K Series back fire

[andrew mckay]

My standard 1.6 K Series has started to back fire off the throttle. The pickup up from idle doesn't seem as good and is very noticeable when toe heeling too. Under power the car seems fine. I was goimg to start with the spark plugs. 

Are there any known issues or probable causes based on experiences out there.

Thanks.

 

[John Vine]

Some questions first....

• When you say "backfire", do you mean that there are pops and bangs in the exhaust or spitting and popping in the induction? 
• What age is the car?  Do you know whether it's EU2 or EU3?  Does it have a distributor?
• What exhaust system do you have?  Is it the standard 4-branch manifold (inside the engine compartment) and a single pipe to the silencer?
• Is the engine misfiring at all?

Pops and bangs are more usual in the exhaust, so I'd start by checking for leaks at the manifold (gasket problem, perhaps?) and at the joint(s) in the exhaust pipe. 

JV

[andrew mckay]

John,

Thanks. Not sure on ECU is an ex academy engine circa 2007 ish but the car is much newer. It doesn't have a distributor and has the standard 4 into 1 exhaust. It doesn't misfire under load but does when revved. There are pops and bangs off the throttle. It hardly used to pop and bang.

[John Vine]

It'll be EU3 (no dizzy, plus coil packs under the cam cover).

If it misfires when you rev, I'd check the plugs and leads first.

Does it tick over smoothly?

(It's a long time since I played with Ks.  Andrew Revill (revilla) is your man -- no doubt he'll be along shortly.)

JV

[CharlesElliott]

Plugs, coils and the springs that connect the coils to the plug tops would be my first port of call. 

[Paul Richards]

JV has it right. Check for leaks on exhaust manifold. 

[revilla]

I'd vote for exhaust leaks too. Manifold gasket, manifold studs all tight, lambda sensor tight, manifold to pipe joint. Look out for cracked welds too.

[revilla]

If it is drawing air in around the manifold it will confuse the lambda sensor so might explain the poor pickup too. I know you might expect holes in the exhaust to "blow" but the strange physics of flowing gases means they often "suck" instead, pulling in air.

[andrew mckay]

Thanks for all the comments I'll,have a good look tomorrow

[andrew mckay]

I've checked the plug leads, can't find any air leaks and replaced the plugs. Engine seems a little crisper, toe heel blip ok a higher revs, hesitates ever so slightly when revved quickly from idle and backfires off the throttle around 2500 revs. Other than that feels ok. Useable but not quite right. 

[revilla]

Have you got an OBDII scanner? If it is EU3 it should talk to one. They can be a bit fussy, make sure the scanner is plugged in and powered up before turning the ignition on as the ECU seems not to bother trying to talk to the OBDII if it decides there's nothing there on startup. If you haven't got one, something like a CANSCAN 900 is cheap on eBay and does the job. It will show up any obvious mixture problems or sensor problems. As a quick check of the lambda sensor you can unplug it and see how it runs open loop on the base map.

[andrew mckay]

Thanks. I think I can get hold of one and will see what comes out.  

[revilla]

If you can, have a look what fault codes there are and record them, then clear them and take it for a drive and check again - this will tell you which faults are live and which are just old ones hanging around. You will probably see several faults logged that are meaningless and can be ignored. It will in all probability be complaining about faults with the post catalyst oxygen senor (the Caterham doesn't have one but the ECU doesn't know that! - anything relating to "Bank 2" or "Bank 1 Sensor 2" is irrelevant but faults relating to "Bank 1 Sensor 1" are meaningful). It will probably complain about the evaporative emissions control system too (same reason, not fitted).

However, probably more interesting in this case is the Live Data screen. Make sure you use a scanner with Live Data capability - this will let you see the readings from a lot of the important sensors.

Have a look at the the readings from the Throttle Position Sensor. If correctly calibrated they always read the same when closed (from memory it's 3.92% but not in front of the car now so can't check). Check that the reading smoothly increases as you open the throttle and returns to the same value when closed again. No need to have the engine running for this. If the closed value looks wrong, you can get the ECU to recalibrate it by turning the ignition on, smoothly opening the throttle to full and then smoothly back to closed again five times, then ignition off (without starting the engine). If it generally not behaving as expected it could be the sensor or the wiring to it.

You can also check the coolant temperature sensor reads something sensible when the engine is cold (about outside air temp plus or minus a few degrees) and hot (probably about 80-90C, should be in rough agreement with your temperature gauge).

Intake air temperature should be reasonably too. If this is misbehaving it will mess up the ECU's calibration. With then engine cold it should be roughly outside air temperature, with then engine hot it will be a bit warmer especially with the bonnet on. I've never found this very accurate but if it is reading something stupid it will give problems. If it says -40C, there's a wiring fault or the sensor has gone open circuit internally. This apparently disables adaptive fuelling and will make it hesitant.

Manifold Absolute Pressure should be close to 100kPa with the engine stopped (i.e. about atmospheric pressure) and something a lot lower (again not in front of the car so can't check but from memory in the region of 20kPa on standard cams, higher with tuned cams).

The fuelling should show "CL" for closed loop once everything is warmed up (it will run open loop at idle until the lambda sensor is up to temperature and reading properly) and the voltage from the lambda sensor should be jumping around all over the place apparently randomly between say 0.1 and 0.9V. In fact it should be cycling up and down regularly but the sampling rates of the scanners are too low to see this so you just see snapshots of the waveform, but if it is steady or way outside of these limits you have a lambda sensor problem or a mixture problem (the narrow band lambda sensor can only really tell the ECU "too rich" or "too lean" but not by how much, so the ECU uses this to cycle slightly either side of perfect so that it can track the correct mixture). Long term and short term fuel trim should show small numbers (within +-15% but probably less). If they are up at maximum trim one way or another you have a mixture/injector problem. Don't worry about any fuelling values shown for "Bank 2", you only have one cylinder bank.

Ignition advance will probably be jumping all over the place at idle as the ECU uses it to maintain idle speed. If it isn't, it's a hint that the ECU doesn't know it's idling and the TPS might need calibrating as above.

If you want to post up any fault codes, sensor readings or findings on here, somebody will be able to comment on what conclusions you can draw from them.

[andrew mckay]

Just to say a new coli pack seemed to do the trick