We have managed a few more hours on this problem this week. Again I thought we had found it when the fault could be certainly
switched on and off by moving the loom at the connector end to the offside connector of the instrument cluster. Aha! Indicates a poor connection here or a dry joint on the connector to pcb pads, I convinced myself .
Connectors all looked good so stripped the cluster down and re-flowed the relevant joints. Then spent 1/2 hour looking for one of the instrument needles.
Fault does not come and go with the movement any more but is back to the 10-15 minute running OK syndrome. At this point the breaker's yard is looking more attractive but I don't really want to give up, I've spent too long on it and don't like to be beaten.
What I cannot get my head around is how we could provoke the fault for a short time by moving the loom at four different places so far in turn but only until we investigate that point whereupon the fault goes back into its cycle of being OK for several minutes and cannot be provoked again during that time.
I think that I now need to eliminate the cluster so as not to get bogged down by it plus I don't know what else to try. What are the implications of trying a cluster from another vehicle for the purposes of elimination?