A strange but good thing has happened, more power!

[johnl]

For a while now I've been thinking my engine had become a bit 'doughy', just lacking verve and perkiness, slightly lethargic and lacking crispness of response. Felt like it had retarded ignition or maybe running way too rich, or something. On rare occasion it seemed to magically 'come alive', briefly, then it didn't, again. So, a couple of weeks ago I cleaned the MAF, which was a good thing to do, not that it appeared dirty. There was a noticeable improvement, but it wasn't anything to write home about (or post on a forum).

Anyway, now it gets more interesting.

For some time now the car has had a really minor issue that was no biggie and so no urgency to rectify (why I've just left it for some time). On start up the rear muffler would momentarily vibrate against one of the brackets I had made to fit the custom rear anti roll bar (because it had about 1mm of clearance between bracket and muffler). It didn't seem to do this while driving the car, though it may have done, just not so obviously.

A few days ago I was doing something at the rear of the car and had it up on ramps (playing with the rear ARB attachment, more in a later topic), so I took the opportunity to adjust the position of the rear muffler. All I did was loosen the clamp and wiggle the muffler a bit rearward to give a little more clearance to the ARB bracket.

What is interesting about this mundane procedure? Glad you asked. The engine started fine with zero rattle from the exhaust, as expected. As completely not expected, the engine seemed to now possess more power, and more torque. It just feels happier, more perky, more willing, crisper, more responsive, generally a 'sweeter' engine. It doesn't ask you to change gears all the time, it's happy to pull a taller gear lugging uphill on a light throttle without complaint and holding a steady speed (whereas before in that circumstance it would gradually lose speed unless you pushed the pedal a long way down, or changed gear). There is also an improvement in fuel consumption.

Why?....

I have a theory. Perhaps, just maybe, the vibration from the exhaust rattling against the ARB bracket was being carried along the exhaust all the way to the engine. At the engine the vibration might have been detected by the knock sensor and interpreted as a symptom of detonation? If so then the ECU might have retarded the ignition?

This might account for a loss of power just after start up (when the 'knocking' vibration was easily heard), though I would have assumed this wouldn't last long (for how long does the ECU retard ignition in response to a detected 'knock event'??). However, maybe the vibration did exist while driving the car, enough for the knock sensor to detect, but not enough for me to detect...

Whatever has caused this, I am very happy about it. The engine is now a sweet little gem and a minor joy, whereas before it was just a little bit disappointing...

Regards,
John.

BTW, how do I disable emojis?

[Citroënbender]

Is it possible a fragment of failing cat passed into an area where it caused less obstruction?

The two "repairs" which gave my '01 TS more poke (other than when the original cats suffered considerable internal melting/failure) were replacement of the crank sensor and replacement of outside spec ignition coils. Cats were the most noticeable improvement when addressed, although I would have said the car felt "soggy" rather than "doughy".

[johnl]

I don't know, but I doubt it.

I also don't know whether my notion that the knock sensor might be mistaking an external 'knocking' vibration for detonation is correct or not. What I do know is that there was clear change in engine behaviour from before to after removing the cause of the 'knocking'. It's a much happier engine now.

A melted cat suggests a problem not to do with the cat, i.e. the melting would be a symptom and not a cause of anything (other than power loss and emissions). Assuming this to be the case, then a new cat would also be at risk of melting. I suspect there would have to be fuel burning in the cat, which means there would have to be unburnt fuel and oxygen getting into it. So, maybe over fuelling accompanied by an air leak (sucking oxygen into the manifold), or, perhaps chronic misfiring (unburnt fuel and oxygen entering the exhaust). You said the coils and crank position sensor were bad, which might have caused such a problem?

Regards,
John.

[Citroënbender]

Yes, initial damage to the cats was from 1300km driving with waterlogged plug wells. It's part of a much longer story with the underscoring motif of pearls and swine.

[johnl]

Plugs underwater isn't a good thing, lots of lost sparks, lots of unburned fuel down the pipe, lots of burning fuel in the pipe, very hot cat...

So, I've done a bit of researching, and it seems that ignition retard due to extraneous 'noise' detected by knock sensors is actually a thing, called 'false knock'. I'd never heard of this before, and am quietly a bit pleased with myself that my diagnosis (aka a guess) seems not unlikely to be correct.

From what I read, exhaust rattling against chassis (or anything) seems a common cause of false knock. Piston slap and noisy valve trains can also cause it, apparently. Any part of the engine or gearbox (or anything attached to them or inside them) rattling against anything might cause retarded ignition and hence lost power and worsened fuel economy...

Regards,
John.

[Citroënbender]

The other one which was new to me a couple of weeks back, a heat damaged knock sensor (Mazda 2, incidentally). I doubt it's normally an issue but remembering Bazz's comments about the crank sensors ageing, it's sort of interesting.

[bazzbazz]

All sensors that use a magnet (Hall Effect ect) are subject to degradation of the magnets due to heat stress.

That's why Crank Angle Sensors are the most prone to this as they are mounted directly in the Engine Block.

[johnl]

Yes. I've had issues with the crank sensor on the Saab, i.e. failed twice.

In their wisdom Saab located the sensor on the bell housing not far from the turbocharger, where it is exposed to a great deal of radiant heat from the turbo and the downpipe (as is the bell housing exposed, so the sensor gets hot from radiant heat as well as conducted heat from the quite hot bell housing). Sure, they fitted a sheet metal heat shield, but it is a pretty minimal effort. In an attempt to head off another failure, I made a more serious shield, larger and double layered polished (reflective) aluminium with fibreglass insulating material sandwiched between the layers. Haven't had an issue since, though this may just be good luck.

Just to note that knock sensors are not magnetic, rather are in effect a very specialised kind of 'microphone' tuned to pick only frequencies within a very specific range associated with detonation. I understand that they use a crystal (quartz?) that vibrates at and around X frequency, generating a small voltage which is what is sent to the ECU.

Regards,
John.

[bazzbazz]

Knock Sensors use piezoelectric crystal, and yes, your quite correct.