Alfa Romeo Forum banner

1 - 20 of 26 Posts

·
Registered
Joined
·
445 Posts
Discussion Starter #1
Since I've owned my 147, every now and then there would be a clattery rattling from the exhaust. This would occur on cold start up, then usually disappear as the exhaust heated up. Uninvestigated diagnosis (speculation) was a slightly loose pre-cat (i.e. the internal core matrix). Since the rattling wasn't constant and not loud I had assumed a minor issue...

The other day I decided to actually investigate this, so pulled off the exhuast (down pipe including pre-cats). I found one of the pre-cats to look intact and healthy, but the other was in very poor condition, with a very loose internal matrix core. So, I cut the precat canisters apart to have a good look inside. The failed pre-cat matrix core was far more than just loose, it was battered, beaten and eroded away to the point that most of it's OD was only about 2/3rd that of the other undamaged precat matrix core (I estimate the core had lost more than half of its' mass). Further, the core OD was much further reduced over about 1/3rd of its' length where the loose core had settled to the bottom of the pre-cat canister, and hugely worn away until part of the core was inserting itself into the bottom of the canister. The shape of this 1/3rd of the core now more or less matched the 'conical' shape of the lower part of the canister, and the shape / ID of the exhaust tubing where it exits the cansister. All in all a complete mess.

All this would explain why the engine sometimes felt a bit fluffy and somewhat down on power, and other times not. I suspect the remnant lump of battered matrix core was intermittently partially blocking the pre-cat canister exit, and at other times not so much. I suspect the exhaust flow was being partially blocked most of the time, but not all that badly. Then at other times the blockage would be a lot worse, as the lump of matrix moved into a position where the most worn part of it would 'plug' into the exit pipe (keep in mind that the matrix is full of longitudinal holes. so would still flow some exhaust gas even it it were fully 'plugging' the exit of the pre-cat canister).

So, I binned the lump of core matrix, and gutted the matrix from the other cat. This isn't all that easy to do, the matrix is fairly robust. I used a drill, a chisel and a hammer to disintegrate the martrix (and wear a mask, the process creates dust that would pronbably be unwise to breathe in...). Welding the pre-cat canisters back together is a bit difficult as the canister material appears to be some grade of stainless steel, and doesn't weld easily or nicely (at least not with MS rods). I used an arc welder, MIG or TIG would likely be better. My resulting welds are not super pretty, but good enough with no apparent gas leaks.

Putting the exhaust system back together (with now fully gutted pre-cats), the engine feels noticably stronger, especially at higher rpm. Previously this engine had lacked power above 5000rpm, now it revs all the way to 7000rpm with significantly more enthusiasm. There is no CEL.

Later, thinking about all this, it occurred to me that the failed pre-cat core had obviously lost a substantial amount of material, and that this material must have gone somwhere. Assuming it hadn't all become powder (unlikely...), then much of it would most probably be captured and still inside the main cat, quite likely blocking that cats' matrix core. So, I pulled the main cat out to have a look. What an amazing mess it was!

There was no internal core, at all. Only some tattered remnants of what looked to have once been two mesh screens (that I think must have been located one in front of and one behind the core matrix). Despite this complete cat failure, and the complete failure of one of the pre-cats, at no time has there ever been a related CEL....

Regards,
John.
 

·
Registered
Joined
·
445 Posts
Discussion Starter #2
Forgot to mention, I've replaced the main cat with the one from my parts car, which at least looks to be OK. Of course this doesn't neccesarily mean that it actually works. The car now has two empty pre-cats and a (probably) functional main cat. There are no CEL warnings, not that the ECU would apparently know to throw a CEL since previously it utterly failed to detect the total failure of the original main cat and one of the pre-cats...

While I had the exhaust manifold / header off the car, I removed the substantial constrictive dents that the factory has taken care to place in three of the primary pipes. The worst of these dents is near the manifold flange on primary pipe number one. It is a deliberate dent obviously meant to allow the header to clear the AC mounting plate when fitting or removing the header. This dent is on the side of the pipe just where it curves toward the middle of the car (further constricting the already most constricted part of this pipe). To allow manifold fitting / removal without the dent in the pipe, I've cut back the AC plate for extra clearance (as the factory should have done in the first place...).

Primary pipes one and two both have significant dents on the top of the pipes (worse on pipe two than three). I can see no reason why these two dents have been made, they don't help with clearances (and can't help with gas flow...). All dents were removed by heating the tube wall to red heat (at least) and pushing them out from inside the tube with custom made 'pushers'.

While I was playing with the exhaust system, I experiemented with deleting the rear silencer. I did this because I do like to hear at least some exhaust note, but this cars' exhaust is so very quiet that I can't hear it at all. I replaced the rear muffler with a section of straight through pipe (some curved sections welded together), which made the exhaust louder without being excessively loud. So on paper this was a successful experiment, but I didn't like the quality of the louder exhaust note. It's a flat droney bass farty kind of noise, not in any way aurally interesting, so the rear silencer is now re-instated. It might just be in my head, but with the rear silencer deleted, the throttle response seemed worse...

Regards,
John.
 

·
Registered
Joined
·
3,390 Posts
It probably would be especially at low revs....a little back pressure is required then to help keep the fresh charge in the combustion chambers rather than all going out past the exhaust valves on valve overlap. I added silencers back on my old Audi Coupe for that very reason..real flat at low revs.
 

·
Registered
Joined
·
445 Posts
Discussion Starter #4
Perhaps. I suppose it is conceivable that at lower rpm the valve timing overlap and harmonic exhaust pressure fluctuations may possibly be counter-productive, and if so then lower rpm power may suffer, possibly affecting throttle response too (?). If so, then some loss of restriction probably created by gutting the pre-cats might have a similar affect, i.e. some increase in higher rpm power with some loss at lower rpm?

I have a spare set of downpipes with intact and seemingly good pre-cats, which I might try in place of the gutted ones presently on the car. I'll think about it after I get more of a feel for my 'new' exhaust set up as it is now (haven't yet driven much with the gutted pre-cats, and a main cat that isn't just an empty box).

There is every chance that the car has had an utterly defunct pre-cat and effectively non existant main cat since I have owned it, and that I have never driven it with a 'stock' exhaust (i.e. 'stock' as in working as it was intended). Without testing it's hard to know whether or not an intact pair of pre-cats might work better than a gutted pair. I would guess that gutted pre-cats may work better at higher rpm, but possibly not so well at lower rpm? But it's just speculating...

I am still quite amazed that there has never been an emmisions related CEL displayed, considering that this car has had only one pre-cat and in effect no main cat fitted, probably for a long time.

Regards,
John.

I wrote above:
"Primary pipes one and two both have significant dents on the top of the pipes ...". Of course I meant: 'Primary pipes two and three'. Three of the primary header pipes are afflicted with 'standard' substantial and restrictive dents, I wonder why AR didn't put a dent in the fourth pipe too?

Regards,
John.
 

·
Registered
Joined
·
445 Posts
Discussion Starter #5
So, if the ECU cannot detect a completely failed main cat (and a concurrently failed pre-cat), then what is the point of the post cat O2 sensor?

Today I experimented with disconnecting the post cat sensor, just to see if the ECU would react. Answer; no reaction, as the ECU lit no CEL. The ECU didn't even detect a 'fault' in the sensors' heater circuit, which surprised me as this is a 'go' - 'no go' failure.

I've recently had two post cat O2 sensors on this car (I changed it a few weeks ago, just on a whim and I had a spare). Both sensors would have been in place when the original failed cat was still fitted. Perhaps both sensors were / are bad, or, the ECU isn't getting a meaningful message from the sensor, or, the ECU just doesn't care...

Regards,
John.
 

·
Registered
Joined
·
427 Posts
Hi, which engine do you have? CF1/CF2/CF3 TS/V6 ? Just incase it is an engine i have expereince with, I'd like to help.
 

·
Registered
Joined
·
2,452 Posts
A 147 2.0 will be a CF3.
IT only has one post cat O2 sensor, not 2 as what John stated. Another issue is that the main cat appears to be functional enough to prevent an engine management code from being triggered.

Yes, TS and JTS engines do appear to like some sort of restriction at low rpm to keep torque from becoming non-existent. This has been noted by other members in the past who completely de-catted their cars.

Good information though, John. Not enough people know about the dent in the pipe at the AC compressor bracket and I hear there is a flood of completely useless, cheap, rubbish maniverters on the market which fit badly and completely ruin the car's performance (147, 156 & GT).

The only reason I am now looking into a Supersprint manifold is to pre-empt silly prices/supply issues of quality maniverters. It is a little more tricky with the JTS but I believe this can be overcome but it is not a cheap solution.
 

·
Registered
Joined
·
89 Posts
Another food for thought thread, as per usual john. I was looking at supersprint manifold too fruity, I am just not sure what the EM would think of it. To my knowledge, these are not made with holes for lambdas.
 

·
Registered
Joined
·
427 Posts
OK I do have one helpfull thing to say then. On our 156 , a 2001 1.8 TS CF3 we have two close-coupled cats, with two sensors by the engine, and one sensor under the car by the other cat. currently our close-coupled cat has a tiny tear in it from a speed bump. This results in a Check Engine Light after about three days of driving. So maybe you need to be more patient waiting for the CEL ?
 

·
Registered
Joined
·
461 Posts
The failed pre-cat matrix core was far more than just loose, it was battered, beaten and eroded away to the point that most of it's OD was only about 2/3rd that of the other undamaged precat matrix core (I estimate the core had lost more than half of its' mass).


Previously this engine had lacked power above 5000rpm, now it revs all the way to 7000rpm with significantly more enthusiasm. There is no CEL.


Please can you explain what 'OD' and 'CEL' mean please?
 

·
Registered
Joined
·
305 Posts
Mighty-147, I would think CEL is 'Check Engine Light'. 'OD', for outside diameter? or overall diameter but i'm not entirely sure that is the case in this specific example.
 

·
Registered
Joined
·
445 Posts
Discussion Starter #12
A 147 2.0 will be a CF3.
Yes, 2.0 CF3 TS.

IT only has one post cat O2 sensor, not 2 as what John stated.
I didn't mean that there are two post cat sensors fitted. There is only one post cat O2 sensor, but I changed it recently, i.e. an older one changed for a newer one (probably, the old one is likely to be the original unit with over 230,000km, the newer one is from my parts car which has 160,000 on the odometer). So, I meant that the car has had two different post cat sensors recently, but not at the same time.

The main cat (and the failed pre cat) must have been effectively non existant while both O2 sensors were fitted, and no CEL was generated by the ECU with either sensor. I think it is reasonably unlikely that both O2 sensors were / are bad, though not impossible (but the ECU should detect this, I would have thought...). So, either the ECU is at fault (i.e. faulty), or it isn't programmed to care about the signal from the post cat O2 sensor...??

Another issue is that the main cat appears to be functional enough to prevent an engine management code from being triggered.
The main cat was in effect an empty box, so no chance that it was functional in any way.

When in good condition, the guts of these main cats appear to consist of a square holed mesh screen directly in front of the matrix core, with another identical screen directly behind the matrix core. The screens appear as a flat mesh which visually hide the core matrix itself. Looking inside my failed cat, the screens can be seen as bent collapsed remnants, tattered with large holes in the middle. There is no sign of any of the core matrix, it has all disappeared (it's catastrophic catalytic armageddon in there...).

So the issue is that the ECU does not appear to care about the condition of the main cat, or by implcation whether one is even fitted or not. Nor does it does not appear to care about the post cat sensor, whether it is connected or not.

Yes, TS and JTS engines do appear to like some sort of restriction at low rpm to keep torque from becoming non-existent. This has been noted by other members in the past who completely de-catted their cars.
I'm thinking to experimentally install the down pipe and intact pre-cats (from my parts car). My suspicion is that my car has had a very substantially damaged pre-cat sensor since I have owned it (one of the two pre-cats), so I have no experience of how the engine feels / runs with the exhaust 'as designed'. I know the accepted wisdom is that pre-cats reduce maximum power, but for a road car I'd be happy to trade some top end power for improved bottom end and mid range. This assumes that the pre-cats do help with this, which I don't know, but might find out...

Good information though, John. Not enough people know about the dent in the pipe at the AC compressor bracket
Probably the most constricted part of the pipe is the tight bend near the cylinder head. Deliberately placing a substantial dent in the pipe right at this point just constricts the pipe further and is almost factory vandalism, yet at least there is a reason for Alfa to have done it (albeit a bad one). The dents on top of pipes two and three are just mystifying in thier pointlesness (at least they are not as bad as the pipe number one dent...).

Regards,
John.
 

·
Registered
Joined
·
445 Posts
Discussion Starter #13
OK I do have one helpfull thing to say then. On our 156 , a 2001 1.8 TS CF3 we have two close-coupled cats, with two sensors by the engine, and one sensor under the car by the other cat. currently our close-coupled cat has a tiny tear in it from a speed bump. This results in a Check Engine Light after about three days of driving. So maybe you need to be more patient waiting for the CEL ?
The main cat and one of the pre-cats must have has been complete toast for quite a long time. I don't think the ECU is listening...

Regards,
John.
 

·
Registered
Joined
·
445 Posts
Discussion Starter #14
Mighty-147 said:
"Please can you explain what 'OD' and 'CEL' mean please?"
Are commonly accepted acronyms.

CEL = Check Engine Light. On our cars it is the illuminated yellow engine shaped icon on the tachometer face.

OD = Outside Diameter. So, ID = Inside Diameter. OD can refer to the OD of any cylindrical or round object. ID refers to the ID of any round hole or tube.

Regards,
John.

PS I dislike this new forum format. Can't seem to edit a post after posting. The quote function seems very disfunctional. Is it just me?
 

·
Registered
Joined
·
2,452 Posts
Hi John, I find the quote works ok.
To edit, just look for the three dots arranged vertically which should be to the right of your ID.
 

·
Registered
Joined
·
427 Posts
Hi John, The CF3 2.0TS I have does detect the tear in our cat, but it takes some time. It may be the case it only pulls a CEL if the engine is up to temperature for a few minutes. This might mean a faulty/low reading CTS would trick the ECU into thinking the engine is still cold and in that situation the cats would not work. Or maybe you are not waiting long enough? Have you tried causing a fault, such as turning the ignition on with the fly-by-wire throttle connector disconnected? That would test a number of things, such the CEL light itself.
 

·
Registered
Joined
·
445 Posts
Discussion Starter #17
All I know is that for a substantially long time this engine was running only one pre-cat that had any chance of being functional, and a main cat that was effectively an empty canister. The ECU had a very long time to recognise this, yet apparently didn't detect a problem (even with an O2 sensor change).

Anyway, yesterday I swapped the gutted pre-cats for the intact pre-cats. Seat of the pants dynamometer (i.e. subjective impression) says that low rpm power is slightly increased, and power above about 3000rpm is significantly increased. Since changing the precats I haven't driven the car all that much and what driving I have done the road conditions were wet, so these are just first impressions. Because of the rain I haven't really played with higher rpm, yet.

Anyway, assuming first impressions hold with more driving, so far this more or less supports what Fruity has said, that decatting isn't necessarily a good thing for lower rpm power (nor seemingly for mid range power).

Regards,
John.
 

·
Registered
Joined
·
127 Posts
Since I've always respected Fruity's and Johnlear's knowledge I would like to ask something relevant since you are talking about precats. Some exaust makers and car programmers here in Greece, have advised me to remove the precats in my 147 1,6ts. They claim that the engine has a much better throttle response, and that it is not a big deal, since the precats work until the engine gets hot. I've always been reluctant about removing them, because my car has only 70.000 kms and it works perfectly (like a Singer sewing machine). I wouldn't like any problems (instability etc in any revs). Furthermore I wouldn't like any lights flashing and the rest. What's your opinion?
 

·
Registered
Joined
·
445 Posts
Discussion Starter #19
My exhausting experience with my 147 TS;

My starting point was a defective standard exhaust system, having a completely failed 'manivertor' pre-cat and a totally failed primary cat (so of the three cats it had only one intact precat). I suspect that the loose and rattling lump of core matrix moving around inside the failed precat was intermittently more and less impeding the gas flow through that single failed precat. When I gutted both precats I felt that the engine seemed to gain some power, especially at higher rpm. I think this was likely due to removing the probable restriction created by the loose core matrix in the failed precat.

I then discovered the failed primary cat (had become an effectively empty can, the matrix having long ago been burnt / eroded away). I replaced this with a good one (visually intact internally). I didn't notice much difference in power or throttle response. I then experimented with deleting the rear silencer, and felt that the engine didn't pick up any power but lost some throttle response (and sounded awful), so silencer went back on.

Next I replaced the 'manivertor' (old one with gutted precats) with another with both precats in apparently good condition. So, now the system is 'fully standard' with all three cats being physically intact. Seat of the pants dyno says that the engine now has appreciably more power than it had with empty precats and an empty primary cat, or with empty precats and an intact primary cat.

However, seat of the pants dyno also detects a somewhat lessened throttle response (with all three cats intact). This is most noticable when 'blipping' the throttle for 'heel / toe' gearshifts, the pedal needing a somewhat bigger prod than it did before, to get a similar rpm 'blip'.

Quote:
"They claim that the engine has a much better throttle response, and that it is not a big deal, since the precats work until the engine gets hot."

Replacing the precats with uniform diameter secondary exhaust tubes may give quite a different result to just leaving gutted precat canisters in place. If so then I expect this his would be due to the sudden and very substatial increase in ID and volume at the emptied precat canisters. So, my subjective experience with gutted precats (adversely affecting throttle response) may not be the same as it might be with precats completely replaced by uniform ID tubes. But given the choice of gutted precats or unmolested stock precats, then from my above experience I think stock is better.

With all three cats present and in good condition I do think that the engine feels significantly stronger, though not quite as responsive to throttle 'blips'. This is something I may get used to, but at the moment it has me fantasizing about lightweight flywheels...

Regards,
John.
 

·
Registered
Joined
·
445 Posts
Discussion Starter #20
Yesterday, with the 'new' fully catted exhaust system fitted, I thought my engine felt to have picked up a bit more power but lost a bit of throttle response. Today I think it has a bit more of a power increase than I had initially thought, and that the throttle response is as good as it was.

I suspect that for the past couple of days, for some reason I may have been driving it a bit more gently than I usually do. Today I gave it a damn good thrashing, and it was great!

Regards,
John.
 
1 - 20 of 26 Posts
Top