Failing Ignition Module?

This sounds more like an air mass meter problem. Does the LPG system get a load signal from the air mass meter?

On Fri, 24 Sep 2004 08:26:01 -0400, the illustrious Mike F favored us with the following prose:

My similar problem was addressed by a new O2 sensor and hollowing out the catalytic converter (which was partially blocked). If the exhaust is only partially obstructed then power loss won't occur until higher revs (which require higher flow).

Lots of power at the top end now but still some pinging likely due to the timing being set by eye/ear. I don't know if unblocking the exhaust would call for a re-evaluation of the existing timing settings.

Maybe it's the old 'banana in the tailpipe', got any enemies?

Just kidding. blurp

[..]

Sometimes I think that 1987 in the UK must have been the dark ages. No O2 sensor, not catalytic converter.

The whole exhaust system is not much more that a year old, so I would be surprised if it was blocked. Also see my other post, that today the problem seems to be absent.

But you're right, a blocked exhaust can give an engine a tough time. I once reversed the tailpipe into a raised grassey bank and killed the engine stone dead until I managed to push it away from the bank and clear the pipe out.

It might do, but more likely require a change in the fuel metering. I'm a little surprised your knock sensor isn't taking care of the detonation, though. But as you can tell, these newfangled (1987) control units are a bit of a mystery to me.

Well now you mention it...

No. Petrol fueling is by basic K-Jet, and the LPG system is open loop, with a venturi mixer (like a very simple carb.) The only electronic wizardry is the EZ-K ignition system. The sensory inputs are - idle position switch, engine temperature, knock and manifold pressure. The only ones I have doubts about are the knock sensor or the temperature sensor. I'm pretty confident that a bad temperature sensor wouldn't create these symptoms, and my tests on the knock sensor (tapping the block with a steel bar whilst watching the timing marks) were inconclusive.

What I figure may be happening is that the ignition keeps on advancing, under load, beyond normal limits, until I back off the throttle due to knocking/backfiring. Even if the knock sensor was bad, this shouldn't happen (or am I wrong about this? Consider that provoking knock on LPG is much more difficult, so if the only normal limit to ignition advance was the knock sensor, then using LPG would result in ignition too advanced for good running).

However, today I decided to run it with the knock sensor disconnected. Went perfectly - couldn't provoke any knock or backfires under any conditions on either fuel. To confirm the faulty part, I reconnected it - still runs perfectly.

I supose it could be down to dirty contacts, but I'm still not really convinced that the knock sensor is the culprit; I have wondered if the hall effect unit, or power stage amp could produce these symptoms.

On Fri, 24 Sep 2004 18:06:04 +0100, the illustrious Stewart Hargrave favored us with the following prose:

Well you may not have a catalytic converter but I'm pretty sure the O2 sensor (aka Lambda Sond) was not optional as it's one of Volvo's bragging points. It should be screwed into the exhaust manifold and mine wouldn't come out without a fight. When it did come out it was clogged shut with a dense chalky residue (I'm told that's from burning coolant).

As for knock sensor, my car's a 1983 240 Turbo so it predates that radical technology. Additionally I believe the grage that replaced the timing belt got the timing wrong and I didn't recognize the "funny sound" as detonation and drove that way for months. Ugh.

Now the pinging is likely permanent but I won't know until my mechanic of choice gets a timing gun.

blurp

[..]

Well they didn't brag about it in the UK. Not until a year or two later. I was going to add one for the sake of making the LPG system closed loop. I had a boss welded to the pipe, but the welder positioned it so that the sensor would have pointed downwards underneath the car leaving about 2" ground clearance. Twit. Anyway the whole exhause system has been replaced since then, and I haven't got round to progressing that idea yet.

I hope not - you should be able to adjust the timing to prevent knock, unless there is something else amiss; badly coked up for example. I've seen the harm detonation can do to a piston. The local forces and temperatures involved can be very damaging and spell the end of an engine. Try and avoid it at all costs.

To me it sounds like it's leaning out (mixture too lean when the problem is happening). At higher engine loads the restriction provided by your LPG mixer and the air flap in the air flow sensor would make the effects of any vacuum leak worse. Did you perchance have the boot from air flow sensor to throttle body off while checking the knock sensor?

The ignition system is programmed with an ideal advance curve, based on engine speed and load. If knock is detected, timing will be retarded from this ideal to eliminate knock.

Weak mixture was my first thought. Running on LPG, it felt like I was running out of fuel. I checked the LPG system in close detail to establish that there was no blockage or somesuch, and also checked for induction leaks. I then discovered the symptoms were evident when running on petrol, too. I haven't taken the boot off yet (the LPG mixer sits in the top of it - have a look here if you're interested in the details

I'll do thattomorrow - I may even unbolt the manifold for a look-see just toeliminate it from the equation. A couple of weeks ago I renewed theinjector seals to resolve an uneven idle, so I've been pretty diligentabout looking for induction leaks.>

I figured this must be the case. Presumably, also, the control unit will detect the absence (or malfunction) of the sensor and retard the ignition a little to a 'safe' advance curve. Thinking on, then, maybe the knock sensor has now failed totally, and the control unit is responding with the safe curve, which is why there was no apparent difference between the disconnected and re-connected sensor earlier today. Is there a definitive way of testing a knock sensor?

ALL MY MANUALS ARE FOR THE 200 SERIES SO I DOUBT THAT WILL HELP YOU BUT HAVE YOU TESTED THE FUEL PRESSURE REGULATOR? ARE THERE ANY SENSORS THAT WILL MAKE THE INCREASED LOAD NOT CHANGE FUEL/AIR DELIVERY WHEN IT'S OBVIOUS THAT THE ENGINE BEING PUT ON AN INCREASED LOAD ie UPHILL SHOULD REQUIRE ADDITIONAL FUEL? IS THE ENGINE RUNING ON A DECENT COMPRESSION? THIS TEST SHOULD BE DONE TOO ESPECIALLY IF IT HAS A LOT OF MILEAGE!

LET US KNOW HOW YOU MADE OUT WITH MY OR ANYONE ELSES ADVICE.

I feel I may have wasted my reply to original post cause I didn't notice on my browser that there were several replies there already, if what I have written makes no sense please disregard it and accept my apology in advance Dan

1987 244 DL 196,xxx klms needs blasted new seats augh 1988 245 DL 237,000 Klms and damn it, it also needs at least one front driver's seat double augh it's leather too and costs a lot i'm sure last was big mistake but wife was scared of underbody rust on the 1987 so she urgently needed a car and settled on 91 Park Avenue piece of crap car lemon with green cap top too. but she likes the power in V6 3800 blarney to that I say

No probs, Dan. I posted it to get different perspectives on the problem.

As it is, the car is now running fine (for the moment, at least). I still have no real idea of what the problem was, and can only think there may have been a dirty connector that I've made good by pulling it off and pushing it on again (I'm not entirely convinced by that explanation, but I'm at a loss for any other).

Very interesting system.

The only way test the sensor is to tap the side of the block with a small hammer and watch what happens to the timing. You have to keep tapping, once the "knocking" stops, the timing is readvanced almost immediately.

Well I've replaced the knock sensor, because there was some doubt in my mind about it. I found the tapping test inconclusive on both the old and new sensors.

But the car has been running perfectly for some days now, with no recurrence of the symptoms it displayed last week. I've no real idea what caused this - I can't figure out a way to make the knock sensor entirely responsible.

Thanks for your help, anyway.