TAKE THOUGHT

I want to share my latest repair with everyone, which demonstrates the value of not leaping to conclusions and following a logical diagnostic path.

The car in question is my wife’s 2006 Vauxhall

Tigra with 74,000 miles on the clock.

A couple of weeks ago she had trouble starting it. She reported that she had to cycle the ignition key a couple of times before the starter would crank over the engine. Then she said the electrical items in the centre binnacle (ventilation fan, radio, A/C) wouldn’t work while she was driving home. She said that when she was driving home, the dashboard lights came on and the gauges made a full sweep. When she got home, she tried to start it again and no crank. At this point I was thinking a Canbus issue or wiring harness problem because of the range of symptoms.

Being self-employed and short of available time, I had to investigate this in 30 minute bursts over two weeks. I had a brief look when she first reported the problem and I tried to replicate the symptoms. Sure enough, the starter would not crank. No click from the starter and no audible click from the starter relay. The fuel pump ran, which is a good sign. The Immobiliser light was performing correctly, so no apparent issue there. The centre binnacle items were all now working, radio, fan, the A/C, HRW and recirc buttons all worked. When I connected my diagnostic kit no error codes were reported, and among the items I checked was for immobiliser key response, and the car was reporting that the key fob battery was good. However I noticed that only two or three of the vehicle system modules would communicate with my diagnostic kit at any one time, and each time I connected the kit I would get a different set of modules communicating. The vehicle battery is new and I checked to ensure it was fully charged, and so I was feeling that sense of doom that comes with a possible CAN line error or vehicle module failure.

At the next opportunity I removed and load tested the relays – all good. The relays were refitted, but moved around so that if there was an underlying relay issue, it should show up on a different circuit. I was working on my own at this point and could not perform a relay operation check by having an assistant cycling the keyswitch. All fuses were intact.

At the next opportunity, I inspected the ignition switch and found all contact positions to be correct and in good order.

Since I had replaced the radio earlier in the year, I removed the new radio to check that there was no issue with trapped or damaged wiring, and could find no issue.

The next investigation opportunity was with an assistant operating the ignition key, and I was able to determine that the starter relay was infact operating (quietly). So I removed the starter relay and set my test meter to the 10A range and used it to make the circuit in the fusebox in place of the relay contacts. I measured 8A, but there was no clack from the solenoid.

I used a set of jump leads (negative side only) to make a temporary earth lead from the battery negative terminal to the engine block, and this made no difference to the current reading.

The starter was removed and tested on the bench and would not draw more than 8A. A new starter was bought and fitted and the car now starts and runs properly.

This was a starter failure like I have never experienced before. Without the sound of the solenoid clicking and with the odd diagnostic kit responses I was nearly fooled into assuming a CAN issue and launching a full-on Canbus and wiring harness investigation. This would, of course not have moved me any closer to the correct diagnosis and would have wasted hours of my time.

I’m glad I followed my diagnostic training (I used to work for a major vehicle manufacturer and provided technical support to the UK dealer network). It’s easy to leap to a conclusion, but it can be a time-consuming and expensive folly. I recall that, over many years, I have supported many trained technicians who simply failed to follow the correct process and resorted to the lazy or desperate measure of diagnosis by substitution. The customer always has to pay for that.

Remember, don’t replace anything unless you know that it is faulty – you’ll waste money, you’ll waste your time, and you might be fitting a part which introduces a new fault into the mix.

I don’t know why the centre binnacle electrical items misbehaved, but it could have been related to a connector contact oxidation issue which was resolved by removing and reconnecting plugs, relays and fuses.

Trust this helps someone...

Tim Baker