Philip Lochner 1995 Land Rover Discovery 3.9 V8i

[Philip Lochner]

Last Update: 3 March 2008

Background

Land Rover Discovery '95 V8 3.9 8.35CR port injection
I have now squirted two of these cars (I'll call them D1 and D2) and will write about them both here so as not to have too many similar success stories.

MS setup for both: MS-2/V3/2.862/EDIS-8 MT2.25p1

Info pertaining to D1:
Info about D2 follows much further down.

A friend was suspecting that his Disco was running unneccessarily rich. I installed an Innovate LC-1 WBO for him and this confirmed his suspicions. We used output 2 and programmed it so that 1.0V = AFR10 and 2.0V = AFR 20. This way, we could use a cheap digital voltmeter to read the AFRs all the time without the use of a PC .

The "leanest" he ever saw was AFR = 14.0 (as I did on my car before I installed MS-2) and this helped him make up his mind to ask me to install an MS-2 for him. But...

He wanted to retain the OEM ECU as a backup so I had to wire it in, in such a way that the OEM ECU could still be put back into service with minimal fuss. This took a bit of careful planning as the OEM ECU also controls the main load relay, engine load relay (12V to injector rails), input from A/C control logic relay and the AC compressor clutch relay based on input from A/C control logic relay. The first two are both controlled by the Ms-2 FP output. The AC clutch relay was a bit more interesting. I used the VB921 circuit with a BC109 tranny to get this to work.

Update: 3 March 2008
Had the car on the dyno recently. Biggest surprise of the day was finding WOT max power at 1000rpm with 26deg BTDC advance!! Max power at full throttle (all rpm) typically came in at 13.5AFR but we saw that one could lean it out to about 15.5 before losing 2Kw in each case.

Update: 3 Dec 2007
Conversion to EDIS-8 completed. Idles more smoothly. Timing visibly more stable with timing light and by looking at the rpm gauge (with lag factor at 90%). Timing map has to be tuned on a dyno otherwise there is too much guesswork. Current timing is a bit more aggresive than before but not much. Because the engine is low CR, there is no pinging to tell when advance is too much.


Update: 9 Nov 2007
Owner reports 7.6km/l at steady 120- 130km/h on open road.
Conversion to EDIS-8 in progress.

Update: 23 Jul 2007
Implemented fw 2.862. IAC now works 99% correct. From a cold start it works as desired but I'm using minimum 8 steps. From intermediate temps it might not implement the last batch of steps, that batch being somewhere between 1 and 7 steps which would cause IAC to remain more open than desired causing slightly higher idle speed but at least its not annoying.

Also found that setting the IAT temp table was important as we experienced lean idle after some heatsoaking. I also replaced the closed element MAT sensor with an open element sensor as the closed element sensor was just too slow to react.

Also moved from 100% TPS based AE to 100% MAP based AE and got much better results much easier!!

Update: 7 May 2007
Implemented IAC stepper over the weekend. Cured RPM spikes by setting "time mask" to 2ms (was 1ms before). Improved (enriched) AFRs particularly at 40kpa and below. Improved AE (still needs a bit of attention).

IAC (GM with square plug) still does not work properly when asked to close during warmup with small increments. This means engine requires a re-start when warm as IAC does not seem to move with small increments from the cranking position after cold start. Hot starts are OK and we do not close it fully when warm to allow air in addition to the base idle circuit.

Update: 1 May 2007

Timing
We decided MS-2 would not control timing as the OEM ECU was not controlling timing and thus this MS-2 is triggered off the coil- terminal.

So, to run on MS-2, the ECU needs to be unplugged. To run on ECU, the MS-2 needs to be unplugged. No other plugs or wires need to be fiddled with.

Installation (yesterday) took one day although he had done some wiring preparation beforehand.

Tuning
Having developed a set of AFR ratios for my own Disco, we applied these "as is" to tune his car, with me driving around town/highway and him operating the PC. EGO feedback was used as an indication to go leaner or richer which went rather quick (less than 1 hour).

First impressions
This morning he reported the car is noticeably more powerful at partial throttle and on the whole more repsonsive. He described the car as being a bit lethargic on the OEM ECU despite LOTS of fuel being applied under all circumstances. We applied the same AFR ratio (12.5) at WOT, so we are not getting more power at WOT. The leaner mixtures (14.7 - 17) at partial throttle interestingly delivers more torque.

Remaining issues
1) IAC stepper must still be put into service
2) Accel enrichment requires further optimisation
3) Now and then RMP spikes occur which seem to co-incide when AE is applied
4) Baro correction table (can only be done when travelling to lower altitudes)

Info pertaining to D2:
With the experience gained on D1, I was able to install this MS/EDIS rather quickly with no hickups during the install whatsoever. Engine fired on the first crank.

I'm using MAP/Baro and "Seperate VE/AFR tables" with 4 SIMULTANEOUS squirts on this car, having found more torque and better throttle response on my 4.6 using 4/simultaneous.

Only thing that is very weird about this install is that my 4.6 and D1 are both very happy with 12º BTDC @ idle. Not this one, this one wants 22º before it idles happily!!!! Dont know what cam this car has but its not a wild cam although it does pull very well for a 3.9.

Owner reports similar performance at WOT but more at higher revs than before but much better responsiveness and power at part throttle.