Hi David,
I'm pretty sure that the only difference between the early and late design MAP sensor on the V6 is the physical shape of the plug - electrically they are compatible.
Nearly all MAP sensors on modern Petrols work the same way - one pin is ground, one is a 5 volt DC regulated supply provided from the ECU that runs the internal circuitry in the MAP sensor (it has a surface mount IC and a few other components) and the signal wire is simply a DC voltage output from the chip on the sensor that varies between 0 and 5 volts with 0 being low pressure and 5 being high pressure. (Actual range will be a bit less, probably 0.3v to 4.7v)
The only difference you'll ever see in design is turbo engines will have a different calibration of the output voltage - on a NA engine the MAP sensor will have a maximum pressure reading of 1 bar, (nearly 5 volts) but on a turbo you'll have a 2 bar or 3 bar MAP sensor - in that case 5v corresponds to 2 bars or 3 bars instead of 1 bar and intermediate values are scaled accordingly but the principle of operation is the same. Obviously because of this you can't swap MAP sensors between turbo and non turbo engines even if the pinout is compatible because of this different calibration - the ECU will interpret the wrong pressure reading from the voltage.
For example a 2 bar sensor from a turbo fitted to a NA engine would only indicated about 2.5 volts at atmospheric pressure instead of nearly 5 volts, so the ECU would interpret that as a pressure of about 500mb instead of 1000mb which would almost certainly set a fault and would definitely screw up running.
The 427mb reading you saw - was that just with the key on or with the engine running ? You should see approx 427mb at idle but with the key on and the engine not running it should read near 1000mb. That's a basic quick check for a MAP sensor.
If you're seeing that just with the key on and the engine not running that's a problem... with the key just on but not running do you get a change in the reading when you unplug the sensor ? Typically the reading would jump to maximum with the sensor unplugged, or about 1100mb.
The fact that you get a different fault code with the sensor unplugged does suggest that the signal wire has continuity, however it doesn't confirm that the wiring is correct - are the ECU and wiring loom both from the same car or were those also mix and match from different models ?
It would be worth checking that the pinouts are correct - on the Xantia V6 MAP sensor the pinouts are:
1 Signal
2 Ground
3 5 volts
If the 5v and Signal wires were somehow transposed due to mixing different ECU's and wiring looms it could explain a fixed reading of 427 that doesn't change with pressure but does change when you unplug the sensor - because the power input to the sensor would be connected to the signal wire loading it down. Unlikely, but worth checking.
Have you got one of those small screw driver style automotive test lights ? If so you could do the following:
Unplug the sensor, leave the key on, engine not running. Clip your earth to the engine chassis then test each of the three contacts in the plug to find which one lights it up - this should be pin 3. Although a multimeter would probably also show 5v on the signal line on pin 1 there won't be enough current to light the bulb, hence using the bulb test to confirm which is really the 5v supply.
While you watch the Lexia pressure reading touch your test bulb between engine chassis and pin 1 and see if the reading drops to a very low figure. If it does you've confirmed that the signal wire is working and wired to the correct pin on the plug. Repeat with the bulb connected between pins 1 and 2 and it should do the same - this confirms the ground wire is correct. Finally connect your bulb between 3 and 1 and the reading should either go high (over 1000) or stay high if it was already high. If that all works you've confirmed the ECU and wiring harness is correct.
Another test to try is to remove the MAP sensor from the manifold, leave it connected to the ECU, key on, engine not running, and push a couple of feet of rubber hose onto the nub so that you can suck on the end of the hose to manually generate a vacuum. If its working properly you should get a reading around 1000 when not sucking on it, but it should drop quite low as soon as you suck on it. You can also put a volt meter between pins 1 and 2 and see the voltage vary as you do this.
If THAT works ok then we've confirmed that the MAP sensor is in fact working and communicating to the ECU. If that's the case there are another two possibilities I can think of....
1) It could actually be a faulty TPS signal. The computer constantly checks for "coherence" between the MAP sensor, engine RPM and TPS opening percentage, as there is a fixed and predictable relationship between them which is stored in the ECU's map tables.
EG at 3000 rpm with the throttle open 20% the lookup table in the ECU tells it the manifold pressure should be within certain limits, say 300-400mb. If the pressure falls well outside this range it will set a fault code for the pressure sensor. You can see where I'm going with this right...? If the TPS is lying this can trigger a MAP sensor error because the two sensors are disagreeing with each other, but because it can't always figure out which one is lying often the MAP sensor gets the blame, when it could be the TPS indicating the wrong throttle percentage.
Have you done any tests/checks of the TPS yet ? It's fairly easy to do a rough test - key on engine off watch the throttle percentage and voltage as you operate the throttle through it's full range. It should be approx 0.6v with the throttle closed and about 4.6v with the throttle fully open and smoothly varying in between.
2) If the TPS checks out there is another possibility that is a little bit disturbing - the problem could be caused by a mismatch between the ECU and the engine itself. You say you've put a Xsara VTS engine into the car with a Peugeot 306 GTi ECU, are you sure those two engines are ABSOLUTELY identical in specs ? Is there ANY difference in CC, inlet manifold design, throttle plate design, ICV, anything like that in the induction system design which would affect the breathing of the engine ?
If that's the case it could well be your problem - because the MAP and TPS coherency checks will fail if the engine breathing characteristics are too different. It expects a certain MAP reading at a certain RPM and throttle opening, but if its a mechanically different engine the reading may fall outside the normal limits that the ECU is expecting, thus it throws up a fault code even though both sensors might be working fine and the engine might be mechanically fine. (but different in characteristics to what the ECU is programmed for) In that case you might have to find a proper matching ECU.
How well do you know the provenance of the ECU and where it came from ? Is it possible that whilst its the correct MODEL of ECU, its not from the correct engine ? The same model of ECU (eg Bosch MP7.0 on the Xantia V6) is used on a LOT of different engines - all the circuitry is the same but the programming and map tables are specific to that exact engine model. Same hardware, different software. It could potentially be the right model of ECU programmed for the WRONG engine...
Hopefully that gives you a few more things to check and try!
) which equates to around 4.7 volts. With the engine running and throttle not open much it should be around 300-400mb or about 1 - 1.5 volts. Unfortunately you can't test it while cranking (and failing to start) because during a crank the ICV will be wide open so there won't be much manifold vacuum. So it needs to actually run to generate vacuum and allow the pressure reading to drop.
