Sounds like you got it under control man.
Let us know how it shakes out.

 

Does your setup still have the fuel temp sensor on the fuel rail? If so, what if connection to that was corroded or cable grounded? Might fool the ECU and impact how ECU handles the hot re-start, since that is a variable resistor sensor.

 

I'm thinking coolant temp sensor.

 

Any and all sensor connections have been kept clean and bright plus lightly coated with KOPR-SHIELD a conductive anti-corrosion surface compound.
Made by Thomas & Betts found at any better electrial supply.
This compound on average reduces connections from readings of 0.300 ohms to 0.100 ohms on average. A electrial version of aluminum based anti-seize compound but of copper powder base.

 

I have three complete spare sets of sensors from 94 & 95 3.9 D1's.
Base of goose neck, two pin coolant sensor, single pin gauge sensor plus fuel rail sensor.
All function with readings rather close to the sensors on the vehicle within a couple degrees as well resistance readings with Fluke 87 and 52 meter. Factory tolerances within reason of sensor production i've been told.

Goose neck temp sensor contacts close at 166*F open on cooling 159*F.
Two pin coolant sensor ahead of temp gauge sensor 2.3K to 2.4K at 70*F to 266 ohms at intake reading of 175*F. Single pin temp gauge 1.67K cold to 80 ohms at 210*F, in pan of water on stove with fluke 52 temp meter.
Fuel rail sensor cold 3.2K to 810 ohms at normal hot engine temp.
This stumble problem is not related to any bad sensor signals to the ECM.

I mentioned before, has anyone measured their fuel rail pressure and encountered gauge needle flutter like i'm measuring or is it a solid steady pressure reading?
At a 1 hour hot restart pressure gauge needle flutter 28 to 36 psi, yes needle over swing I bet it's 30 to 34 psi.
After 1 to 2 minutes idling needle flutter settles down to 2 psi then down to 1 psi.
Fuel pump is talking to me.
Question, what brand by manufacture of replacement fuel pump is recommended?
Been told some replacements require the power connection splice is required, no biggie.
Not a use only a Rover dealership pump reply with their inflated prices times four.

 

There have been a number of posts of users buying the Airtex pump from a parts store. Would you consider that the needle swing (pressure changes) might be related to cavitation or bubbles (hot fuel) inside the fuel lines getting compressed, then slowly being expelled through the injectors? The ECU has a plan for dealing with that hot restart with the factory setup, but have your mods changed the game to make that situation occur at a temp below the window of concern for the ECU to initiate hot retart sequence?

 

I think you nailed it with the twitching needle readings. Just for fun I did mine,:

Stone cold - overnight - "0" PSI
Turn key to make dash lights come on - fuel pump starts - PSI to 38 solid.
A few seconds later, multifunction relay goes clunk and pressure starts to drop, pump was shut down because of no crank.
Crank truck, cold idle around 1K, PSI 30 and steady, no fluctuation.
Let truck warm, shut down, pressure to 28 PSI.
Come back in 10 minutes, pressure at 30 PSI (fuel expansion)
Crank truck, again at 30, rev engine to 3000, pressure drops to maybe 26 PSI

The twitching needle could indicate one pump winding bad, so it stutters, and the effective volume will be less of course. At some point during the long overnight rest the pressure drops away on mine. This could be old age on mine, lack of service by the present and previous owners (I'm number 6 in a line of fools who fell for this rolling lump of coal teapot). I did replace fuel filter about 6 months ago.


Another thought - since the ECU deals with hot restart by increasing the pulse width to the injectors, to move the bubbles out of the system, you could see that on an oscilliscope connected to an injector wiring. A stone cold start you would see one pulse width, at "hot restart" it should be different. At least with the factory ECU program. What does your chip vendor say about how they handle hot restart in their program? The stumble certainly sounds like bubbles in the fuel.

 

This is nice knowing a fuel pump by brand name to look for.
With the little pressure swings at still over 28 psi fuel bubbles I can not see as a problem.
For the bubbles being expelled thru the injectors I find hard to believe as at idle 95% of ther fuel is bypassed back to the tank by the regulator.
Also the cooler fuel from the pump to regulator is less apt to turn to vapor plus the fuek rail can not cool down the fuel within the 10 to 20 seconds it takes to clear up this stumbling problem acting like vapor lock.
At zero or 4 to 5 psi in a carbed engine I can see vapor lock but not under 28 to 33 psi with a hot shutdown.
With the meter set to duty cycle it reads the same be it a stumbling hot restart or cleared out 15 seconds later.
These mods have been running perfectly the past two years, intake runner extension only effects max rpm TQ range.
Injector upgrade to later 4-hole Bosch vs Lucas single squirt is a way better design in atomizing the fuel. Four small streams with a shallow fan pattern vs one big stream, you pick which one will atomize better.
There is a reason why 4-hole injectors came out in app 2002 vs single squirt design of the 90's, I call it progress. As far as ECU not handling the above mods I find very remote to none at all.
To ask a question for Chris Crane at RPI (the chip maker) even with one to the point simple question you will receive a reply, "send the part or motor to us and let our tech crew tune it up". Reask same question but reworded, same useless replies besides they are in England vs me in California. Best to **** in a hot bucket and call it yellow snow.
This RPI Tornado chip has and still works flawlessly with power and TQ increases plus fuel mileage the past 12 years, kept installed even for smog testing, passes with ease plus proper F/A ratio to redline. Not the lean out of upper end rpm's of a stock Rover chip.
Read up on RPI Tornado chips you'll understand their direction, correcting fueling not for the maximum hp possible, they design practical fuel control not dyno shouting numbers.

 

With your readings you do not hold pressure overnight, I question this.
On your idle of 30 psi then rev to 3,000 with only 26 psi makes me wonder if your engine fuel demand is higher than the fuel pump volume output?
When I quick rev from idle the fuel pressure increases from 30 to 40 psi, this a fueling demand for heavy engine loading from the ECU to add more pressure.
With the vacuum line to regulator removed 39 psi rail pressure increase vs 30-31 psi.
I can see a new fuel pump in the future, plus an adjustable regulator as the 30 psi I read is already below the 33.99 to 38 psi minimum spec limits.
Why chance installing a replacement fixed regulator that might also read lower than spec limits? Covering my butt as $127 is the best regulator price found so far.
Fuel pumps I find $54, $64 to $89 plus different brands of manufacture.
I'll post a few photos once I have time and help, I promise.
.....~~=o&o>.....

 

My D1 is in a sad state of tune. But even with leak down, and even with lower than optimal fuel pressure, it does run, and consume fuel at a 14 - 16 mpg rate. All of that means little, but what is of interest is the "not twitching". Can't recall a post in the past year that referred to twitching fuel pressure.

However, my ride is capable of going 0-60 in 3.2... gallons. And I was interested enough in your situation to get out my gauge and make few test readings, in the quest of knowledge. But you may find that other forums may have a greater tech level participating on them. We are more newbies, off road modders, and DIYers over here.