S40 1.8 2002 Hesitation & rough running

[Pete_c20]

Car
Volvo S40 1.8SE BiFuel (but symptoms same on LPG as petrol)
127K miles

Symptoms
Hesitation on acceleration sometimes nearly stalling. Power reduced. On acceleration a noise like a small blow in the exhaust is heard - sounds like a fluttering noise.
Excessive fuel consumption.
Long term fuel trim +5% rich
Code showing in Dice 'ECM-41 Knock sensor (KS) Faulty signal'. (See below)
Has never shown an engine check light for anything.
Passed very recent emissions test same as it always has
Spark plugs are a normal colour and are are all similar.

EDITED TO ADD: Idle is smooth and stable. Starting is very good.

Tests done
Compression test
DRY 175 178 178 165 (all in psi)
WET 211 215 218 200
No obvious vacuum leaks.
No obvious air leaks (I was looking for an air leak as the fuel was rich)
No obvious pressure at the filler cap either way with engine running. Oil is clean.
Lambda sensors before and after cat looked at. Both have an output that slowly oscillates between 0v and +5v the post-cat one is a low pass filtered version of the pre-cat one. I was expecting a sawtooth like waveform oscillating around the 1 Volt mark. I'm assuming this is correct as I've heard of some that have 0-5v output.

The ECM41 showing in Dice - from what the indy Volvo chap tells me, faulty knock sensors are very rare but this is a specific knock sensor failure code. So I've just replaced this with a genuine Volvo part but 2nd hand (as failures are rare ). The wiring has been metered from the knock sensor connector on the firewall back to the ECU for continuity and also shorts. Both the old and new knock sensors produce an output but I cannot check they're correct as I don't know what a good output looks like. Both are similar. The battery was disconnected and the power lines shorted out for a while after the replacement. No change.

Questions
1) Obviously, does anybody have a line on what might be causing the hesitation and excess fuel consumption?
2) What is this part? Volvo part # 08631228


The short stubby pipe comes from the top of the PCV system and the part mounts on, and the longer pipe projects into, the timing chain end of the inlet manifold. The 2 pin connector is driven by 12 volts straight from the engine management main relay. I'm assuming that it's a solenoid valve and allows the PCV system gases to enter the inlet manifold? If it is a valve should the electrical input open it or shut it? Right now it appears to be permanently open regardless.
In the S40 wiring diagram its down as '9/12 Crankcase breather' if I have the right part.

[Pete_c20]

Erm... I think I'm going to answer my own question on the part number 08631228 shown above. It's not a valve, it's a heated elbow. It draws about 1.2A on 12v and this means it'll dissipate about 14.5W, enough to get quite warm/hot. I'd guess this stops the gases condensing out before being released into the inlet manifold. I would guess that the main pipe lined in what appears to be copper helps this.

But still the original question - What causes the hesitation?

[Pete_c20]

I'd like to edit the OP and clean the thread up now, but cannot. So I'll add another post.

I mentioned the idle was smooth & stable. What I should of said is that the idle speed is stable but the idle is lumpy. There's an slight shudder to the engine and the exhaust pops. This has only appeared with this problem.

[villar4sotrondio]

Hi

I have a 2000 1.8 non GDI, 61k, with hesitation on pulling away. This is an intermittent problem. In my case, the car idles very well and power and fuel consumption are good.

I did clean he MAF with electric contact cleaner 2 weeks ago and although the car is now even smoother it has made no difference to the hesitation issue.

[DXMachina]

Could it be valve seat recession, perhaps? the original poster's issue that is...

Factory fit LPG cars diidnt have valve-saver and did tend to suffer a bit from poor LPG burns eroding away the valve mounts. If the exhaust valve isnt closing properly that could potentially cause higher fuel consumption due to lost compression

Not sure whether a compression test would show that or not...

[KBB]

A faulty knock sensor will cause the ignition timing to default to about 15 deg ATDC leading to hesitation and lack of power. The ECU will do this with the absence of a knock sensor signal to protect the engine as it can no longer detect pre-ignition.

With you scant tool if you can do the following it could possibly reveal some key info:

In live data call up STFT & LTFT, O2 Sensor B1S1 along with ignition timing. Note the readings at idle and again at 2500RPM held for 30 seconds to alow fuel strategy to stabilize.

I note your compression test results but to be honest that just proves the cylinders are sealing, it will not give any info on the engines ability to flow air. Additional test is what is known in the industry as the "dynamic test", as the name suggest its a running compression test.

Conduct the test on each cylinder as follows:

First disconnect coil to that cylinder if it is a DIS system ground the HT cable, Coil on plug simply disconnect. Disconnect electrical connector to the injector to prevent cylinder wash.

Fit compression gauge, start and idle engine on 3 cylinders. when engine is running bleed off the gauge and allow pressure to build note the reading.

With engine still running momentarily floor the throttle, note the new reading on the gauge.

Repeat the test on each cylinder and post the results. This test will give clear results as to the mechanical health of an engine when analyzed against you cranking compression results. Good to rule out mechanical issues before chasing fuel, ignition and engine management related faults.

[Pete_c20]

Some notes on the following cylinder tests.
In each case the open throttle test (momentarily flooring the throttle) was done for long enough for the pressure reading to stabilise, or roughly stabilise, whilst I was watching the gauge. It often took a second or two for the engine to actually rev, and when it did rarely was the throttle held open for more than a second to achieve the test.

Where a pressure range is noted this indicates a non-stable reading and the range over which the gauge flickered.

The ignition system is DIS (which I assume to mean distributorless ignition system). Two coil packs - One pack sat on top of cylinder 4 also drives cylinder 1. The other pack sat on top of cylinder 3 also drives cylinder 2. There is enough room on the looms to pull each coil pack up and out when doing cylinders 3 & 4, turn the packs upside down, and then reconnect their chassis earthing with a clamp. I found that when testing the coil packs on the bench that loading one spark plug slightly effected the other plug on the same coil pack. Also see further down the effect of unwanted arcing.

Although my gauge seems to fall back again when the pressure drops it is bled off between each set of readings. I'm wondering here if the non-return valve in the gauge is okay. See later for the remedy.

Testing starts from engine cold. All testing done on Petrol.

CYLINDER 1 - IDLE (cylinder next to timing belt)
160 psi reasonably stable.

As these tests are new to me, and I was waiting for the engine to warm up, I bled the gauge a few times and noted the readings that followed whilst it idled. As it warmed the readings dropped. The readings were never stable. It dropped and stabilised at about 130 psi. Also as this was the first reading I lifted the airbox lid and parted it from the air filter to see if it made any difference. The difference was barely noticeable. The airbox was closed back up. Engine is warm now.

CYLINDER 1 - OPEN THROTTLE
220-230 psi

After this high revs test and another gauge bleed the cylinder idled at 80-100 psi.

CYLINDER 2 - IDLE
80-100 psi

CYLINDER 2 - OPEN THROTTLE
220-225 psi

CYLINDER 3 - IDLE (Wet! I forgot to pull cylinder 3 injector drive out. Ooops!)
70-100 psi,
CYLINDER 3 - OPEN THROTTLE (Wet! I forgot to pull cylinder 3 injector drive out. Ooops!)
190 psi

Cylinder 3 test re-run.....

CYLINDER 3 - IDLE (now dry)
110-130 psi,
CYLINDER 3 - OPEN THROTTLE (now dry)
Peaked at 230 psi then immediately tailed off to 60-80 psi. Back to idle and it stabilised at 120. Due to this I repeated the test.

CYLINDER 3 - OPEN THROTTLE 2nd try (now dry)
peaked at 230 psi and then fell back to, and stabilised at, 205 psi.

At this point, due to the differences between these readings, my previous readings from the traditional cylinder compression tests and also the uncalibrated old compression tester being used, I pause to check the compression testers non-return valve function and also double check its gauge reading against something of much better quality. The non-return valve has a small intermittent leak and is replaced. The gauge is slightly out but only 5% max (not bad for a 15 year old Gunsons product).

Testing resumed and re-run/re-started.

CYLINDER 4 - IDLE (yes these tests run 4,3,2,1)
87-95 psi

CYLINDER 4 - OPEN THROTTLE
174 psi

The new non-return valve seems to have helped the readings there's less gauge flicker.

CYLINDER 3 - IDLE
80-90 psi
CYLINDER 3 - OPEN THROTTLE
138 psi

CYLINDER 2 - IDLE
50 psi (Yes, Fifty!).
Side note: I noticed (and heard) the unused spark plug connector arcing to the engine block during this. It was arcing through what looks to be the injection point for the plastic injection molding about half way down its shaft. When I've inspected these for cracks none were found and I still can't see one. This is a pinhole. In the bore I would say it doesn't matter due to the air gap, but lying on the block it shows up. 5 turns of electrical insulation tape didn't stop it arcing either. In the end I pulled out the HT lead at the coil pack end. After this couldn't arc out, this raised the cylinder pressure from 40 psi to a consistent 50 psi.

CYLINDER 2 - OPEN THROTTLE
Peaks at about 150 psi then rapidly falls to 75 psi. Back to idle and its the same.

CYLINDER 1 - IDLE
68-72 psi
Side note: Same arcing problem with the un-used HT lead as noted before, with it arcing this was consistently 50psi.

CYLINDER 1 - OPEN THROTTLE
Peaks high (uncertain as to how high due to the flickering gauge but round about 120psi) and then very quickly dropped to 0-15 psi (!). Returning to idle and its 50-60 psi

Due to the 1st reading this was repeated but the result was the same.

Even though the non return valve on the compression tester was replaced with a high quality one rated at 25bar/377 psi it still showed flickering readings. I assume down to the inertia of the non-return valve opening and closing, the pulsed air flow and small air volume beyond it.

I'm saying little about these readings until you've commented on them KBB. I have my own suspicions! The tests have been eye opening. I'm amazed at the variation in readings to the point of wondering about publishing the results!

Some proper ODB2 data to follow tomorrow as the software has landed and I've set it up tonight but it's too late to run tests at 2500 rpm in a quiet neighbourhood now.

[Pete_c20]

Originally Posted by DXMachina

Could it be valve seat recession, perhaps? the original poster's issue that is...

Factory fit LPG cars diidnt have valve-saver and did tend to suffer a bit from poor LPG burns eroding away the valve mounts. If the exhaust valve isnt closing properly that could potentially cause higher fuel consumption due to lost compression

Not sure whether a compression test would show that or not...

It's an interesting point DXMachina especially in the light of the above results. I asked this question of Volvo a long time ago when the car was new. They said as it was factory fit there wouldn't be a problem. I know that valve seats got harder with the introduction of unleaded petrol but wondered at the time was there any difference between unleaded 'hard' and LPG 'hard'. This car has been run 99.9% on LPG, looked after and not driven hard, and it's only done 127K miles so we'll see!

I'm late coming back to the thread as notification of a thread reply hadn't come through, probably due to me replying to my own post initially.

[960kg]

Have you replaced anything else for new, yet.......eg: given it a good service?

[DXMachina]

Originally Posted by Pete_c20

It's an interesting point DXMachina especially in the light of the above results. I asked this question of Volvo a long time ago when the car was new. They said as it was factory fit there wouldn't be a problem. I know that valve seats got harder with the introduction of unleaded petrol but wondered at the time was there any difference between unleaded 'hard' and LPG 'hard'. This car has been run 99.9% on LPG, looked after and not driven hard, and it's only done 127K miles so we'll see!

I'm late coming back to the thread as notification of a thread reply hadn't come through, probably due to me replying to my own post initially.

I can confirm that factory fit LPGs did not have valve seats hardened for LPG, they were identical to the Petrol versions and they did have to replace a lot of them under warranty due to serious erosion