Firstly honourable mentions to rhinoman and pitmole, both of whom have been a direct help in this problem. rhinoman even offered to fix my ECU for me but at that point I wasn't absolutely sure it was the ECU that was dead.
The Symptom:The Vitara had been standing for at least two years, during which it had never started properly. The owner at the time (pitmole) had tried manually feed fuel into the throttle body and the engine had attempted to start. Therefore the engine, spark, fuel pump drive etc were all good. On these Vitaras the fuel pump and starter are triggered by the ECU only after the immobiliser has been disarmed so the immobiliser was also working as intended.
The Diagnosys:My personal investigations after buying the Vitara off pitmole showed that the fuel pump was priming correctly, that the spark was solid although the spark plugs were caked. I too verified his experience by manually feeding fuel in and verifying it started.
Upon disassembly of the injector I discovered that the reason for non-starting was that the injector wasn't releasing fuel into the throttle body. The injector itself seemed clean and crud free but limitations of my electronics equipment at the time meant that I couldn't verify that the ECU was correctly instructing the injector to open.
It was at this point that rhinoman responded to one of my posts on the subject and then responded to a couple of PMs. He said that the injector drive on the ECU was a weak point and offered to help identify/fix it. I wasn't convinced it was an ECU problem at this point and shelved the investigation until I had the right tools.
Break to three weeks later when a conversation with pitmole about putting a carb on it to get it running brought out a link to running Daihatsu carbs on Suzuki manifolds. pitmole noticed
this link in the menu bar showing that someone's injection Samurai had stopped working. His investigation revealed that the electrolytic capacitors on the ECU PCB had reached the end of their useful life, leaked electrolyte and rendered the ECU inoperable.
The Solution:It took me half an hour to figure out how to remove the ECU without taking the entire dash off. For reference the easiest way I found was to remove the right hand speaker cover, then the speaker itself. Behind it you can see the steel box housing the ECU and can guide a Philips screwdriver up into the front retaining screw. The rear retaining screw is an inch or so further back and you need to hang upside down in the foot well to see it clearly. Once the bolts are out the ECU can be manhandled around the back of the pedal box and disconnected from the two looms.
Removing the four ECU housing screws revealed a two layer PCB. It wasn't that complicated but surprisingly was dotted with Mitsubishi logos. A project for further investigation is to see whether this ECU is used on any Mitsubishi vehicles.
Even a cursory visual inspection revealed two critical things:
1) Someone had attempted a repair before. The warranty seals were intact before I got my hands on it so by implication this was a Suzuki technician. Or maybe they were more careful with the warranty seals than I was.
2) The electrolyte had clearly leaked out of two capacitors, corroding the board. One of these leaks was around the above repair so I assume that was what triggered the above repair job. I noticed that the larger capacitor which had leaked also showed a large bulge, implying it had completely failed. There were also scorch marks on the inside of the ECU case.
The repair job wasn't an amazing piece of work and missed a track which had been corroded by the electrolyte. I went on a trip to Maplin and purchased two capacitors for each capacitor on the board plus a few things to supplement my toolset. In particular I purchased some solder braid as I only had a solder sucker.
The remainder of the exercise was quite straight forward. I removed each electrolytic capacitor in turn and inspected the tracks to identify the intended track route. I used a scalpel to scrape the solder-resist off the relevant tracks and used 7/0.2 electrical wire to patch in the track sections corroded by the electrolyte. I repaired the previous repair and reinstalled entirely new capacitors for all electrolytics.
Care was required removing the old capacitors and it was here that the solder braid came into its own. I found that the most reliable way of removing the capacitors was to heat each leg in turn and gently bend the capacitor to release that leg. Care should be taken to ensure that the solder is melted on both sides of the board but not so much heat that the capacitor and PCB were over heated. A temperature controlled soldering iron is a huge benefit here.
After removal of the capacitor the solder braid can be used to clean out the holes making it easy to install the new capacitors with fresh solder.
The Results:Following a jump start the Vitara now runs! It's lumpy as hell, sounds like at least one cylinder (possibly three) aren't firing but it's running! The fuel is at least two years old, there's probably a ton of condensation in the tank and the spark plugs and leads will need replacement. I'll replace them and clean out the distributor and try it again.
It goes to show that if you're reasonably handy with a soldering iron and DVM and take the time even the electronic systems of reasonably modern vehicles aren't that scary.
Learn from my experiences and fix your ECUs!
Edited for grammatical correctnessification
