If it's indeed the switch in the dash i found one here in Bulgaria selling it for 5 bgn

Brilliant, thanks Todor / Hughie, I'll try that if I have to

Thanks Toe, also. All noted

I have a plan: I have to go to Birmingham this weekend anyway to pick up the keys to my new place, SO I'll hire a car, go to Brum and do the handover, then go to the unit and retrieve the spare Koitos and also a spare switch if I have one - I have a couple of spare dashboards, which if I recall correctly still have all the switches in, so fingers crossed.

BUT if I don't have a spare switch - Todor, I might be asking you to get me that one! A pity you've already sent the rad!

Then back to the south, replace the switch / lights / adjuster.

Then I just have to figure out why the car failed on emissions yet again, when yesterday it had new oil, plugs, and clean filter, a tank full of good quality petrol (i.e. not the cheap supermarket stuff) and also a bottle of Redex emptied in there, and today I drove it for an hour just before the test, so the engine was hot.

If the cat needs replacing, this will be the 6th cat in 7 years. The problem MUST be the car. I need a masterclass in tuning up the engine / troubleshooting the idle/mixture, if that is indeed the problem. Any offers?

S

Sounds like a plan batman

I'd definitely say the engine is at fault if you're burning through that many cats which part of the emissions test did it fail on? CO? Lambda?

It failed on:

Fast Idle test:
CO (0.66, max is 0.30)
Lambda (1.064, max is 1.030)

Second Fast Idle test:
CO (0.58)
Lambda (1.091)

Passed the natural idle test - just, was 0.49, max is 0.50


S

Hi Serena,
Oh dear, sorry to hear you are having cat problems again!

Emissions problems can be hard to track down because it could be any one of a number of sensors playing up. I wouldn't condemn the cat or the lambda probe just yet, (although it could just be either of those).

I'd say the lambda reading is a good clue, it's far too high. Is there a leak in the exhaust - because that can upset the lambda and put everything else out, as the lambda then tells the computer to adjust the fuelling to try to compensate, and that makes it worse.

Or - I may also go back to what I said last year - is there oil getting through to the cat and lambda and contaminating them? It wouldn't take much oil, because over a year it amounts to a lot!

As for the headlights, yes the height adjuster is now on the MoT. If it's the headlight motor then usually they go one at a time, so look at the wall in front of you as you adjust the switch and see if only one of them moves up and down. If neither move then it's more likely the power to the motors, so yes it could be the switch, (or are they on their own fuse? - must check the Haynes manual!)

Anyway, good luck and keep smiling. Let's hope you don't like the hire car

Cheers John

I thought it was odd for the leveller to "suddenly" go - probably it's been gone for years, as I never touch the dash switch.

I tried the light against the wall test tonight, and neither beams are moving so I suspect the switch is gone. The switch isn't lit up, either, which adds to my suspicions that that's where the problem is. I'll replace it this weekend, anyway.

The O2 sensor was replaced about 5 years ago, and the cat was new last year. In fact, about 6 months ago, as the one fitted at last year's MOT fell to pieces under me during the summer, and had to be replaced.

If this cat is now dead, Kwik Fit will replace it as the work is under 3 yr warranty (the warranty on the cat being the only reason I've gone back to Kwik Fit, though they're pretty good at this branch)

I did laugh though: Butch was in there briefly last year, and today someone I'd never seen before at the branch said "Are you the person who's got two of those?" - so apparently I'm famous there and so are the cars

S

Be careful when changing parts on the headlamp leveling system. There are 3 different possibilities of which version Mazda has chosen for your car. So if you use Koito headlights, you have to use the switch with the blue plastic part.
If you mix parts nothing will happen, but it also won't work. There is a little workaround, I'm currently compiling an instruction how to use different HL parts in one system.

Good info there from Rumpel, I wasn't aware the switch was changed between years, so good shout!

I definitely think something else is causing your emissions to fail every year, as jinglebells says

Trying to think back, I know you've replaced the distributor O ring (or whole distributor) - could your ignition timing be out far enough that you're not getting a complete fuel burn?

Are you burning through oil at a noticable rate? If you're having to top up the oil regularly, and you can't see any leaks, it's burning away!

Same for coolant - if it's not leaking, but it's disappearing, it's likely burning!

If you've a blowtorch, you could remove the O2 sensor, gently clean it with a fine wire brush, then blow torch the tip of it until it's glowing red/orange/yellow hot, then sharply blow any crap out of it, let it cool and refit it. That alone got me through an emissions test once

Vacuum leaks anywhere?

No, oil and coolant are fine, I check them pretty regularly. The only ongoing issue with oil has been the leaky rocker cover/gasket but that only leaks outwards, surely?

I did have idle issues, which I solved by adjusting the idle valve on the intake, but that hasn't been going on for a whole 6 years.

The fuel filter probably needs replacing - at least it's been a looooong time since it was looked at - would that affect it? I was trying to do that a few days ago, as you know. No chance to take off the battery etc. yet.

S

Sorry to hear you're still have emissions control issues with Sundance, Serena.

Just a suggestion: if you're still getting a leaky rocker cover gasket, check the PCV valve and breather hose from the rocker cover to the intake for blockage.

Excess blow-by could also be an issue, this is recycled back through the oil breather/rocker cover into the intake. This can cause emissions issues and unstable idle.

To that end, at idle take the oil cap off (use a rag if there's no shield and oil can splatter and see how much "puffing" you get out of the cap. It should more or less puff in and out wilth little smoke. If it puffs out smoke a bit like a chimney or steam train (on a smaller scale) you may wish to have the compression tested and/or otherwise check for blow-by or maybe valve stem leakage.

If the cat is being contaminated by oil it should be able to be seen on the honeycomb structure- it'll be coated/clogged with black, carbony type crud.

If there's too much fuel going in, it can overheat the cat as the cat tries to burn it and disintegrate or melt the honeycomb.

Conversely if the mixture is lean the exhaust gas itself can be too hot an possibly damage the cat (along with other parts, like your pistons and valve/valve seats).

If there's coolant reaching it, again the honeycomb becomes contaminated.

The other this is perhaps the cats they are fitting are not good enough, i.e. not up to the job anyway.

Good luck with the diagnosis and fix.

Patrick

Cheers Patrick - I've just had a quick look at the car on this basis.

I opened the oil cap during idle, absolutely nothing came out, all that happened was a change in the engine noise (it started growling, basically).

However the PCV valve has oil around the rim, and it feels rather too loose for my liking - so I'll replace that anyway. I imagine, if that's not sealed properly that will cause mixture problems?

S

Hi Serena,

yes a loose/leaking/blocked PCV valve can cause idle problems. At idle there's high vacuum in the intake area due to the closed throttle. So blow-by through the PCV can be ingested, the engine idle is designed to cope with and expect this.

If the PCV valve sticks open or leaks, too much gas gets in and interferes with the idle (lean)- it can be rough and/or stall.

If the PCV valve sticks closed:
- pressure builds inside the engine and causes oil leaks through the gaskets and seals.
- it can also mean the blowby gases mix with the oil more, causing sludging- and modern Japanese engines (yes this is still one- basically anything past the 2vlaves/cylinder period) really don't like that. - because the ECU is designed to cope with the PCV breathing during idle, a blocked PCV valve can mean the engine runs too rich- perhaps failing emissions.

Patrick

The valve still rattles - which I gather is a good sign? I wonder if maybe the rubber seal on the rocker cover PCV valve inlet is worn and not sealing properly. I have a spare rocker cover in the unit too, so I'll fetch that at the weekend & see if swapping it makes a difference.

S

Many thanks! I'll bear that in mind when I do the swap next week

S

The rubber ring on the rocker cover is hardened itself from years of exploitation.Judging by mine, you have to be carefull no to break it when removing installing it from the rocker cover when cold.
I tried it cold once,twice, then warmed up the car and it pops right out.I changed mine with another used one,same thing when hot - lose.I wrapped the PCV valve in black isolating tape to make it thicker - WIN!

Great minds think alike ... I was thinking about that just now (I have the valve on my desk to clean it).

I'll do that

After 6 cats (probably 7 next week ) I'll try anything !!!

S

Yes if the PCV rattles it's a good sign generally because it should mean the valve is opening/closing. But some of them are spring loaded so rattling may not be a good sign! Usually a replacement is fairly cheap so it may not be a bad idea to replace it anyway.

As for the cat, get it and cut it open or at least have a look down the inlet and see if there's actually anything wrong with it. If there's nothing wrong with it, no point replacing it and costing yourself $$$.

Note whilst your CO is high indicating incomplete combustion, your Lambda is also relatively high indicating a lean mixture.

So how can the CO (indicating incomplete combustion) be high if the mixture is lean? Incomplete combustion- a misfire, in particular, a lean misfire. So check your plugs, leads and ignition system. A misfire on (say) one cylinder will cause hydrocarbons to get to the cat, plus an excess O2 in the exhaust for the O2 sensor (unaffected by HC). The ECU may then try to enrich the mixture to fix the high O2 exacerbating the HC in the exhaust and the CO production and burning out the cat.

If this is the case then yes the cat will need replacement as a symptom BUT you need to fix the underlying issue first. A new cat will simply disguise the issue...until you need your 8th.

Other common causes of high CO are:
- Oxygen sensor failing
- dirty/clogged air filter
- MAP sensor failing (do you have one)
- Faulty Throttle Position Sensor (TPS)

Common causes of high Lambda are:
- leaks inc. in the exhaust system
- possible faulty oxygen sensor

From the following page:


It may be worth getting a proper 5-gas analysis and before the cat if possible. If your local mechanic isn't investigating or suggesting this it sounds like he probably should...a replacement cat is a good way of boosting his profit margin each year.

Unfortunately it's not OBD2 otherwise you'd be able to get the ECT, lambda, O2 sensor and injector trim readouts to make a better diagnosis.

Patrick

Hi Patrick

Yes, I agree, all good stuff; but are you familiar with "Kwik Fit"

You are quite right that the "mechanic" should have made a few checks before sticking on a new cat just to get it through the MoT, but ... Kwik Fit! As Serena says, the warranty looks good.

If I recall from a few posts back, Serena said the current cat was fitted by Kwik Fit last year, and yes they are profit-motivated but not to the point they will risk having to honour the warranty.
It will be interesting to see what they say, I hope they now don't try to wriggle out of the warranty by saying the failure was caused by an external factor which they forgot to check last year!

Patrick: thanks, I'll be at my parts unit tomorrow so I'll see if I have a spare tps sensor. The O2 sensor was replaced at the start of this saga a few years ago, so I'll put that to the bottom of the list to replace.

Any idea if V6 and 1.5 TPS sensors are interchangeable? As I know I have 2 V6s sensors in the unit.

I don't have a spare map sensor - anyone got one? Again, are they interchangeable with V6? If so I'll borrow Butch's.

John: Yes, I don't generally trust Kwik Fit but until the emissions issues are resolved I'll stick with it: especially as under the warranty they have to give me a new cat.

S

Jinglebells: I am not familiar with 'kwikfit' but I can guess from their name exactly what they do and how quickly (and perhaps for how much). I guess it may be akin to the USA 'jiffylube' which I hear about on various other chat groups.

It comes back to the old "quality, cost, time: pick any two" adage. Or from Lord of the Rings: "short cuts make long delays." Or Ian Fleming's "Once is happenstance. Twice is coincidence. The third time it's enemy action." (Serena's at 6 or 7 now)

Serena: yes stick with the warranty, if they complain there's some other issue causing the cat fail remind them that's what you brought it in for in the first place...so their cat replacement didn't actually fix the original issue and they can honour their warranty by fixing it properly now and not just replacing the cat (assuming you didn't just sign a contract to replace the cat). Point out if they know it's not the cat and that the cat is being damaged by something else that's an effective a posteriori admission of guilt that they didn't diagnose and fix the issue properly in the first place!

Patrick

Hi Serena,
Just summarising where I think we are with this:-
- The CO reading suggests the cat is dead, but it could be a fuelling/ignition issue;
- A new cat might get it through the MoT this year but will fail again next year;
- The lambda reading suggests something else is wrong (as Patrick says, could be the MAF,
the TPS or the PCV, or a leak in the exhaust) in which case a new cat might not be enough.

I wonder if your garage is as helpful as mine was, they did an intermediate emissions test for me, after some repairs, before doing the official MoT retest, because you only get one re-test before paying the full fee again.

I'd be pretty confident it hasn't been misfiring, or you would have noticed it in your pre-test driving.
But a small amount of excess fuel (or oil) isn't immediately apparent when driving.

I don't know whether the V6 parts (sensors) will fit a 1.5 but my guess is that the MAF will, but the TPS won't (because the inlet manifold must be different) but like I say that's a guess. Worth a try because replacing all the sensors with new ones is an expensive business.
Good luck this week.
Hope the hire car wasn't a Mazda 3 because you'd like it too much and want to keep it!

Cheers guys, yes I think that's about where we are

Just checked the V6 MAF against 1.5, they are different - looking on eBay for a spare 1.5 MAF, as I don't seem to have one.

I have 2(!) spare V6 TPS sensors, but comparing photos online, I think they are different to the one for the 1.5. However I will check tomorrow. The solution, it seems, is to buy a cheap 1.5 throttle body from eBay as mostly the TPS sensors are still attached.

Anyone any idea if MAP sensor is same on both cars?

S

MAP sensor does not exist on our F's, they got MAFs.
There is huge difference betwenn the two sensors
MAP - Manifold Absolute Pressure meter
MAF - Mass Air Flow meter

MAP measures the negative/positive pressure in the intake plenum, MAF measures the mass of the air that goes into the intake manifold.

I belive i got one spare tps for 1.5(and for a V6)
By looking at them, i'm not sure they are compatible.For sure they will have different mounting points and thus making the reading incorrect.The resistances could be different too.
Also the MAFs on both cars are different(for me), the V6 one is with larger diameter and calculates the air going through it for the bigger motror, the 1.5 is smaller in internal diameter so it calculates the air for the smaller engine.. the calculating actually is done by the ECUs not the sensors sensors just gives different resistance when the air is with different quantity.

If you need a 1.5 throttle body give a shout Serena, I have the one here from womble.

Ah, cheers Paul, it has the TPS?

That's actually what I'm looking for, but on eBay most are still attached to a throttle body

S

Serena, may i ask, what is the problem with your (Sundances) TPS that you want obviously to change?

Hi mazza, my understanding is that one of the many diagnosis posts above says a faulty TPS could be the cause of high CO.
Whilst I agree that may be true, there is no guarantee that this is the problem with Sundance. I think Serena is willing to try anything as being under pressure to get Sundance through the MoT.

Bear in mind we have done these diagnoses from a distance! Not having seen or heard the car, it's impossible to be certain. But my feeling is that a faulty TPS would be faulty enough to cause a bad "feeling" when driving it, or to make it stop altogether, and Serena would have noticed something amiss.

But having said that, changing the TPS for a known good one won't do any harm.

Hi Serena,
May I ask, when you changed the plugs, what did the old ones look like, i.e. light brown/grey tips (which is fine) or dirty black (too rich) or electric blue (too lean)?

If the plugs looked good then I think the fuelling/air and ignition can't be too far out.

Sundance is 22 years old on Wednesday, I've had him for 20 of those and I know for a fact that none of the relevant sensors have ever been changed. So yes, I'm changing anything that might have quietly died without me realising, and is now causing the problem - or contributing to it.

The plugs looked very normal, i.e. a bit furred up, but not terribly. Mostly just light brown/grey deposit. but they'd only been in there a year or so, I changed them as a precaution. The oil DID need changing, it was long past it.

The engine sounds fine, in particular when it's warm you can hardly hear it, it ticks over without any concerns.

S

Have you tried to see if there is some error codes stored in the ECU?
If there is, they might point you in the right direction.
I know from experience that faulty TPS gives a blink code as Blue's TPS was with bad connection..