Testing
“The Overcooling Syndrome Overcome”

Following the modification of the cooling system, extensive tests were carried out using our Zetec 2.0 Litre. We also had the benefit of access to data from testing carried out on an 1800 Sport Carbon and an 1800 Speed Sport, with unmodified and modified cooling systems.

Road Tests

Our Westfield cooling system was modified during the build, so no data was available for the unmodified system. During 2000 miles of motoring over the summer and autumn of 2000, no problems with overheating or overcooling were observed. In open road conditions the indicated temperature was a constant 90ºC.. In low speed, low airflow conditions the indicated temperature reached 100ºC. at which point the fan operated and cooled the temperature down to 95ºC. at fan switch off. Owners driving the 1800 Speed Sport and the 1800 Sport Carbon report similar temperatures and operation to our car.

Static Tests

Chart 5 Westfield Speed Sport 1800 Zetec Unmodified Average Difference in Temperature 15.8 ºC, Cooling Fan came on after 24 minutes, Afterthe test a 5 mile drive in open road conditions the VDO Gauge stabilised at 70ºC Click on the diagram above to see the full picture

Having observed the performance on the road we needed to verify the accuracy of our observations. Tests on an unmodified Speed Sport showed differences in the VDO gauge readings and the temperatures reported by the Laptop PC, connected to the ECU, averaging 16ºC. (See Chart 5). These differences needed to be understood and the true result determined. Was the VDO gauge or Laptop accurate? VDO stated that the accuracy of the gauge was 2.5% FSD (full scale deflection) or +/- 3ºC. The average discrepancy reported by the Lap Top software gave cause for concern. The laptop reading originated from the Ford Sensor mounted in the thermostat housing via the ECU. Our car had a Ford Mondeo Sensor mounted in the identical place but unused, because our car has carburettors fitted. We decided to use the Ford sensor to check the accuracy of the gauge and laptop display

Testing our Zetec 2.0 Litre

The first task was to determine the temperature / resistance characteristics of the VDO and Ford Temperature Sensors. Using data we obtained from VDO, Ford, Haynes Publications and some mathematical analysis, we were able to deduce the thermal coefficients and plot the Resistance / Temperature characteristics of the sensors fitted to our car. (Refer to Chart 1 & 2, by clicking on the diagrams below).

Test 1 

We emulated the test on the 1800 Speed Sport. The vehicle was set up on axle stands in the garage and in still air. Ambient temperature was 2ºC. The engine was run at idle speed observing Ford Sensor resistance and VDO Temperature Gauge Readings at 1-minute intervals. The object of the test was to compare the results with those from the Speed Sport and to verify the data we had calculated for sensors. (Refer to chart 6).

Chart 6 Westfield 2.0 Zetec Modified cooling System Test 1

The results were as predicted by the resistance / temperature data for the Ford Sensor and the VDO Gauge readings experienced during road testing. There was a close correlation between temperatures reported by the Ford Sensor and the VDO Gauge readings.

The thermostat can be seen to open, after 14 minutes, with a corresponding reduction in temperature as coolant flowed through the radiator. The engine continued to warm up to 100ºC where the Fan was observed to come on.

Chart 7 2.0 Zetec – Speed Sport 1800 Comparison

Comparison of our test, with the test on the Speed Sport 1800 (Chart 7) shows similar readings for both VDO Gauges. The Ford sensor broadly agrees with the VDO Gauges, The laptop appears to be over reading by a significant amount and is therefore suspect. The modified cooling system on our 2.0 Zetec lags behind the Speed Sport, at first, but this is explained by the fact that the test were done at ambient temperatures of 2ºC and 16ºC respectively. After 10 minutes both vehicles have reached the same temperature. Subsequently the modified cooling system of 2.0 litre Zetec is always ahead and hotter than the Speed Sport. The 2.0 litre thermostat opens at 14 minutes and 3 minutes later on the Speed Sport. The cooling effect of the thermostat opening is much more pronounced on the 2.0l Zetec engine than the unmodified Speed Sport. By the end of the test, the 2.0 Litre Zetec had gone through 2 cycles of the cooling fan, to one of the Speed Sport.

Test 2

Following on from Test 1 it looked more likely that the VDO gauge was accurate. To prove VDO gauge accuracy we attempted to measure VDO sensor resistance / time as the engine warmed up as we had done for the Ford Sensor. However the results were inconclusive for two reasons. The resistance of the VDO sensor is much less than the Ford Sensor.

The resistance readings were measured using a digital multimeter. The low value of the sensor resistance (287 ohms at 40ºC) combined with the multimeter voltage caused some internal heating of the sensor and a corresponding error.(There is dissipation power, D mW/ºC, quoted for all thermistors. This is the power that must not be exceeded if the self-heating error is to be prevented) 

The low value of resistance meant that over the temperature range 80ºC to 100ºC the readings were very volatile and difficult to determine accurately

We designed a test circuit to overcome these problems. The circuit used very low voltages across the sensors, to keep the thermistor power below the critical value, and then scaled these voltages to a value between approximately 5 volts and zero volts over the range 40ºC to 120ºC for both VDO and Ford Sensors. The accuracy of this circuit was to within 0.5ºC. Calibration tests showed that operational and design performances were almost identical (See Charts 3 and 4). (Click on the diagrams to see the full picture.)

During the design stage we produced a look up table, to convert voltage to temperature, at 0.5 ºC. intervals over the operating range of the VDO gauge. This table was dynamic, in that we could enter calibration data from the test to compensate for ambient temperature and the climatic conditions of the day.The results from the test were interpolated using the look up table and further interpolated between adjacent cells to obtain accuracies of less than 0.5 ºC.

For the test, the engine had been pre–warmed to just below 40 ºC. whilst the test box was calibrated. Using a digital multimeter for each channel we took readings at 1-minute intervals as the engine warmed up. We concluded the test after three cycles of the cooling fan. (Refer to Table 1)

There is a high degree of correlation between the temperatures recorded by the two sensors. (Refer to Chart 8). The discrepancies after 22 minutes are caused by the fact that we only recorded voltages to 2 decimal places and not 3 as previously. At worst case this data degradation could produce temperature differences of between 2 and 4 degrees between the two sensors. 

Chart 9 concluded our tests by comparing our results with the data we received from the unmodified Speed Sport and the modified Sport Carbon and Speed Sport. For consistency we chose to compare temperatures recorded by the VDO Gauge. In all cases the modified cooling system engines warmed up faster than the unmodified Speed Sport. The Thermostat opening can be seen to occur earlier and with a more marked cooling effect than the unmodified cooling system. The 2.0 Zetec engine appears to reach fan switch on temperature faster than the 1800 Engines. Different ambient air temperatures, the different methods of modification, (Alloy pipe / Mondeo bottom hose) could cause the differences. Other factors could be the efficiency of the radiators or the differences between ducted and non-ducted nose. Overall the modified engines performed in a similar manner with no signs of overheating or overcooling.

Conclusions

The Westfield cooling system can be modified using a separate bypass pipe to increase engine temperature. Some expansion will take place in to the header tank but this is normal and should not overflow providing the cooling system is not overfilled. Because of the reduced size of the Westfield expansion tank compared to that fitted to the Ford Mondeo, as a precaution, we would recommend filling and maintaining the system coolant level to half way between the min and max marks in the header tank. 

The 2.0 L Zetec Engine behaved exactly as predicted and as observed in test and road conditions. The temperatures observed on the VDO temperature gauge fitted to the car are corroborated by the test data produced and are a reliable guide to engine temperature.

Zetec 1800 engines can be modified using a modified Mondeo bottom hose, (Ford FINIS 1013384).

The modified cooling systems of 1800 Zetec engines performed in a similar manner to the 2.0L Zetec engine. Operating at 90ºC in free airflow conditions warming to 100ºC in still or low airflow conditions. The fan operated at 100ºC and cooled the engine down to 95ºC at switch off. These temperatures are consistent with published data for the Ford Mondeo

Engines fitted with Ford “Red Top (Ford FINIS 7001614) thermostatic switch need to be replaced with Ford “Orange Top” (Ford FINIS 7001611) switch to achieve normal fan operation.

Acknowledgements

E.M Engineering acknowledges the assistance given by: -

Westfield Sports Cars Ltd, VDO – Keinzle Ltd, Haynes Publishing, Ford Motor Company, Steve Richards – Oakwood Data Ltd,Andrew Mumford – E.M. Engineering, Tim Hoverd.