K-series sudden-stop (electrical)

[Myles]

Engine rebuilt two-three years ago following a cambelt failure. It then sat outside on the coast for over six months before Richard Nicolson was kind enough to drag it over to Edinburgh on his trailer . I didn't drive the car until the mot - and on the way over, it stopped dead a couple of times in traffic. Instant cut. The engine would turn over but not restart immediately. Basically parked the car on return and am just trying to get it running again for the next mot

 

Fired it up this afternoon on fresh fuel and a charged battery. Seemed ok. Ran it up and down the close a couple of times and then left it idling to check that the fan was kicking in and that the coolant bottle won't overflow (dodgy cap last year I think).

 

Just after the fan kicked in, the engine died again - instant cut again. Refused to restart (emerald telltale showed that rotation wasn't fast enough to be recognised). Juggled a few cars and hooked up the unloved Focus for a jump. Instant start - so battery will be replaced next week but is almost certainly a red-herring for the cut out - low volts would just give a gradual 'death'.

 

From memory on the trip to/from the mot last year, the car never cut out (or showed any signs) whilst on the go - just at idle.

 

So - what could be the problem? I think I might have a spare crank position sensor at my other place - that's worth trying (although I thought they were basically bullet-proof).

 

Any other ideas? the rest of the electrics appear unaffected - nothing else appears to cut out for instance.

[tbird]

Could battery be fubar-ed enough that when fan cuts in it drags voltage down so far that Emerald stops working

 

Tim

 

Edited by - tbird on 14 Apr 2012 17:52:47

[Myles]

Unlikely, I'd have thought. All I've read about, say, alternator leads coming off mid-blat, suggests that you'd get a progressive death until the ecu/injectors could no longer cope - not the switch-like kill I've got.

 

The crank sensor could be the culprit - but is it a common failure mode to go intermittent? And aren't they pretty bullet-proof? Not much to go wrong!?!

[Johnty Lyons]

I'm not sure tbird isnt right Myles the ECU will die very suddenly when the electronics stop working they are very voltage sensitive Jump start her again and attach a digital meter and watch the V reading whilst on tickover when the fan cuts in.

[mav]

Agreed. 1st thought is defo battery. Probably not enough at idle / tickover to keep things running.

 

 

 

Edited by - mav on 14 Apr 2012 22:09:40

[Mankee]

I was suffering from very similar problems, which I thought was my battery. Turned out that my alternator wasn't charging at all. The car idled OK on about 12.4 volts but as soon as you tried to drive it, it was dire. When the fan cut in, the car just cut out. Was able to jump it straight off another car though. Either hook your laptop to check the Emerald voltage readout or bung a multimeter across the terminals.

[tbird]

Myles I aggree with you about Alternator lead falling of, a good battery will politely decay to a voltage low enough to stop ECU, but if Battery already failing then the fan could well be enough to drag the Alternator output down, especially when idling

 

Tim

 

Edited by - tbird on 14 Apr 2012 21:57:06

[Myles]

Ok, thanks. I'll be ordering a new battery on Monday whatever happens. Think I've left my dvms at my flat (and the ancient laptop with the emerald software on it) so will have to check next weekend.

[charlie_pank]

sounds like battery or alternator to me. Perhaps watching the voltage across the battery terminals while it warms up to the point the fan kicks in would tell you what you need to know.

 

Does the battery have enough oomph to turn the fan after the engine has died?

 

Is the fan triggered by the emerald or by the rad thermo-switch?

[Myles]

New battery ordered but to answer....

 

 

Yep - battery runs the fan quite happily after the cut-out. It'll turn the engine over too - but sluggishly and the Emerald telltale never flickers off Red (indicating that the revs aren't high-enough to recognise rotation).

 

The fan is not controlled by the Emerald - it is controlled via either the rad-thermostat or via a dash override switch.

 

From what I recall of last-years stressful run to the MOT (conked out at least twice due to this problem in rush-hour traffic), the fan switching on is not a guaranteed initiator for this problem - it will quite happily conk out while idling fan or no fan. I do always have dipped lights on when driving, though - so it's quite possible that the car gets into a cycle of running fine (normal driving), idle for a while, battery goes critical (through either the fan cutting in or just the alternator not being able to keep up with the lights etc.), engine conks out.

 

I do have a voltage gauge in one of the SPA instruments - but it's way over on the pax side (fuel-level/volts gauge) and I know it displays something slightly-less than the across-the-battery voltage.

 

Anyway, I should get the new battery this week and will bring my dvm etc. across when I return to the gfs for the weekend - if the alternator isn't performing properly (always harboured doubts about it being borderline), I'll whip it out and get it serviced.

 

Does anyone know/recall if the Emerald telltale signal is available as an output - or does it only drive the LED on the side of the box? It might be handy to have a repeater on the dash.

[charlie_pank]

I take it the engine never cuts out at anything above an idle?

 

You can log the emerald output to a laptop if I remember correctly. Perhaps then playing back the data would tell you what events lead up to a stall?

 

Once it has stalled the emerald thinks it's not turning over fast enough - so this MUST be one of a few things:

 

1) Engine isn't turning over fast enough - dead battery, alternator (might charge ok at speed but not idle) dodgy connection(s), starter motor

2) Emerald getting wrong message from crank/cam sensor on engine speed - problem with sensor(s), problem with sensor wiring/contacts/interference from HT leads, increased resistance with temperature?

3) Emerald is misinterpreting data - send back to Emerald

 

I guess you could pull out the cam-sensor and crank sensor and test them - perhaps swap with known good ones? I'd lend you mine if I still had the K. Or could you turn the engine over by hand and check that the sensor is registering the 'pings' at the right points? Here suggests they're both Hall-effect sensors, so you should be able to test them with something ferrous. I don't know what the tolerances are on those particular sensors are, but you might find that during the rebuild one of them was mounted ever so slightly too far away from its target...

 

Edited by - charlie_pank on 16 Apr 2012 13:21:29

[Pooh_R]

You suggested the car conked out more than once in a journey. I can't see why it would be the battery if you were able to re-start the car without a jump? If the battery was bad enough to cause the engine stall , I can't see it recovering enough to then start it. This scenario would require the battery to be bad enough to pull the voltage low and overcome the output of the alternator.

 

My suggestion would be the alternator is stuggling, it's output will be least at idle, even when in good condition.

 

 

 

Edited by - Pooh_R on 16 Apr 2012 15:05:36

[charlie_pank]

When 'a friend' of mine forgot to re-attach the alternator leads and conked out, simply waiting for half an hour did allow the battery enough time to recover and then start the engine without assistance.

 

 

[Johnty Lyons]

Did offer to rewire it at a time My boy still think it need done.

[David Brown]

Could you not take the battery out the Westlife and try it? (I am sure there will be a good reason)

[john milner]

I think it's something on the alternator side. Years ago I had a motorcycle. One night riding home I pulled up at a set of lights and the drain of the brake light was enough to stall it. Leaving it a while allowed the battery to build up enough to turn the engine over. I went to a nearby garage and had the battery put on charge for half an hour and was able to get me home. The cause was a knackered alternator resulting in a flat battery.

[Myles]

Did offer to rewire it at a time My boy still think it need done

 

Did you? I thought you recoiled in horror and refused to touch anything on the drivers-side of the firewall 😬

 

Charlie - yes, from memory (didn't dare take it out on the road at the weekend for fear of being stranded - and my jump-start pack is clearly knackered too!) - the car only stalls at idle (or thereabouts). The battery is a (goosed, I think) 'Red Top' style battery - not sure how well they recover after a rest.

 

 

David - the battery in the Westfahlia is indeed identical to the one in the Caterham - but it is buried sooooo deep that you really, REALLY don't want to try and remove it unless you have to. From memory, it's buried between the front wheels roughly where the cruciform is on a S3 - it's extremely tricky to even get a trickle-charger to the terminals (previous owner had a set of fly-leads wired in - but took these off when he sold the car).

 

All ref Alternator - it's certainly been somewhat suspect over the years - a bit marginal at times. I think it got toasted when I put the first (abortive) engine upgrade in - the plastic case shows sign of slight heat-damage - but it kept on 'working'. As I have a second 7 to hand, I might whip it off and drop it in locally for a refurb - and get it set to the voltages Powervamp recommend (no more than 14.5V - but anything above 14.0 a bonus).

 

[Myles]

Latest update:

 

New battery fitted (powervamp 25)

Engine fired up first time.

Voltage at battery initially 14.5+change - dropping gradually after long idling to 14.0 +/- change.

Voltage stabilises fine when lights and/or fan are on.

Engine idled fine for probably 10+ minutes. Coolant got up to 85 degrees in bypass pipe - fan hadn't cut in on it's own, so I flicked the override. Engine continued to run fine and temp sloooowly started to come down.

Was getting a bit bored so decided to rev the engine to push more coolant around and cool it more quickly.

 

Immediately obvious that there was a problem - revs start to die as soon as I tickle the throttle.

 

Tried to coax the revs up, but the engine just petered out.

 

Pulled plugs - wet, mucky and I could see that the tops of the pistons were damp (sounded like the injectors continued to fire after the engine died until I cut the power).

 

I cleaned the plugs, reinserted them and tried to restart. Bit sluggish turning over and it gave a quick pop through the exhaust, but caught and idled and rev'd fine.

 

Don't have my old laptop with the serial port and emerald software here so can't tweak idle map. I'm guessing that it's overly rich - and just sooting-up. Dunno quite why it would do this now as the map hasn't

T changed.

 

I do have a wideband lambda on the car - haven't calibrated it within memory, but it was showing high 15s at idle (lean???) so don't quite know if this adds up.

 

 

 

Soon, I think I've had multiple issues contributing to the cutting out. The sudden-cuts really do smack of being electrical to me - but I can see that if the idling is badly set up, and I'm sitting in traffic - that trying to rev and move off might result in a stall and refusal to restart due to crudded plugs etc.

 

Any thoughts?

[CageyH]

Re-calibrate the wide band in free air.

It sounds like it thinks it is running lean, so adding more fuel.

This can't help.

[GPBox]

Get a USB to Serial converter from Maplin etc...to be able to plug into the Emerald...cheap and easy....

[Myles]

This is an older version of the emerald and isn't running in closed-loop mode, so the afr shouldn't be an issue as far as I'm aware.

 

I have recalibrated it though - and I did pop in to my flat and pick up the old laptop. And it still works! And I can still remember (or guess) the password! And the cable (which is effectively hard-wired into the car) still works!!!

 

I've just fired it up and it was running a little rich to start with. Not sure what the ideal afr at idle is though - so beyond randomly pressing buttons until I found the live adjustments, I've left it as it was.

 

I only idled it for a few minutes but checked that the engine responded to the throttle every so often. I then shut it down, realised that I would have to push it up the slope into the garage and so tried to restart to drive it in. It took a goodly amount of cranking to fire - the emerald was recognising rotation (telltale) - but it wouldn't catch for many, many seconds.

 

Haven't pulled the plugs to check for fouling as gin was calling!

 

I recall that we never really put much effort into tweaking the startup. The idle advance feature is enabled, but not sure if the low speed ignition map is ideal. Can anyone (Oily for preference!) give me some pointers? Ta.

[charlie_pank]

Hi Myles, AFR should be 14.7:1 at idle.

 

I wonder if you've got that issue where the HT leads interfere with one of the sensor signals when you start to rev it up. Are you EU2 or EU3?

 

If you're fouling the plugs then you must be running too rich - perhaps the coolant sensor thinks it's colder than it really is? Or perhaps the TPS has been nudged so it thinks the throttle is open wider than it really is? Didn't you make some MAP mod for touring in the alps?

[Myles]

I don't think I made any adjustments for the alps on the basis that we would be yo-yoing the altitudes and it wouldn't be worth optimising for one - certainly didn't take the laptop.

 

I'll see what adjustments are made by the emerald next time i fire it up, but once warm, Im pretty sure that it had settled down and wasn't making any fuelling changes.

[Jason Plato]

knackered petrol ?

[Myles]

I drained and replaced with a gallon of fresh last weekend.