Technical Oil Smoke on Startup

Currently reading:
Technical Oil Smoke on Startup

BenjaminU

New member
Joined
Apr 22, 2022
Messages
10
Points
1
Location
Windhoek
Hi there. Long time lurker here.

I have been running a Fiat Bravo 1.4 T-Jet (110kw version) for a while (read: sine 2008). I am posting here, simply because there are a few vehicles with this engine in the Fiat range.

Through the last few years, and by virtue of downright horrible service centres, the engine ate a few turbos (It's on number 5 now). While I would have been tempted to absolutely bin the car in late 2020 when Turbo #4 went, my wife loves the car.
So a mate (an ex fiat mechanic) and myself spent the holidays completely disassembling the engine and replacing everything that looked remotely dodgy. While opening it, we noticed years of carbon build-up, and had the whole engine acid bathed, sandblasted, refinished, etc.
Virtually everything is new:
Gaskets, spark plugs, water pump, oil pump, most valve stem seals (rookie mistake - ordering 8, instead of 16 seals), turbo, oil filter, fixed a very slight starting crack in the exhaust manifold, Timing, Fanbelt, etc. Safe to say it was a somewhat lengthy project.

The parts we didn't replace were the hard plastics around the top, as well as the bearings (since they are fine), the Manifold oil separator and those 8 valve stem seals (we took the best of the 16 we took out for that purpose).

Since then, the car ran fine for over a year.

Lately, it has been smoking in a weird way, though:
Nothing when the car is all cold and just staring.
Oil smoke as the car comes up to temperature (roughly 1-1.5km into a drive), on idle and thereafter when pulling away from a stop (normal acceleration, no flat-footing). It's quite a LOT of smoke, but it stops once the car is fully up to temperature.
Oil consumption is quite high on the car, as a consequence.
Now - I have two suspects here:

Option one is the Crank Oil Separator box... that was NOT replaced, and is a quick replacement (albeit a surprisingly expensive part). I did once try cleaning it, and chunks of carbon came out, but that didn't change much. It may be that the diaphragm is just shot. My logic here is that, as the engine runs, it does its job, but deposits all the oil on idle, due to the vaccuum not being strong enough to overcome the spring tension. There MAY be an underlying issue here, insofar as that there may be an actual vaccuum leak somewhere, but on visual inspection I don't see where that would be.

Option two is the one I am dreading, being that one of the valve stem seals gave in (probably the old ones). Logic applied: As the oil warms up a little, it gets into the combustion chamber on idle. At higher RPM, that seal is compressed too much of the time for oil to leak in there, and at temperature, the seal expands a little.

Is there ANY way that the smart people in this Forum can think of to test for the options, ideally stopping short of "replace one, see if the problem persists", since that could become very pricey (and I frankly don't have the time to attend to it myself right now, so I would need to hand the car to an actual mechanic if it's more than the oil separator or a hose)?

Thanks in advance.
 
Update:
The Air/Oil Separator (Segregator - Part Number 55208531) has been replaced.
While that offered SOME relief for the first few days, the problem persists.

Further observations:
1) It only happens while the engine is warming up.
2) It happens mainly on idle, occasionally while slowing dramatically and gearing down.

To me, this suggests a vacuum issue somewhere, but it could also be an issue of excessive crankcase vapour when the engine is cold.
Any thoughts in this regard would be appreciated, since it seems most of the remaining checks would involve taking the engine apart more substantially to even start looking for the problem...
 
Further update (and mainly an update to maybe assist anyone that may experience similar issues):
Time was spent with various spanners and parts over the past few days:

Segregator was pulled off again, and I prodded around the hole where it pops onto the engine. Made sure there were no obstructions (some oil gunk, but nothing that I would be concerned about). I also checked that oil dropped into that flows through to the block. It does.
Oil was changed, Oil Filter changed (and appropriate cussing at the location of the oil filter, the 27mm spanner not fitting properly, etc).

Water was topped up.
Now: It still smokes, but does it a full minute earlier (making it 30 seconds into the morning commute, rather than 90-120 seconds in), and less. It also STOPS earlier (by 45 seconds in, where previously it hadn't even started smoking, it's stopped).

Safe to say I am absolutely puzzled by this. I was thinking cylinder rings (i.e. real blowby), since i checked back on my shopping list from the rebuild, and the valve stem seals are all brand new. But that would not start and stop at the same spots.
It also doesn't seem to be temperature dependent, since the timing of this is not engine-warmup dependent (or at least, based on the timing of the smoke and how that changed, I don't see it as very likely).

SO here we are... *shrug*
 
Of course this could be quite a number of things - and turbo seals are always suspect? - But a good place to start is with a compression check if you suspect ring/bore wear might be a problem, and it's a very quick and non invasive procedure so you've lost nothing if it doesn't produce a definitive result. If the simplest form of compression check turns up some anomalies I like to follow up with a "wet" compression test which will tell you a lot more. Best of all though is to do a cylinder leak test - again, non invasive but a bit more difficult to perform at home as you need the gear to do it (compressed air and a cylinder leak tester for starters).

Here's quite a good video showing you the procedures:

Of course this is a V8 so the engine has two banks of cylinders so the layout is a little different to our straight 4 engines but the procedure is the same. There are several details about his test I'd comment on. Firstly, as you can see, this is an old "muscle car" with a carburetor so his plug colours are going to look darker than our more modern fuel injected weak running engines so our plugs typically will look "whiter" on a "good" cylinder. Also, when it comes to the leak testing, he is able to very easily remove the pushrods due to the rocker design so that he can ensure the valves on the cylinder being tested are closed. Not something you can do on most European designs of engine (although you could have on the old Vauxhall Viva - anyone remember them? - especially if it's an over head cam design as so many are these days, so I prefer to very carefully locate the piston on the cylinder under test at TDC on it's compression stroke and locking it in place by selecting top gear with handbrake fully applied and wheels chocked. - When you introduce the compressed air, it will try to turn the crankshaft unless you're very precisely at TDC so keep your fingers out of the way of fan belts etc! - It's important that the valves stay closed during the test. I also think it's important to carry out the test with the piston at TDC because that's typically where maximum cylinder wear is to be found and thus ring gaps will be at their maximum. If you allow the crank to rotate then the piston will be further down the bore so you are not taking the measurement at the point where max wear is likely but also the cam may have turned enough to be very slightly opening a valve which would make a nonsense of the test.

By the way, when you're listening for the hiss of the air escaping (exhaust pipe for exhaust valve, throttle body venturi for inlet valve, oil filler hole for piston ring blowby, it makes things much easier to stick a bit of flexible tube - fish tank or fuel pipe tubing works well - in your ear and stick the other end where you want to listen. Saves you getting nagged by your better half later due to her noticing your filthy ear when you're out at the Pizza parlour with the family!
 
Last edited:
In addition to the above - regarding leak down testing - Don't know if this is of interest to anyone, and of course it's only of use to someone who has a source of compressed air, but thought I'd pop it in here anyway.

In the garage I had access to a leak down tester - you can actually buy a "budget" brand quite cheap online if you think you'll use one often enough, for instance: https://www.onbuy.com/gb/uk-petrol-...feizI_GlIlgAw--MAMQck3lpvRkJ1s3BoCXVcQAvD_BwE However the number of times you're going to need one at home is not going to be often so the money may be better spent on something else.

However there can be times when one would be very useful, especially if you are trying to perhaps cost a repair for a relative or advise a friend on how much repairing their engine might cost. It's likely that in this sort of situation you will have done a "dry" compression test, followed up with a "wet" test if the dry test went badly. Perhaps you've found one - or more - cylinder/s not giving a good result especially if the wet test gives an increased figure. In this instance you know that probably the bore &/or rings are worn but a wet test tells you nothing about how the valves are sealing. This is where being able to listen for the "hiss" is invaluable and you can do this by making a simple adaptor for your airline. Just take an old spark plug which fits your engine, smash out the ceramic and centre electrode and weld or braze on a short connector to which you can screw an airline connector. Here's one of the ones I made (I have a couple for different plug sizes):

P1100322.JPG

In fact you'll notice it's got one of the "old" PCL connectors on it which I used with my old compressor. Need to change it for a "euro" connector so I can use it with my new compressor.

Then, with the cylinder you want to test at TDC on it's compression stroke, screw the adaptor into the cylinder and plug the airline into the adaptor. Watch out, just as you would with a "proper" tester, in case the crankshaft rotates when the air is connected. Now you have a cylinder with inlet and exhaust valves closed so a listen up the exhaust pipe and down the throttle body/air intake should immediately establish if you've got valve problems too. You must remove the air filter because the air filter element acts as a very good silencer of the hiss!

Of course because you don't have the leak down gauges it's much more difficult to use this to diagnose blow by the pistons because you'll always hear a hiss at the oil filler - no engine has rings that seal 100%. However, having done the dry and wet compression tests you'll already have a pretty definitive result for blow by anyway. If you do a lot of leak tests you'll get to know how much of a hiss from the oil filler is acceptable and when it's hissing too loudly but it takes a lot of experience to do this and I doubt if my ear is now "tuned in" sufficiently (after a good 25 to 30 years away from the shop floor) to do this myself any more - I wouldn't trust myself.
 
Firstly, thank you for your responses. They are extremely helpful.


Another wee thought. Reusing valve stem seals - you say you only renewed some of them? Did you put the new ones on the inlets or exhausts? If the old ones went on the inlets that just might be problematic?
I was wrong about that.
All valve stem seals are new (a year old). The Tappets were only part-replaced, keeping the old ones on the exhaust side.

Also, my assessment that it was temp driven must be way off. Late last week, after posting the above update (claiming that it smokes earlier in the drive), it didn't smoke until a good 3-4km into the drive, which is a lot later than it did before.
To me, this means it's not temperature driven.

It also stops after a while, which just makes it double-weird to me.

A former-mechanic mate of mine recommended that I should just run the motor properly warm (so some longer road driving) at some point, but I currently cannot afford a engine rebuild, and don't want to risk cylinders climbing out of the block or some such occurrence.

What is perplexing to me is how inconsistent the smoking is.
 
Last edited:
just a thought, but pull the turbo piping at a low point and see if there's liquid oil lying in there. My old Cordoba 1.9tdi had a problem with that which would sometimes make it smoke at random times. When I spotted it I thought at first it was turbo seals but it turned out, thank goodness, to be a breather problem

Once again, thanks for the reply.
This morning, for example, it was not smoking at all.
When it does smoke, it smokes like a fully broken turbo (just that it does it off-throttle, where a blown turbo tends to do that on throttle) - think THAT MUCH smoke.

As I have it, the "breather" on this engine is one part, being the segregator - that's brand new and working. I tested that oil runs off back into the crank case as well.

What I did notice is oil traces in the air piping going over the "far side" (cabin side) of the engine cover. The hosing there is also in quite a bad way, making me think that (regardless of whether that is or is not a contributing factor) it's probably a good idea to replace that.

I also have a spare turbo gasket, so I may see if I can find any leakage there and possibly replace that (though I doubt that - again, a new part, same as the turbo - less than 5k km on them). The Turbo seals themselves could be (would surprise me, but that doesn't mean it's impossible), but these turbos are notoriously crappy to service in any way, so it would likely be cheaper to just replace the whole turbo.
 
Hey, I have the same issue with my 110kw t-jet, but only during the colder months: when its < -10 degrees. It does the full smoke eating oil maybe twice during the winter randomly. After it warms up its running fine, no issues during the summer at all. I changed the segregator that had a bit of water inside for a new one with no big difference. I do mostly city driving and short routes and It might also get a bit more water condense. My air hosing is also cracked a little, but I didnt get to change those yet. Following the thread.
 
So - an update:
After some more checks, the following was additionally discovered:
No power past, say, 110kph.

So the following was done:
Cylinder Pressure tests (dry, not wet - yet): came back well within the spec.
Oil pressure tests (since the engine had massive carbon buildup once upon a time I felt this prudent): Within Spec, at 53 PSI warm, 60 on startup.

Turbo seals were dropping oil into the intercooler and the exhaust side, mostly while the oil was colder (hence the settling).

so:
New Turbo, new air intake pipe (the old one was getting cracks), cat had to be replaced.
While we were at it, new front shocks and brake pads, as well as the one engine mount that was showing age.

All of that, only to start the car and be greeted by plumes of smoke, which won't settle anymore. Fun.
Next step, then, is pulling the block, and checking basically everything:
Valve stem seals, Piston rings, Top gaskets and checking for cracks in the block.

Total price of that - don't ask, but less than anything even remotely comparable would be to buy.
 
Has the oil drain from the turbo been replaced?
Some replacement turbo suppliers will not give a warranty unless both oil supply and drian pipes are replaced with the turbo.
You say there was some oil residue build-up, and I would expect this to happen in the turbo drain too. If it is restricted, oil will leak from the turbo into the inlet or exhaust, as it has to go somewhere. Oil in the inlet will be sucked into the engine, and I'd expect this to be on startup, gradually reducing. Oil into the exhaust will wait until the exhaust is hot enough to burn it, then it will produce smoke.
 
Has the oil drain from the turbo been replaced?
Some replacement turbo suppliers will not give a warranty unless both oil supply and drian pipes are replaced with the turbo.
You say there was some oil residue build-up, and I would expect this to happen in the turbo drain too. If it is restricted, oil will leak from the turbo into the inlet or exhaust, as it has to go somewhere. Oil in the inlet will be sucked into the engine, and I'd expect this to be on startup, gradually reducing. Oil into the exhaust will wait until the exhaust is hot enough to burn it, then it will produce smoke.

Yup - both Oil lines into and out of the Turbo are clean and checked to be working properly.

The thing smokes right on startup (unlike it used to prior to everything), and just continues - it doesn't settle anymore.
 
Valve stem seal leaks will usually change with conditions, more on the overrun when inlet vacuum is higher. Consistent oil smoke would tend to be piston rings, or cracked head or block, or leaking head gasket, where the oilways pass through, although that would normally only affect one cylinder. A spark plug check might show if one or all cylinders are oily.
Turbos bring another oil source. Leaking seals on the inlet side will pour oil into the cylinders, and this would fill all the intake pipework with oil, and the intercooler, as the oil will only pass to the engine as a mist, so only when hot. Burning oil will contaminate the catalytic converter. The exhaust side seals are more prone to failure, as they run a lot hotter, and will dump oil directly into the exhaust, where of course it burns. So the smoke might not be the result of any engine issues.

After a run, once it has all cooled, I think you then need to be able to examine the turbo. The inlet side is of course easier to view, but I think the exhaust is more likely to be the issue.

With so many failed turbos, the cause needs to be found and addressed. Turbos need a good quality oil, changed at the correct intervals, the car's history before you bought it could have contributed to the original failure.
Replacement turbos will only be as good as the reconditioner.
Turbos should always be allowed to spin down before turning the engine off, let the engine idle for a few seconds after arriving anywhere, use the time to turn off lights, wipers, heater fan, etc. Never 'blip' the throttle before turning the engine off. A practice from years ago, that was supposed to help the next start, which is nonsense, but with a turbo, it spins it up, then the oil supply is cut off when the engine stops, leaving a spinning turbo on dry bearings. If the journey has been a fast run, with the turbo hotter than normal, let the engine idle for a couple of minutes.
 
So: The turbo is brand new.

We know why it ate turbos in the past few years (oil gunk and carbon buildup everywhere - that was resolved).

The last turbo (lasted 2 years) was working harder than it should due to a catalytic converter that was, simply put, done (understandable after that many problems). That was leaking oil out the seals on both ends, and has been replaced.

We've been careful on turbo-practices for the last few years.

I have literally had 5 mechanics from all walks of life look at this car by now.
It's simply time to pull the block and see what's what there, frankly.

In effect, at the stage where we are, it makes no sense (from what I can see) to run further tests, since we're pulling the block inevitably.
If anyone can think of any other tests that could be done that could point to a potential fix that doesn't involve ripping the block apart, I would be most interested.
 
Hey Benjamin, you're not the only to face this issue. Pretty much similar issue and same diagnostic procedure has been done. The result is same as yours.

In between for some time it had stopped and it has started again..

Have couple of culprits in my mind
1. Turbo oil return line
2. Second ring or the compression ring being bad/loosen.
3. The obvious oil segregator (changed many but yields different results which eventually ended up being same issue)

So let me know what is current status of your car.
 
So: The turbo is brand new.

We know why it ate turbos in the past few years (oil gunk and carbon buildup everywhere - that was resolved).

The last turbo (lasted 2 years) was working harder than it should due to a catalytic converter that was, simply put, done (understandable after that many problems). That was leaking oil out the seals on both ends, and has been replaced.

We've been careful on turbo-practices for the last few years.

I have literally had 5 mechanics from all walks of life look at this car by now.
It's simply time to pull the block and see what's what there, frankly.

In effect, at the stage where we are, it makes no sense (from what I can see) to run further tests, since we're pulling the block inevitably.
If anyone can think of any other tests that could be done that could point to a potential fix that doesn't involve ripping the block apart, I would be most interested.
So started the car today and apparently no smoke . yesterday it was smoking on the first/warm up start.

Somewhere I feel like enough of this, check the rings and get over it.. But how is that car making good power if I blame rings the culprit..

Been fiddling around the segregator area. Make sure to tighten the bolt which hold the seg box on the valav cover.. The one particular bolt which holds the manifold , seg box, and goes in the valve cover..

Looking forward for your response.

You're not alone :D
 
Back
Top