I apologise for the length of this post but I wanted it to be as clear and concise as possible for all interested. I undertook what is detailed below in good faith to help me, MMC and other Roadster owners. I should point out that I take no responsibility for any issues which may become evident for an ECU fitted to a Morgan Roadster by MMC in the future.
I purchased my 3.7 Roadster last summer, it is the Roadster built early 2013 for the Marfan Charity run to Nordkapp. From early on I have had various discussions with Willams Automobiles, Steve Morris and Mark Evans on the rev hang issue with this model. MMC has now given me permission to let others know what I have been up to, we all agreed to confidentiality as no one wanted to get others hopes up and possible disappointment. For those who do not know me I have owned many powerful cars over 35 years, I have done track days, sprinted and I race in hill climbs, so am a reasonably experienced driver.
Issues as evident on my car with my style of driving was as below:-
1. When accelerating and disengaging the clutch revs would hang causing the vehicle to continues to accelerate on selecting a higher gear eg. 3rd to 4th.
2. If hard acceleration was being used as above the revs would momentarily rise 200-300 rpm.
3. It was possible due to either of the above to loose rear traction on wet or slippery surfaces due to the above.
4. When cruising (say on a motorway) releasing the accelerator to slow down would result in overrun, eg. the car continued at the same speed for several seconds.
MMC had been working on an ECU development for the Aero ARV6, there was an opportunity to possibly develop a simpler solution for a road going 3.7 Roadster requiring just an ECU change. I offered my car to trial this solution once it was ready and MMC were satisfied that testing of it was sufficient to have confidence putting it on a customers car for trail. Prior it was dyno tested on an engine and was run on two other factory cars before mine. The development in my opinion undertaken by MMC and MBE (the ECU developers) for the past 8 months has been significant and has not been simple. The Ford supplied ECU is effectively locked down and as such any replacement has to work in the same format to enable it to interface with the Roadster instruments and engine monitoring systems/sensors. I cannot go into technicalities (which although I work in IT) as I am not skilled in this area, although I do understand the underlying issues to replicate and emulate a replacement ECU which is then configurable.
In early January this year Mark Evans at MMC asked if I was willing to conduct the trail, which I agreed to do. The new ECU was fitted by Chris Bridgens of MMC and setup in conjunction with Jeff of MBE at the factory. Possible measurements were discussed on how to judge its success, and all agreed it would be best for me to monitor and feedback how I felt best, so as to not lead me into any conclusions. I took the following measurements in detail.
Every trip undertaken was measured accurately using a GPS system for mileage and speed.
Indicated Speedometer, Odometer, Mileometer, Average MPG, Economy readout, and Temperature readings were monitored and recorded.
Any performance issues or problems with the car were noted and reported
Fuel consumption was measured accurately on fill ups to ascertain accurate mpg and comparisons with indicated mpg on the car.
Over the next 8 weeks I drove a varying length of trips, on motorways, A roads, local urban driving and to and from work, ranging from a few miles to 160. Conditions varied, dry, wet, and on the edge of icy. When able I pushed the car, accelerating hard to try and induce Rev hang and other issues previously mentioned. At the end over testing over 1700 miles were completed.
Testing Results1. From day one of the new ECU the Rev hang and points 2 to 4 were gone.
2. Initially the engine would occasionally cut out on start up, this was rectified on the second visit to MMC.
3. I identified anomalies with MPG, Economy meter, Odometer and Mileometer readings against actual. These were addressed and rectified on the 2nd visit to MMC, with changes to the ECU programming.
4. After the second visit to MMC I experienced occasional cutout on slow approach to junctions. This was rectified after the final visit to MMC, by adjustment to ECU cam sensor mapping, an anomaly only experienced on my car to date. This relates to High Tension electrical noise transfer from the HT coil cap nearest Cam 1. MBE informed us that the new ECU has inbuilt filtering and this was tuned to removed the noise for all Cam sensors, they have experienced this before with ECUs requiring a loom harness adapter.
5. After the second visit to MMC I experienced once the car going into limp home mode - fixed 2000rpm (occurs when ECU is confused or unable to get accurate readings). I have been advised this can be forced to occur on any 3.7 Roadster or Mustang if the accelerator pedal is stamped on, effectively switching it from low voltage to high very quickly. It can assume there is a fault and for safety puts the car into limp home mode at 2000rpm. Adjustments and checks were made to ensure that the ECU would react correctly and put the car into limp mode when appropriate.
6. After the 3rd visit of the car to MBE and MMC unfortunately the car went into limp mode once unexpectedly and once deliberately induced by myself. The car returned for a 4th visit on 3rd March to MBE & MMC, after this MMC discussed with Ford and it was established that there are 3 iterations of the ETB(accelerator) and some early Roadsters had the 2nd, the case with my car, this requires slightly less sensitive ECU voltage parameters, and MBE made the required changes. I collected the car on 13 March and have had no limp home mode re-occurrence to date, or any other issues. If I do find any I will report back.
SummaryMy final testing has shown that the fuel consumption has not been affected and I believe is as before the ECU change, performance has not been affected and if anything I feel the car is smoother under acceleration and just as quick to respond. My Roadster is now lovely to drive and gear changing can be reasonably quickly, just how it should be in my view. On the final visit to MMC arranged for the car to have an emissions test at a local MOT centre and the car passed. There is one minor issue outstanding, the Economy meter does not work correctly, I am not concerned about this as I do not consider it important on a Roadster, MMC may or may not choose to investigate.
Summing up the Rev hang and other issues 2-4 above are no more, the car is a pleasure to drive and I am now very happy with the way it behaves and is as one would expect. I would recommend that any 3.7 Roadster owner who has the aforementioned issues
talks to their main dealer and MMC to get the changes required applied to their Roadster.
In my opinion MMC has undertaken a great deal of work and investment into getting a solution for this problem and some may not appreciate that it was a difficult and challenging thing to do. I would like to take this opportunity to thank Steve Morris, Mark Evans, and all at MMC who worked on my car and Jeff at MBE who worked on the ECU, and the trust they placed in me to do the trial. We all think that the result is good and should be a generic fit for any 3.7 Roadster.
I think that Morgan in conjunction with Aero Racing have come up with a very satisfactory and cost effective enhancement to the 3.7 Roadster.
Final ResultsMark Baldwin on 12 March took my car home and drove around 65 miles round trip and tried to induce the limp home mode and was unable to do so.
Drive home from MMC on 13 March a total of 153.1 mile(GPS), recorded odometer miles 152.9.
This journey involved 90 miles on M5/M4 at 70+mph, and 10-15mins in heavy near stationary traffic, a detour to Williams Automobiles, and the remainder was on A roads at 30-60mph.
Fuel fill up 25.6 Litres or 5.6 gallons, average calculated mpg 27.38. Indicated mpg 30.9. The MPG anomaly I suspect will be due to fill up discrepancy of around 2 litres from when MMC filled the car and I did. Previous recorded fuel usage during the trial has regularly indicated average mpg of 30 to 31.
I also performed several hard acceleration tests in different gears when able, revving the car up to 6500rpm and was unable to induce the limp home mode.
Two further short drives undertaken locally in drier conditions on 14 March and once again under hard acceleration and no limp home mode has been induced.
I am planning to drive my Roadster this coming week to the Exeter Park and Ride using dual carriageway and A roads and will report if I have any issues as a result but I am not anticipating any.
Emission Test ResultsConducted at Malvern Motor Services on 12 March 2015
Fast Idle Test 2500-3000rpm
CO = 0.09 Pass max 0.20
HC = 37 Pass Max 200
Lambda = 1.022 Pass Min 0.97 max 1.030
Natural Idle Test 450-1500rpm
CO = 0.02 Pass Max 0.30
Happy driving
