt now so we will see if that does it

It would be good to head from the pump rebuilder if they can identify a leak before they rebuild it.


You have a major leak, and the pump is a clear candidate. Earlier someone asked about whether the fuel return path/valves were open? A mistakenly closed return line valve can cause pump damage, and might explain the sudden leakage.


Beyond that I think you need specific knowledge of how the fuel system is plumbed on the perkins so you can identify possible leakage points. For example, so engines deliver fuel to the injectors via passages in the cylinder heads. A failed seal where the injectors fit into the head can allow fuel to dump into the crank case. Other engines have all the fuel contained in dedicated pipes from injector pump to injectors, so no escape path for the fuel into the crankcase other than through the pump itself.


I think you confirmed that you don't have a mechanical lift pump, and an external electric pump instead? A mechanical lift pump is a possible, if not likely leak point.

 

[STB]

The Perkins 6.354 is very straight-forward w.r.t. the fuel path. It goes into the lifter pump, out of the lifter pump, through the filter, into the injector pump, and from there into each injector, and from there out of each injector into a common return rail, and from there back to the tank. There is also a bleed path from a bleeder hole in a banjo bolt at the fuelcfilter back to the tank via the return rail. All outside of the engine block. No internal passages. None.

Fuel can get into the engine only where this path interacts with it: each injector, lifter pump seal (interaction because the engine drives the lifter pump, not fuel path), injector pump seal (interaction because the engine drives the injector pump, not fuel path).

Fuel can make its way into the oil via exactly one of these interfaces or via the normal combustion path, in normal or abnormal volumes.

Given the volume of fuel, without heavy white diesel smoke, it is dramatically unlikely that injectors are dumping enough fuel into the engine to cause this.

Without a huge amount of blow by, and likely also a smoking problem, it is pretty unlikely that the engine has an internal problem, e.g. with a ring, bad enough to cause this. Having said that, I've see. These engines seeingly run surprisingly well on 5 cylinders.

If there is a leak or crack or hole along the supply or return path to or from the injectors, is into the bilge. These paths are all external to the engine. A return leak can not cause this problem. The path isn't near oil. I guess the exception would be if the valve cover were off or if the oil dipstick were out and it drained from the engine or area around into the engine. Or, maybe, got sucked into the air intake. Or something like that.
But the engine room would be red and slimy beyond doubt first.

An /obstructed/ return path, allowing fuel to back up and pressure to form in what should be a free flowing path, could conceivably cause an injector problem or temporary or permanent seal failure that could possibly cause this. My model of this failure mode isn't super good. Something would have to give or misbehave. If it was at an injector or injector pump seal, it could be a problem.

If the return checks good, I'm betting the shop finds and fixes the problem at the injector pump. While you are paying them to rebuild the pump, if you want, you can ask them to pop test the injectors for free. Shops usually do that as a courtesy.

If that doesn't do it, it is leak down test time. But it is hard to imagine that type of problem without a lot of blow by.

 

[twistedtree]

The Perkins 6.354 is very straight-forward w.r.t. the fuel path. It goes into the lifter pump, out of the lifter pump, through the filter, into the injector pump, and from there into each injector, and from there out of each injector into a common return rail, and from there back to the tank. There is also a bleed path from a bleeder hole in a banjo bolt at the fuelcfilter back to the tank via the return rail. All outside of the engine block. No internal passages. None.

Fuel can get into the engine only where this path interacts with it: each injector, lifter pump seal (interaction because the engine drives the lifter pump, not fuel path), injector pump seal (interaction because the engine drives the injector pump, not fuel path).

Fuel can make its way into the oil via exactly one of these interfaces or via the normal combustion path, in normal or abnormal volumes.

Given the volume of fuel, without heavy white diesel smoke, it is dramatically unlikely that injectors are dumping enough fuel into the engine to cause this.

Without a huge amount of blow by, and likely also a smoking problem, it is pretty unlikely that the engine has an internal problem, e.g. with a ring, bad enough to cause this. Having said that, I've see. These engines seeingly run surprisingly well on 5 cylinders.

If there is a leak or crack or hole along the supply or return path to or from the injectors, is into the bilge. These paths are all external to the engine. A return leak can not cause this problem. The path isn't near oil. I guess the exception would be if the valve cover were off or if the oil dipstick were out and it drained from the engine or area around into the engine. Or, maybe, got sucked into the air intake. Or something like that.
But the engine room would bebred and slimy beyond doubt first.

An /obstructed/ return path, allowing fuel to back up and pressure to form in what should be a free flowing path, could conceivably cause an injector problem or temporary or permanent seal failure that could possibly cause this. My model of this failure mode isn't super good. Something would have to give or misbehave. If it was at an injector or injector pump seal, it could be a problem.

If the return checks good, I'm betting the shop finds and fixes the problem at the injector pump. While you are paying them to rebuild the pump, if you want, you can ask them to pop test the I jesters for free. Shops do that as a courtesy.

If that doesn't do it, it is leak down test time. But it is hard to imagine that type of problem without a lot of blow by.

Thanks. That's very helpful, and it does indeed sound like a real short list in fuel to crankcase entry points. I also agree completely that it's not a leaking injector or blow by - it's just way too much fuel with no other symptoms.


Personally I would feel a lot better if the pump rebuilder put some pressure on the pump and identified the leak. I have a real pet peeve about "fixing" stuff via replacement or rebuild without conclusively identifying the problem. Anything else is just guessing and swinging in the dark. They might be well educated/experienced guesses, and sometimes that's all you can do, but I always feel better when there is a clear smoking gun.

 

[STB]

Personally I would feel a lot better if the pump rebuilder put some pressure on the pump and identified the leak. I have a real pet peeve about "fixing" stuff via replacement or rebuild without conclusively identifying the problem. Anything else is just guessing and swinging in the dark. They might be well educated/experienced guesses, and sometimes that's all you can do, but I always feel better when there is a clear smoking gun.

I can't imagine them not bench testing it first. They usually do this so they can explain any differences thebowner might notice-- and to protect themselves by documenting things if it had major issues coming in and isn't fixable or fromnwarranty claims by double checking known problem points.

But, I've been surprised more than I'd likely lately.

 

[Jeff F]

The only way the IP can dump fuel into the oil is through the shaft seal, right? This seems like a very specific problem that should be easy to test for.

Sent from my moto g play (2021) using Trawler Forum mobile app

 

[MakinDo]

Agree with all. The injectors were done about a month ago and they were pretty dirty. Lift pump was changed 2 months ago but I might change again since sometimes they fail and are relatively cheap. The injection pump was leaking fuel down the side of the block. Expecting a report on Monday on what they find on the rebuild. I did verify the fuel return was clear all the way from the injectors to to the tank.

 

[STB]

The only way the IP can dump fuel into the oil is through the shaft seal, right? This seems like a very specific problem that should be easy to test for.

Yes and yes! It'll show right up when they test it on the bench.

 

[ronobrien]

Rid yourself of the diaphram type on-engine lift pump and replace with an electric pump and you will never have diesel in your from from that source again.

I'm with you. The location of one of my lift pumps make access very difficult. So better locating an electric makes it much easier. Also the failure of an electric pump is pretty obvious, where as the failure of the mech pump diaphragm could lead to an engine loss, or at least shortened life due to oil dilution. All in all I agree with the electric pump idea.

 

[MakinDo]

Turns out my injection pump needs a major rebuild. The diesel place indicated two major components needed to be replaced. Fingers crossed that was the problem. Sorry dont have the paper in front of me now to tell you what the 2 components are.

 

[MakinDo]

So I have the new injector pump in along with the new injectors. Upon starting and idling lots of white smoke. Not sure what's going on. Each injector was cracked to check to see if that was an issue but no change. Also advanced and retarded injection pump timing but the smoke was still heavy. Any ideas? Faulty injection pump timing / flow on rebuild? An air leak somewhere in the system? The injection pump was installed with engine at top dead center and according to the alignment marks.
Thanks

 

[twistedtree]

How long did you let it run? Maybe it just needs to burn off residual fuel or assembly lube?

White smoke would be unburned fuel, or coolant. Once you have smelled both, you will always know the difference. Fuel will be acrid and make your eyes burn. Coolant has a very distinct smell, but I struggle to describe it.

 

[Dwk]

When starting our Perkins after injection pump rebuild the Engine smoked excessively. Twistedtree has a good point,it may need to run longer to burn off residue fuel. Beside the smoke how does it seem to run,start after the pump installation?

 

[MakinDo]

How long would you say. I ran it long enough to get up to temp

 

[KnotYet]

How long would you say. I ran it long enough to get up to temp

Talk to the shop(s) that rebuilt the IP and injectors and report what they say.
Please.

 

[MakinDo]

One part for the injection pump took a while to come in, maybe they forgot to do a calibration step

 

[jleonard]

It was many years ago, but when I had the pump rebuilt on my old T6.354 there was no smoke upon initial start-up. Other than the "typical" start-up smoke.
As soon as the engine temp reached 140 F it was clean.
What was very noticeable was how the engine sounded after the rebuild.

 

[sunchaser]

It was many years ago, but when I had the pump rebuilt on my old T6.354 there was no smoke upon initial start-up. Other than the "typical" start-up smoke.
As soon as the engine temp reached 140 F it was clean.
What was very noticeable was how the engine sounded after the rebuild.

Jay
+1
We just had a Bosch rotary injection pump rebuilt and like you noticed how much smoother and quieter the engine was. No smoke or smells upon startup either.

 

[STB]

Make sure your fuel valves are fully open and that you have fully bled the system. While it is clearing air out of the system it runs like trash. But this phase will pass within a very few minutes, at worst, as long as there are no major suction leaks. How long did you let it run run?

Also....did you get the copper rings out of each and every cylinder before adding a new, fresh one under eaxh injector. If you didn't get the old one out and add a new one, you've changed the height of the injector and the spray pattern, so it can run badly.

Also, when cracking each injector, one at a time, the expected result isn't "no change". The engine should run noticably worse with the injector cracked.

If one cracks an injector and it doesn't run noticeably worse, either (i) it isn't cracked enough, or (2) there is a problem with the cylinder associated with the injector, most likely a bad injector.

So, you want to loosen the supply (not return) enough for the fuel to squirt out and the injector to not pop. Have a rag ready.

The basic goal is to let the engine run with and without the injector to confirm the contribution of the associated cylinder is normal and isolate any bad cylinder(s) otherwise.

 

[MakinDo]

Thanks. I have the feeling I have residual air in the system from it sitting for a while when I was chaging out parts. Hopefully it's a relatively easy find.

 

[STB]

If that is the issue just let it idle for a bit. Once running it will bleed itself.

 

[MakinDo]

Still getting alot of white smoke. My injection pump does not have the inspection plate to really hone in on the timing. Any ideas, what am I missing

 

[STB]

There are a bunch of adjustment you can get to on that pump, but there really isn't a "turn down smoke" knob. The problem isn't likely to be an adjustment, especially since you just had it rebuilt.

Can you post a photo or short video of the smoking?

Is there any slicking of the water? Does it settle in or rise right up and disappear? Is there any blueness to the color?

I'm trying to get a sense of how bad and if it is smoke or maybe coolant and if there is any oil in it.

From my earlier post and not addressed in the forum...

Also....did you get the copper rings out of each and every cylinder before adding a new, fresh one under each injector. If you didn't get the old one out and add a new one, you've changed the height of the injector and the spray pattern, so it can run badly.

Also, when cracking each injector, one at a time, the expected result isn't "no change". The engine should run noticably worse with the injector cracked. Is this the case?

If one cracks an injector and it doesn't run noticeably worse, either (i) it isn't cracked enough, or (2) there is a problem with the cylinder associated with the injector, most likely a bad injector.

So, you want to loosen the supply (not return) enough for the fuel to squirt out and the injector to not pop. Have a rag ready.

The basic goal is to let the engine run with and without the injector to confirm the contribution of the associated cylinder is normal and isolate any bad cylinder(s) otherwise.

 

[MakinDo]

The smoke has no blue in it. It generally rises up and drifts away. There is no slick on the water from the smoke.

With the injectors cracked the engine does run worse but the amount of smoke remains relatively constant.

 

[STB]

So....that is interesting information. Diesel fuel and coolant/water can both be white smoke. When it is diesel, it is unburnt diesel (vs grey or black smoke, which is incompletely burnt diesel).

Diesel tends to hang around longer and lower than water/coolant, which tends to rise upward more quickly and dissipate a little faster.

White diesel smoke tends to slick the water if there is enough of it, whereas even a lot of water/coolant won't do this.

Water tends to have no smell, but you can sometimes smell a sweetness to a 50-50 diesel mix. By contrast, diesel tends to smell like diesel.

Water/Coolant tends to be easier on the eyes, whereas diesel can sometimes cause one's eyes to burn a bit.

If we are seeing an huge amount of white smoke, which is rising quickly and dissipating, without a correspondingly large slick on the water, and especially if the smoke doesn't smell/taste/eye-water like diesel, I'm betting it is coolant.

I'd get a boat neighbor or truck driver who's been around diesels for a while to assess things.

Usually that much water is caused by a head gasket issue.

You can try use a test kit like this one to test (Note: I don't have this exact one, just something that seems similar):
-- https://www.amazon.com/dp/B08FHMW6B1/ref=syn_sd_onsite_desktop_33?ie=UTF8&psc=1&pd_rd_plhdr=t


As another test, you can check for exhaust gasses in the coolant using this, which I do have and works well:
-- https://www.amazon.com/gp/product/B000NPDL76/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

Of the two, for the situation you have, the pressure tester is probably the more likely to be more useful.

 

[MakinDo]

Cold start smoke

Cold start smoke

 

[STB]

Cold start smoke

I don't have any pictures handy....but those pictures are pretty typical.

 

[MakinDo]

Did you see the pictures I posted. Waiting for the engine to warm up now so I can send some warms ones

 

[STB]

Yep. Those don't look abnormal. Some of those engines do run cleaner. But cleaner isn't typical.