UKC

Probably not going to get anywhere but I thought it's easier to post here before I start to trawl car-specific forums...

I've recently had a problem with my car (2004 Fiat Punto, 1.2 litre, 16 valve) in really wet conditions. That is, when it's really hammering down and in particular when there's a lot of spray coming up off the road it start to judder a bit (as though something's wrong with the timings?) and loses power (it will just about maintain 60mph on the motorway). The yellow "check engine" light starts flashing but doesn't stay on (which is annoying - if it stayed on I'd be able to get a fault code read).

Anybody else experienced this? Is there a quick solution? My own (clueless) assessment is that it's water getting into something electrical in the engine.

In reply to Nick Russell:


If the engine light is coming on and the engine is loosing power, I'd probably say the same. Likely to be the OEM seals one of the engine sensors (TPS, MAP etc) or ECU connection. Where is your ECU located?

One check for this would be to wrap all engine sensor and ECU connections well with insulating tape when the engine is dry ensuring to seal to the wire. If the problem goes away then you know its one of those... can unwrap one at a time to see which one it is. If that doesn't solve it, I'd look at the ignition circuit (low voltage side) for the same problem.

In reply to Nick Russell:

Cracked distributor cap or coil?

In reply to sleavesley:


Don't think a car of that age would have a dizzy... though don't know the car specifically so could be wrong. But if that was the case, it probably wouldn't register on the ECU so unlikely to give a warning light.

In reply to crayefish:

yea unlikely as you say. Sounds like symptoms of what you have described already or the coil pack breaking down <or HT leads> due to moisture in the engine bay

In reply to wilkie14c:


I doubt it will be any part of the HT side of things as the ECU will have no feedback about that and thus no warning light. I am not even sure the 12V side of the coil pack would produce a warning light either as it's an output, not an input to the engine. But with how complex modern ECUs have got, I could be wrong on that one.

In reply to crayefish:

F in Law has a clio of the same vintige as the OP, its on its 3rd coil pack now! Fortunatley they are only 45 quid and that incs the ht leads. It does exactly as OP describes, rough running <esp in 1st> and goes into limp home mode. I think you are right though, it ECU won't get a signal saying 'coil pack problem' but the lambda sensor will report incresed CO2 cause of unburnt fuel and the ECU will try and deal with that.
OP - It used to be easy to recreate these problems with a misting sprayer, you know from the garden centre or hair dressers? Spray mist onto various components while engine is running and see if it turns lumpy. You need to do this in the dark so you can see any arcing. Keep your limbs off the car while doing it!

In reply to wilkie14c:

Good point! Forgot about the lamda sensor.

All this is exactly why I prefer to build all my engines with carbs So much simpler for my brain!

In reply to crayefish:

Gimmie a timing strobe anyday!
Modern engine men are software interpreters/electronics engineers these days with main dealers adopting the easier practice of replace until you hit the jackpot. Not much home mechs can do thesedays with modern fuel injection. fortunatly it doesn't go wrong that much.

In reply to Nick Russell:

My car (a 10 year old Rover 25) had similar symptoms a year ago, a cylinder started misfiring (that's the juddering and power loss plus flashing engine light) when wet or cold. Got progressively worse and also wrecked my catalytic converter as I left it (unburnt fuel + expensive cat = broken cat) - this will happen to you too if you don't fix the problem.
Check which cylinder is misfiring by using a OBD2 reader (can get them off ebay for about £15-30) and swapping the ignition leads and coils around while reading the error codes for which cylinder is misfiring. If swapping makes no difference try swapping the spark plugs to see if it is a dodgy or misaligned one of them.
For my car it was one of the HT leads available for £20 from Halfords (the catalytic converter was another £200 though and would have been more if I hadn't been able to fit it myself).
There are other things that could be causing it, but these are the common problems which are cheap and easy to fix with a bit of knowledge and a Haynes manual.

In reply to Nick Russell:

Even though the light is going off, it should have stored a code.

As Wilkie says, most likely a coil pack, common fault on Puntos! No easy way to check with out a code reader though.

You could try liberally spraying the electrics with WD40 as a quick fix, it will displace any moisture, stop tracking across a coil etc.

Cheers Adam

In reply to Nick Russell:

If it is a coil or ht lead problem then it should record a fault. Looking at the plugs could provide some ideas, look for differences in pairs as you'll likely have a pair of coils with each one driving 2 plugs. I'd say that the lambda sensor is probably a good place to start looking as is gets all the spray and repeated heating and cooling of exhaust components can kill insulation materials.

In reply to wilkie14c:

Agreed! My new race engine I'm building has quad downdraft dellorto carbs. Now that's how an engine should be tricky to set up but demon when it is.

In reply to All:

Thanks for some useful responses. It gives me a few ideas and I'll have a poke around.

In reply to wilkie14c:


It is especially bad at low revs, though I'm not sure about just first gear. Started it up yesterday and it was really rough in idle until it had been running for a few minutes. No warning light at that point.


This kind of fits actually. The rough running often occurs first, then a minute or so later the light starts flashing. I assume the flashing light is indicating 'limp home mode'?


I took it through the car wash yesterday. It didn't do a good job of washing the car (next time I think the money would be better spent on a bucket and a sponge) but it did replicate the problem! I'll have a go with the spray bottle...

In reply to butteredfrog:


I'll look into getting a code reader - as somebody upthread says, about £15-30 on ebay, so not much. When I asked at the local garage, the guy didn't seem to think it would have stored the code (but maybe he was just reluctant to take a job where he might have to do some detective work so fobbed me off). Could it be that the cheaper readers don't have access to this kind of code?


If it works, this (or the electrical tape suggested by crayefish) could be a useful temporary measure. I'm driving to Belgium this weekend and would rather not have it playing up there.

In reply to Nick Russell:

I reading your first post I was also going to say problems with the HT side of things.

My old rover used to have exactly the same problem, rough running in the wet and sometimes wouldn't start.

Drying out the inside, spraying the cables with WD40 and replacing the HT cables solved the problem.

You could see where it was shorting in the dark, in the rover's case the coil was right in the corner of the engine bay, the HT lead to the distributer had to pass all soughts of bits of metal, any one it would short out on.

In reply to Nick Russell:
If the light's coming on then it's a detectable fault, probably not the HT side of the ignition circuit which is what's traditionally vulnerable to water (it's probably got coil-on-plug ignition anyway which tends to stay dry).

Chances are it's a sensor misreading due to water in the loom, one that's needed for clean running of the engine but maybe not essential (depends how bad the mis-read is).

Maybe something low on the engine, crank position sensor for example.

Or maybe you're getting water in past the air filter (check it's seated right) and into the air metering sensor (usually mass flow and would be sensitive to contamination).

Either way, the ECU should provide fault codes. If it has loads stored then look for a loom connector that's leaking. If it's just one then look for that sensor and its wiring.

jk

In reply to Nick Russell:

A £15 to £30 pound code reader is only going to read generic EOBD codes. So would probably not show anything up. To give you an idea, our scanner (Snap-On Solus Pro) £3500 plus yearly subscription, has vehicle specific software and even this is not perfect, its just another tool.

Try this; run it till your engine is nice and warm and dry. With the bonnet up, mist the plug leads with water from a spray bottle. Does this create a misfire?

Plug leads are about £10, coil pack £25, WD 40 the plug leads, If you see an improvement, I would change the leads, (unplug them one at a time so you don't mix them up), see if this cures things.

If the leads don't cure the problem, change the coil pack.

Its either 2 packs, one for Cylinders 1 and 4, and one for 2 and 3 or one pack that feeds all 4 cylinders, whatever they are easy to change, held in place with allen screws.

There is a very good chance that changing the leads (and possibly coil pack) will cure your problem. Its the commonest fault. All for the same prices as a cheap code reader that might not tell you anything.

Cheers Adam

In reply to Nick Russell:

If the MIL was on it should still store a code(s), but some cheaper readers don't pick up historic/intermittent codes. If you have a reader that hasn't picked up the code it could be worth recreating (with a hose) and logging the code then. The PCM should store a code for the LT side of the coil.
If you can't read the code(s) it might be worth disconnecting the sensors and looking for corrosion on the pins.
Depending on how lumpy it is depends on which sensors I would start with, it it is a constantly lumpy feeling I would start with air flow/temp sensors, but it is more cyclical and sounds like a miss/partial fire I would go with the coils/injectors.

If you need a quick fix WD40 should do the trick.

Hope this helps

A modern engine man

In reply to jkarran:

I don't think its a sensor problem jk, OP says a definite loss of power, "struggles to maintain 60mph" that sounds like a miss to me.
Also a misfire would throw O2 sensor and switch the light on.

Its one coil pack bolted to the cyl head and short plug leads from memory.

To the OP:

Take care around the plug leads with the engine running. A tracking plug lead can give you a hell of a belt.

In reply to butteredfrog:


But any fault causing the engine warning light to come on could put the ECU into "limp home" mode possibly explaining the loss of power. Although limp home mode on a my ford restricts you to around 30 mph on the flat.


I agree with this.

In reply to gethin_allen:


Does it not lower rpm limit to around 3000rpm in limp home? Actually I can't remember a petrol ever going into limp home. Quite common with diesels as any fueling fault has the potential for a runaway engine, as engine speed is controlled by fuel flow.

A misfire would still run up to the normal redline.

In reply to butteredfrog:

I'm not really sure what 'limp-home' would mean or do in the context of a NA petrol engine: Open the fueling and ignition control loops. Perhaps limit revs. Set any active controls (Variable valve timing, variable cooling etc to safe states). For a forced induction engine I guess it additionally limits boost pressure one way or another.

I think my R1 motor has a 'limp' mode triggered by various missing non-essential sensors which limits revs to about 8.5k I guess more to indicate there's a fault than to protect anything! The details are hazy, it's a long time since I read the manual.


Maybe not under load, depends how badly it's missing.
jk

In reply to jkarran:
I can't really see what the point of Limp home on a petrol engine would be, protect against overheating under certain conditions? Like you say protect VVT etc.


Like you say depends. Brings me on to another thought, If it is a misfire its more likely to be a plug lead than a coil.

Nick says it struggles to maintain 60mph, a duff coil would drop 2 cylinders, a plug lead just 1, In a 1.2 Punto, that's got to be the difference between 60mph and scraping along at 30 in second gear with an obvious miss?


It should still rev under no load conditions, "limp home" would set an artificial rev limit,

In reply to butteredfrog:

To prevent under fueling of the engine. If the engine is run very lean for a significant amount of time, you get detonation due to the flame front not reaching the cylinder peripheries and thus it knocks. Then things start to melt.

Either way, it doesn't need to go into 'limp home' for the engine to loose power if a sensor is down. The ECU either simply reverts to an emergency map that can run without the sensor to get you home, or the engine runs rough because it has no idea of engine load (in the case of a MAP sensor) or throttle position (if the TPS).

I'd still check the ECU and sensor connections first because its FREE Always good to rule out the stuff that doesn't cost first. No point buying a coil pack if that isn't the problem, even though it sounds like it's likely for that car.

It should also be fairly evident if cylinders are missing rather than just running rough. Dropped cylinders have a very distinctive sound that will be there regardless of load or engine speed. Where as rough running will likely be worse at one end of the load or rev range depending on the ECU/sensor/problem causing it.

In reply to jkarran:
Could be anything to do with the accelerator pedal (very unlikely in this situation), throttle, losing the crankshaft sensor and having to guess using the cam shaft sensor, fuel pressure(more so with DI vehicles), the list goes on.
The severity of the limp home is controlled by the OEM and some PCMs have different levels of limp home for different faults. So it is guess work to understand if/which limp home it is in.
I would go for a simple approach of looking for water inside connectors (if you can't read the codes)

In reply to crayefish:

A backup map doesn't usually have an obvious effect performance though. In fact, the customer usually turns up at the garage when the fuel consumption drops, not when the light comes on.

Nick says he has an obvious drop in performance in V. wet conditions. This fault rectifies itself in the dry.

ECU and sensor connections are "perminent" faults, in that once you have water ingress into a plug, corrosion sets in and the fault will be present until you do something about it. Water ingress into an ECU plug would probably stop the car. To get to the ECU it would have to be a very deep puddle, the ECU is tucked behind the cylinder head on the Punto.

Loss of power when very wet, ok when it dries off, Engine lamp illuminated only when fault present, can still maintain 60mph just. It has to be a plug lead. A misfire in one cylinder isn't that obvious at higher rpms (until you unplug the knock sensor).

In reply to Nick Russell:

Possibly HT leads split. Had something like this on an old motorbike years ago. Ran sweet in the dry, a pig in the wet and was not obvious at all.

In reply to Nick Russell:
This kind of fits actually. The rough running often occurs first, then a minute or so later the light starts flashing. I assume the flashing light is indicating 'limp home mode'?

Just reread this, a flashing MIL should indicate misfire, assuming FIAT have set gone with the standard. So check probably worth checking the ignition circuits and coils.

In reply to butteredfrog:

Bridging of two or three sensor wires by water can lead to an incorrect signal and then the engine will run rough. You don't need corrosion to set in for that to happen. We had a similar problem with one of our race cars in the rain where the electrical guy forgot the rubber seals on one of the sensors.

In reply to crayefish:

Ah, what race cars? I used to work in TGP.

In reply to crayefish:

True, corrosion is a good guide if it has been happening for a while.

In reply to butteredfrog:

Cool. TGP is bike racing right? We ran motorbike engined cars. Not that powerful (100hp with restrictor) but at 200kgs they were pretty quick. I know naf all about bikes but done a fair bit of work on bike engines, including supercharging, turboing and other stuff.

In reply to dapoy:

Yeah good indicator of long term probs.

In reply to crayefish:

Thoroughbred Grand Prix, F1 Cars between 1970 and 1984.

We had an Ensign N180, Ex Tiff Needell. Originally developed and built with a budget of £80000.
An Arrows A6, Ex Marc Surer.
A Shadow DN8 (the one in Samson tobacco colours) We never ran that, beyond turning it over every so often.
Also two McLaren M10B's, Formula 5000 cars

Other stuff in and out of the workshop as well

The A6 is a nicely built car, the N180 is held together with cable ties, but still good for 200mph.

In reply to butteredfrog:








Oooh very nice stuff! Much higher class stuff than I have been into then What did you do there? I would have loved to have worked on cars like that.

Our cars were very basic in comparison. I started off in engines with my first team; had great fun supercharging a GSX R600... the torque curve from 5000 to 10,000 rpm was literally flat! Used to drop exhaust valves though. Since then I have ended up mostly designing suspension and drive-train stuff as got into FEA. One year did a lovely 550g upright and an 80g (or maybe less, I forget) rocker arm that I was rather proud of. We've worked to budgets of £25k but some teams had budgets 10x that size so we were never that competitive.

In reply to crayefish:

As you can imagine TGP cars are rich man's toys, and need a lot of TLC. I worked for a firm called SNE racing (defunct now). The main job was maintaining the two TGP cars, prepping them and getting them out to various circuits. The pit crew consisted of me and my mate, Steve the boss (but we never saw him anyway, it was a standing pit lane joke that Frank Williams could move faster) and a couple of helpers.

A lot of make do and mend, stripping and setting gear ratios, damage repair, engine removal, you name it really.

Repairs usually involved making some smaller components, anything bigger/complex we used a local engineers. Engine rebuilds twice a year, new engines as needed.

In reply to butteredfrog:

Yeah you'd have to be rich to get into that game. It's why I stick to off road racing when I am not in a team... it's one of the few forms of motorsport that I can afford!

What engines were you running typically? Were you using race prepped road engines or were they dedicated race engines? My experience has been with the bike engines on the formula side, then rover straight 4s and v8s for my car. Though I do have a 4.7ltr Porsche v8 that I want to put in... but not for another year as I can't afford the custom bell housing and ECU currently. Plus I need to finish making my new firewall for the rear mounted tank, coolant system for rear rad and wiring loom that has already taken over a year. I leave the loom for a few months while I work on other stuff and when I come back it takes me a few weekends to remember what the f*ck I did previously.

I've read all replies with interest, thanks again everyone. I've got a bit of time to look at it tonight, hopefully sort something out. To be honest, even if I do anything to it, I might not know if it's worked until I drive on a wet motorway. Of course, given the way this winter's going, that could be quite soon!

In reply to crayefish:

Never knew there were so many petrol heads on here.

I leave the loom for a few months while I work on other stuff and when I come back it takes me a few weekends to remember what the f*ck I did previously.
I found shadow boards helped me and saved so much time.

In reply to dapoy:




Stupidly I didn't think of that when I started! Doesn't help that I have a hugely complicated dashboard that with 12 relays, 16 switches (each circuit fused individually under the switch) and more fuel pumps than a space shuttle. lol