Alfa Romeo Forum banner
21 - 40 of 125 Posts

·
Registered
Joined
·
109 Posts
Not sure where you're getting that 44°C from. Tested on car, engine cold and idling, MAF unplugged IAT is reading minus 39.8°C. Same measurement hot engine-idling, MAF unplugged.
Good luck with your problems.
 

·
Registered
Joined
·
225 Posts
Discussion Starter · #22 ·
I have mistyped from the phone. I read somewhere that the value was -45C. I wanted to know what happens before unplugging the MAF and giving it a go.
The car started OK this morning when I took the kid to kindergarten, so it seems that I have somehow managed to solve this issue.
 

·
Registered
Joined
·
225 Posts
Discussion Starter · #23 ·
No, as suspected, the issue persists.
1st start
Just seconds after that, the 2nd start was just fine
 

·
Registered
Joined
·
109 Posts
Have you tried swapping out your main injection relay?

Engine bay fuse box 3rd one back from maxi-fuses. Red colour.
 

·
Registered
Joined
·
225 Posts
Discussion Starter · #26 ·
Have you tried swapping out your main injection relay?

Engine bay fuse box 3rd one back from maxi-fuses. Red colour.
Will give it a try. Anyway I booked a visit to my trusted indie for next week, since I am unable to identify the issue and he said that it is worth to have a look at the fuel pump and some seal at the back that has a tendency to fail. Will keep you posted since this seems to be a tough one.
 

·
Registered
Alfa 159 2.0 JTDm Lusso 6sp.
Joined
·
6,931 Posts
My money is still on the fuel pump or pump relay.....
 

·
Registered
Joined
·
109 Posts
Well if that fuel pressure plot is with engine at operating temperature then is about right at idle (435- ish PSI). Your indie is talking about the internal diaphragm of the High pressure pump - a common problem on BMW N73.
 

·
Registered
Joined
·
206 Posts
Thanks that's really helpful. I am guessing that on cold start acceleration and running are both fine? I have scanned mine today from cold. New scanner and can't quite figure out how to migrate the data but I am reading 55bar which roughly equates to your KPA figure.
 

·
Banned
Joined
·
642 Posts
My money is still on the fuel pump or pump relay.....
Have you tried disconnecting the temp sensor and checking it’s value. It should be 2500 ohms~ at ambient.

Just changed mine and the difference is remarkable. It is associated with Valve Timing at start - up, but the absolute detail of how is not apparent from anything I have read. It is probably also linked to spark advance/retard,

If S/A and valve timing are wrong - it will be difficult to start the engine.

The changed conditions with my engine are difficult to attribute to a device that cost only 7 quid. So the extent of the NTC sensors influence upon the engine came as a bit of a shock.
 

·
Registered
Joined
·
225 Posts
Discussion Starter · #35 · (Edited)
Have you tried disconnecting the temp sensor and checking it’s value. It should be 2500 ohms~ at ambient.

Just changed mine and the difference is remarkable. It is associated with Valve Timing at start - up, but the absolute detail of how is not apparent from anything I have read. It is probably also linked to spark advance/retard,

If S/A and valve timing are wrong - it will be difficult to start the engine.

The changed conditions with my engine are difficult to attribute to a device that cost only 7 quid. So the extent of the NTC sensors influence upon the engine came as a bit of a shock.
Hey, Brian. Is that the Air intake temp sensor, which is integrated in the MAF, or the Coolant temp sensor?

BTW I am inclining towards the fuel pump version, since I noticed smell of unburnt petrol at the back of the car and 2 days is starts rough, as if it was starving.
 

·
Banned
Joined
·
642 Posts
Hey, Brian. Is thatthe Air intake temp sensor, which is integrated in the MAF, or the Coolant temp sensor?
As someone said, not likely to be air sensor in MAF, it’s just a diode and there is next to nothing through it.

I’m on about the cylinder head NTC sensor at cylinder 2 exhaust port (coolant temp) - just to the left of the dip - stick. If you pull the plug off and stick a DVM across it, it should read ~2k5 ohm.

It’s as close as they could get it to the Man - cats, which they need to monitor for emissions reasons. They probably do some clever software manipulation to calculate {guess - timate} the predicted temperature value of the manifold cats, from that sensor.

This is linked to getting the cats to “Light-off” quickly and maintaining them at light - off, after you slow from 70 to zero on British motorways in rush hour - and for urban driving.

How the hell they do all this is a mystery. But Alfa designed - in so many “Options” an EU funded study gave manufacturers to meet Emission Regs. Only Alfa seem to have used them all - which is why it got such high ratings. It also makes it difficult to diagnose when any bits go wrong.

But the device has such a fast changing temp/impedance characteristic, the ECU will be looking for a pretty damned accurate response, to mimic what the Softies have embedded in the program.

It is only recently I found out where it was, thanks to a bit of research and some belated, but non the less welcome information from the forum.

However, looking at the practicality of it all, I think I will be changing mine every two or three oil changes. Or as soon as it stops dripping water from my tail pipes on tick-over. Mine is less likely to blow now, as the man cats have gone. But for those who still have them, it is a very cheap way to eliminate it from the equation, particularly given how often and extreme the temperature changes are in normal use. Not saying for definite it is your problem, but it sure as hell was one of mine.
 

·
Banned
Joined
·
642 Posts
Returned from shopping - always a joy, particularly when there are zillions of mums and dads picking up the kids from school and coaches parked up everywhere to ferry the remainder back to the villages.

Grabbed a ratchet, extension and socket, and fitted the new NTC sensor - only took 3 - 4 minutes. I really didn’t know what to expect, other than perhaps engine running a bit quieter on start - up. I lost about a pint of Paraflu in the process, but the old one came out a treat. Mind you, so it should given the engine is a brand new rebuild.

Start - up ever so quiet - comparatively speaking - much more subdued. Engine quickly warmed and water droplets from the tail pipes, masking the heavy soot.

So has this NTC sensor always been faulty and has it been making life even more difficult to get the engine to perform better? The sensor came from the original 159 engine - I asked the garage to move all the sensors across to the new engine when it was being installed.

Not really knowing too much of the original engines history, other than having had its chains and tensioners done, it seems as if it was faulty. The front bank cat showed evidence of overheating as it had a large area of scorched metal on the front face. Could this have also caused the NTC sensor to fail?

It may well have done as on testing it intermittently showed a short circuit, an open circuit and occasionally a value which did not change with temperature.

Running conditions have changed and a few test miles should give some indication as to whether this is the last, or just one of a few remaining niggles. Bottom end torque is brilliant and she piles on the revs very quickly. But the racket it produced under hard acceleration just seemed a little too loud and I have long recognised it was due to over-richness at tick-over/low revs - Valve timing having changed dramatically.

From what I have read, the sensor temperature determines when the exhaust camshaft starts to retard, ie, when the port temperature is high enough to maintain ” Light-off” in the Man Cats.

If; as this one has, it fails, the camshafts won’t assume the correct angle, but how much error this creates over the full rpm range is unknown. As it invariable happens under acceleration, this must be at a time when the camshafts ought to be adjusting to new angles. So did any errors result in prolonged early exhaust valve opening, generating excess noise in the exhaust system? And noise levels only dropping as they approached the correct angles at cruising speed? The engine is very quiet and resonance-free at cruising speeds.

The sensor came from Auto doc - Berlin and I was starting to believe it would never arrive, what with all this EU/Brit tariff nonsense. However, it’s a Delphi Sensor - I fitted a set of Delphi coil packs recently - and they were recently bought by Borg-Warner. By the time postage and vat was added it was about 20 quid delivered - 7 quid for the sensor. The six coil packs cost ~ 180 quid, so I‘m pleased with the price.
 

·
Banned
Joined
·
642 Posts
High voltage devices do not like excessive heat! I remain convinced, wrt the issue of coil pack failures, Manifold - cats are the root - cause of them and a multitude of peripheral failures, which are wrongly attributed to other factors.


Temp in celsius Ohms kilo ohms

-10 0.00600 6.00

-9 0.00590 5.90

-8 0.00580 5.80

-7 0.00570 5.70

-6 0.00560 5.60

-5 0.00550 5.50

-4 0.00540 5.40

-3 0.00530 5.30

-2 0.00520 5.20

-1 0.00510 5.10

0 0.00500 5.00

1 0.00490 4.90

2 0.00480 4.80

3 0.00470 4.70

4 0.00460 4.60

5 0.00450 4.50

6 0.00440 4.40

7 0.00430 4.30

8 0.00420 4.20

9 0.00410 4.10

10 0.00400 4.00

11 0.00390 3.90

12 0.00380 3.80

13 0.00370 3.70

14 0.00360 3.60

15 0.00350 3.50

16 0.00340 3.40

17 0.00330 3.30

18 0.00320 3.20

19 0.00310 3.10

20 0.00300 3.00

21 0.00290 2.90

22 0.00280 2.80

23 0.00270 2.70

24 0.00260 2.60

25 0.00250 2.50

26 0.00240 2.40

27 0.00230 2.30

28 0.00220 2.20

29 0.00210 2.10

30 0.00200 2.00

Of course this does assume a linear increase.

Of course, the logarithmic characteristic of the sensor is linearised by the input circuitry of the ECU.

Details taken from a Toyota Forum. There may be differences in absolute values between theirs and Alfa’s but, essentially they all follow the same “Law” and it gives an excellent illustration of how sensitive these devices are to temperature. The ECU “must“ see this characteristic as a function of cylinder head temperature change to accurately modify the operating parameters of the engine - and swiftly.
 

·
Banned
Joined
·
642 Posts
A NTC device to my knowledge, is created by the junction of semiconductor material and a metal. Whereas historically, it was thought that semiconductors would “age”, as thermionic valves do, this has been found not to be the case. However, the junction of an NTC device can fail due to oxidation of the metallic element device.

Excessive temperatures however can also cause the semiconductor element to break down.
 

·
Registered
Joined
·
1,059 Posts
The issue described by the OP sounds like a problem with the in-tank fuel pump (not the high pressure one on the engine). The only way to test it is to tap into the fuel line and attach a gauge.
 
21 - 40 of 125 Posts
Top