Erratic idle on an RV8

Mike Simmonds (Woodcote Green 1504) from Berkshire posted a message on the bulletin board seeking help with an erratic idle on his RV8. Roger Parker (Green 4092) the well-known MGBV8 guru posted a reply.
(Jun 02)

Mike Simmonds said "I have an RV8 with an annoying problem. With the engine warm, tick over is fine at 800rpm, if a little high I feel. But when slowing down after reasonably fast driving, the tick over is very reluctant to settle down, slowly passing 1,500rpm to settle back at 800rpm after 10 seconds or so. Effectively this gives the engine no braking at all. I have checked as far as I can the speed sensor in the speedometer cable, the advance/retard vacuum and the throttle stepper motor but all appears to be fine. I intend changing the stepper motor to try and rectify the fault, but at £70 I would like to be sure before I do this. Any clues anybody? Any help would be greatly appreciated".

Roger Parker replied on the bulletin board - "the Lucas Hot Wire fuel injection system is in today's terms crude. It is only a fuel control system with engine speed input arriving through the ignition pulse which is exactly the same way as the tachometer works. The crudity does offer some advantages in terms of the scope for the able owner to rectify some faults. The Hot Wire system first appeared fitted on the 2 litre 02 series engine fitted to the MG Montego in 1984. It was later fitted to the 1.6 litre series engine of the Rover 216 EFi and the 2 litre M16 engine of the Rover 820 and the early 220 and 420 Rovers.

One issue that was a common fault was failure to achieve correct "idle speed control" through a combination of error conditions. The most common fault condition is where the initial start sequence involves turning the ignition switch straight through from off to cranking. These systems need a period of approximately 2 seconds to compile a full system status check and assimilate values for the main engine functions. This issue was addressed with the 1994 Lucas GEMS8 engine management systems where a built in delay was fitted so that even if you did switch straight through from off to cranking the system would see a 2 second delay before allowing the starter to crank the engine.

With the four and eight cylinder Hot Wire equipped cars it is quite possible to create a fault condition whereby the system is then operating with an incorrect value which then impacts directly on the "idle speed control".

The most influential value for the idle speed control is the coolant temperature and this is also a sensor which has a record of being less than 100% reliable. Remember that on the V8 there is a completely separate coolant temperature sensor for the temperature gauge so you do not know if you are seeing the correct values from the coolant sensor for the injection system. Coolant sensor resistances vary depending on the temperature applied to the sensor, but as a guide at 20oC the sensor should have a resistance of approximately 2,500 ohms, and
at 80oC approximately 350 ohms. This gives some reference to check the sensor as any higher figures than 1,000 ohms when the engine is genuinely at working temperature means a faulty reading that will imitate a partially warmed engine which sees extra fuelling and indexing of the stepper motor to increase idle speed.

The next most common problem is one of poor connection. Bear in mind that all the values that are seen by the ECU require changing resistances. By definition, bad connections create their own resistances which in individual cases are an addition to the sensor resistances and result in incorrect voltages seen at the ECU.

Slow decline from high idle speeds is also a characteristic of these systems. The original Lucas information indicated that if the system took up to between 10 and 15 seconds for the idle speed to settle after the vehicle comes to rest, then this was acceptable. In reality 5 seconds is a far more realistic figure. The road speed sensor is again a common part with some of those Rovers I mentioned. It generates four pulses per cable revolution and is actually effective from approximately 5 mph. Below this speed and the cable speed is too slow. This sensor is present only to ensure that idle speed control is disengaged when the vehicle is moving, otherwise having the engine speed controlled outside the driver input when moving at slow speed in a traffic queue would (does) often present difficulties.

Lastly I would mention the lambda sensors (oxygen sensors) as if one or both have developed faults then the resulting signal to the ECU will cause a range of problems with poor running. It is possible to tap into the signal wires from these and read voltage fluctuations which are fluctuating quickly once the sensor is hot. Slow voltage changes when cold are normal, with marked increase in the rate of fluctuation occurring after a few seconds. However if the changes are less than an average of one per second with a fully warmed up engine, then the sensor may be faulty.

There are other conditions that cause running problems, but these are often affecting maximum power or simply do not allow the engine to run. This does not mean that there are not other factors that cannot influence the running of the engine, and for this reason and MG Rover dealer or a Land Rover dealer or a specialist with Testbook or the newer T4 has to be the simplest route to see a complete diagnostic check of the Hot Wire system.

Editor: it is worth reminding readers that there are a number of RV8 Workshop Notes on the ECU and other diagnostic matters which should be read in conjunction with this very useful note. Underneath that classic car exterior, the RV8 is a very modern and sophisticated car in the engine and electronics department. See RV8NOTE130 on hunting down a misfire where Peter James found that a well equipped specialist with the diagnostic equipment was able to identify several problems and solve them rapidly.
Copyright reserved by the V8 Register