Possible fuel or sensor issue?

[David00]

Need some advice, please:

My NA SW20 with a 3SGE engine was SORN for over 2 years. In hindsight should have drained the fuel down... Turned it over every three months or so and ran it hot for 10 mins, just to check it hadn't died.

Got it back on the road last week, passed MOT etc. Drove around 20 miles to and from the test centre, briskly on the way there to get the Cat working again. All seemed OK, a little sluggish perhaps.

The next time I drove it, it was fine for the 30 minute drive. But on the way back there was a noticeable difference. Engine revving up and down of its own accord. Sounded unwell. Poor acceleration. Didn’t stall, but kangaroo’d. Smelt like it was running rich. Seemed like a fuelling issue - and after 2 years I’m sure the fuel was degraded - but I didn’t experience anything like this for those first few drives. Also it was dry that day.

With the tank almost empty I refilled it with premium unleaded. Such a different sound, acceleration back. Fine at 4,000 RPM, but then I wouldn’t notice a small hesitation or surge as there’s so much angular momentum. But still some kangarooing - nowhere near as much as before, and only really noticeable at lower speeds. Also rough idling, surges for 2-3 secs then drops down.

I’m wondering if the first few drives I did with stale fuel has damaged a sensor, or something else is amiss. There are 18 possible fault reasons for erratic idling in the 3S-GE engine manual, the first one is fuel quality that I've fixed. Before I work through the remaining 17, can anyone suggest the cause?

Thanks,
David.

[Zircon]

I would advise that you drain the fuel tank completely.
If the really old fuel you had in there contained ethanol, then it is almost certain that there is a fair quantity of water in the bottom of the tank that is being sucked up and sent to the engine, causing the kangaroo effect. Even non ethanol fuel deteriorated over time and could cause a similar kangaroo problem.
If you have a drain plug on the side of the tank the job can be completed without too much trouble. If not, you need to bridge the fuel pump relay contacts, then remove the outlet pipe from the fuel filter and temporarily replace it with another pipe to drain it into a suitable container. With the relay bridged, just turn the ignition key to the position before the starting mode - the fuel will be pumped out in a continual stream until the tank is empty or you switch the key back to the removal position. Do not let the fuel pump operate for more than a minute or two as it could become hot. Then let it cool down, say 5 minutes, before repeating the procedure.
Hope that helps and maybe cure the problem.

[David00]

@Zircon - I'm familiar with the concern around water being absorbed hygroscopically by ethanol. But I don't think it's a water issue, for the reason further below.

Even if the fuel had degraded, adding nine parts of fresh fuel to one part old should have significantly reduced any stale fuel issues. Plus my SW20 is normally filled with Esso Synergy Supreme + which contains no ethanol.

Since coming off SORN I tend to start the car and let the engine run for 5 mins or so before getting in - mainly so I get some heat in the cabin. But I've realised that if I drive it from cold, there is no problem initially. Whether idling or driving, it runs smoothly when first started.

However as it starts to heat up (about 1/4 on the temp gauge, where 1/2 is hot) the car will intermittently surge slightly - perhaps another 150 RPM for around 2-3 secs - then drop back for another 10 to 20 secs

So I am wondering if this is the known temp sensor issue. I have run the car up to heat with the coolant cap on first turn - so water can expand out - but haven't seen any bubbles. What else can I do to ensure the sensor is wetted?

Has anyone had first hand experience of the temp sensor issue, and does this sound like it?

Thanks,
David.

[Juliuswws]

Hi David,

I would really change my fuel filter after all that 3 years old fuel...
As for the coolant temp sensor, it should be located near the oil pressure SW switch. There is a junction there for the coolant passages. The sensor for the gauge there as well but that's not important.
ECU uses a different sensor for the coolant.
It should measure around 4.3k Ohms when cold (it will vary with ambient temp too but should be around that value)
You could also measure it at the ECU, just google the pinout for yours. That pin called "THW"
It will be the middle connector.

Let us know,
Cheers,
Julius

[David00]

I appreciate the suggestions.

I’d really like to hear from someone who’s had the coolant issue. As I’m getting the same engine behaviour when at idle or 3,000 RPM - and from what I’ve read since posting previously, that’s not likely to be ECU coolant temp related. I’ve tried to pull the temp sensor header off - but as is often the case it’s stuck and I’m reluctant to force it as usually that leads to other problems…

I’m not convinced it’s the fuel filter. If there was a blockage, that would tend to affect high demand first. But I’m seeing the issue whether low or high fuel flow. So replacing the filter - more cost and time - isn’t a preferred option at this point. Plus I changed it in 2019 as preventative maintenance.

I could work my way through all the diagnostics steps in the 3SGE manual to find and fix this. But that will take some time, and will be expensive as I throw parts at it.

Does anyone recognise the symptoms please? As it starts to heat up (about 1/4 on the temp gauge) and when hot, the car will intermittently surge slightly - perhaps another 150 RPM for around 2-3 secs - then drop back for another 10 to 20 secs. That’s at 1,000 RPM or 3,000 RPM, and whether steady speed or accelerating.

Plugs seem OK. Distributor cap fine, and dry. Understand the HT leads can cause problems, but this feels more like it's ECU & sensor related. Or possibly an issue with the fuel pressure regulator - but I don’t know if that would manifest itself with the symptoms outlined.

[David00]

OK, here’s what I discovered:

The problem turned out to be no spark across a plug gap on one cylinder for a few seconds at a time once the engine was warm. I assume some unburnt vaporised fuel was passing across the O2 sensor as a result of this, and the ECU was getting confused by the exotic exhaust gas blend, helping to mask the root cause. That also explains why it smelt as if it was running rich!

The HT lead connection from that plug to the distributor cap wasn’t great. There was some black oxidation on the pin that clipped into the distributor cap. But enough for a spark to easily jump across.

Cleaning this pin with Jenolite has made a major difference. I can now hear a single misfire occasionally when idling, but no issues when driving.

As the engine gets up to temp, something somewhere is either making it easier for for a spark to jump across some part of the HT circuit from the coil onwards to ground - possibly perished insulation, or the plug temp is making it slightly harder for a spark to jump across the plug gap.

The spark always takes the path of least resistance. I don’t know where this point of low resistance is, but it’s at a threshold level. The small amount of oxidation shouldn’t normally cause the spark to bridge across elsewhere. I'll probably order a set of new HT leads.

Hope this may help others at some point.