LUCAS Epic start trouble

[NielsDK]

The car is a FIAT Ulysse, which is identical with Citroen Synergie (Eurovan, Evasion) or Peugeot 806, so hopefully it qualifies as french enough for the forum.
It has the 2.1 TD engine, now clocking 339000 km (210000 miles), and in general I'm very satisfied with it
However, June 6th I parked at the local COOP, and returning from the store I was utterly unable to start the car. It had been running with absolutely no trouble 30 minutes before. Starter motor running fast, lots of battery juice, no firing at all. Two and a half hours later, with the assistance of Falck, the Danish version of AA, the car started after activation of the collision fuel cut-out switch. Previously, I did not know of the existence of this switch, maybe I should have read the manual.
Then everything went fine for another month, but then all of a sudden I was once more unable to start, quite shortly after having stopped the engine, which was running perfectly OK. Happily, I went and pressed the collision switch - but this time it did not work. The car was transported to a mechanics' shop, and left outside, since it was after hours. Next morning, I went there on my bicycle, to explain and demonstrate the problem, and agree on the mechanics' actions. However, as I was demonstrating, the engine started trouble-free.
This was repeated four or five times, the car cannot start shortly (up to several hours) after having been stopped, but will start easily the next day.
My local diesel pump expert (although maybe a greater expert in Bosch pumps than Lucas ones) would not even take the car in for diagnosis, explaining that he was very certain that the pump was worn out and needed replacement, and that a diagnosis would only increase the cost (which, to my estimate, will anyway exceed the value of the car). I did change the pump once before, the old one failing at 298000 km, probably the vane wheel pre-pump, as the filter and the tank wer fille with a couple of spoonfuls of steel chips. I have not disassembled the old pump yet, but I shall in order to satisfy my curiousity. At that time I was lucky to get a used pump, mileage unknown, for a cheap price, about £ 100, from a failed engine. But I do not expect to be that lucky a second time.
I did not and I still do not think that the symptoms are like those of a worn pump, and having no certain opinion of which of the ESOS valves or maybe the timing or pump rotor position sensor that was malfunctioning, I decided to exchange them all with the ones from the old pump. Anyway. they come in a set, being soldered to their wire harness.

It worked, now the start problem is history.

Does anybody (maybe CitroJim, the Epic enthusiast) have a good guess at which valve or sensor is the troublemaker?

And, by the way, if I should need a replacement pump some more miles into the future, does anyone know, which pump types will do as replacements. Mine are marked TYPE XUDL.P01 EPIC R8640A102A, but Eurovans (Ulysse's, Synergies, and 806s) are less common than XMs, Xantias, 406s with XUD11 2.1 l TD engine, so the chances of finding a used pump may improve with the car population.

Rebuilding for a mechanical pump is not an issue under Danish rules, this would require a new approval and registration of the car, and the cost is prohibitive. And veg oil running is strictly forbidden, this is a tax evasion issue. But it seems that refitting of mechanical pumps is popular in the UK, does that leave you with an abundance of Lucas Epic pumps?

Finally, does anybody know the proper opening pressure for the injectors?

[CitroJim]

An interesting one Neils.

I've known Bosch pumps be harder to start when hot but never an EPIC. It doesn't sound like a worn pump to me either, nor worn ESOS valves. The normal symptom of worn ESOS valves is a failure to stop immediately and timing issues.

A worn pump is unlikely to generate enough delivery pressure to start either hot or cold.

Many, many issues with EPIC pumps are down to air leaks and this needs to be carefully investigated first. Many EPIC pump installations have a transparent pipe between the filter outlet and the pump fuel inlet. This is perfect for observing any air in the fuel lines. If yours does not have such a pipe then fit one, if only for test purposes.

Also, the leakoff pipes between injectors can be a source of air ingress and are best replaced. The biggest culprit in this area can be the little blanking cap on the last injector which can split.

Any air ingress will cause the internal pump hydraulics (which use the diesel fuel as the hydraulic fluid) to fail to work correctly. and cause fuel to drain from the pump and fuel lines back to the tank.

So, check for air ingress first.

Also, try vigorously pumping the priming bulb to see if a bit of assistance to the lift pump helps starting.

It may not be the pump. You may have a very weak timing signal from the crankshaft sensor which is good to start and run when cold but too marginal when hot. Also, low compression when hot can also cause bad starting. This is rare on a 2.1 but needs to be eliminated if all else checks good. Finally, glowplugs are required to start the engine whether hot or cold. An Indirect Injection engine such as the XUD11 will never start without adequate pre-heating.

Any EPIC pump originally fitted to an XUD11BTE (2.1TD) engine should be good as a replacement.

I have no idea what the injector opening pressures are unfortunately.

I trust that helps a little Neils?

[kenhall1202]

Adding my small two penn'orth to Jim's comprehensive advice above, the Delphi injector needle lift pressure is quoted as 163.5 +- 3.5 bar (source Citroen 2001 Car Book 2)
Link http://www.peli59.net/xantiafaq/XANTIA_ ... N_2001.pdf" onclick="window.open(this.href);return false;

[NielsDK]

Thank you Jim, thank you Ken.
+/-3.5 bar seems a quite tight tolerance, I wonder how many 1/100 mm of shim that will correspond to?
Glad to learn that any EPIC from an XUD11BTE should do, in case I will be needing another one.
I have transparent pipes, between filter and lift pump, and also between bulb and filter, and the return line from the lift pump to the tank, the first distance at least. Not much air in the fuel though, not when the starting failed and not now, when the trouble seems to be over.
Pumping with the bulb, and venting the lines, did not help at all. But changing to the old valve/sensor set, from my old, failed pump, did the trick.
Therefore, I do not really believe in a worn pump, or air, being the cause of the trouble. But anyway, I will renew the leakoff pipes (rubber hoses, I think, having become quite hard with age).

At engine stop, I think the stop solenoid is de-energised, and the small spring will push the viton? rubber tip aginst the inlet hole for the high pressure pump, preventing filling of the pump and the return stroke of the opposed plungers (if I have understood the pump design). At the same time, I imagine that the pump control could stop the drain of diesel from the rotor cavity, and open fully for the inlet, moving the rotor to zero index position. If, then, either the stop solenoid sticks in closed position (it did seem quite loose at removal, though), or the feed or drain valve sticks, I think it could account for no fuel delivery. Or, if the rotor position sensor gives a wrong feedback signal, making the ECU believe that the fuel index is good for starting, although it may in reality be zero? But anyway, this is guesswork.

Starting is now fine, cold or hot, and I will enjoy every one of them. By the way, compression must be okay (it could not be repaired by changing the valve set, and idle running was always stable and slow). When the engine is hot, it will start without waiting for the glowplug light to go off, but I understand that the electric heating will be active anyway for a couple of minutes (I think from an earlier reply from Jim?)

Weak signal from the crankshaft position sensor - I would think that would be revealed by a fault diagnosis readout - am I right? Or by the orange lamp coming on in the dashboard during running (which never happened).

I am still puzzled, but my engine is running.. maybe we need a Lucas EPIC forum?

[CitroJim]

That's getting more interesting Neils

So, I take it you swapped the ESOS valve set complete? Did you renew the seals and O rings or use the existing ones?

The ESOS valve with the red Viton tip is the stop solenoid and if this does not seal properly on an engine stop command then other measures are used to stop the engine which results in the delay as the governor valves are used instead. The Engine Management Light illuminates if this is the case.

I'm still trying to work out how swapping the ESOS assembly may have helped. I've known solenoid coils fail when hot in other areas and I really do wonder if the stop solenoid was marginal enough that it could pull and hold the stop plunger in when cold and running but unable to do so when hot.

Yes, even on a hot engine post heating will be in effect for a while - certainly whilst the engine is idling.

I really trust all continues to work fine from now on in Neils

[kenhall1202]

"+/-3.5 bar seems a quite tight tolerance, I wonder how many 1/100 mm of shim that will correspond to?"

Approximate rule of thumb is 0.05mm change in shim thickness will change pressure by 3bars.

[NielsDK]

Thanks once again Ken and Jim.
Now I'm starting without problems, there is really not so much urgency. I was asking of the opening pressure of the injectors, because I have kept the old ones that were installed with the pump that failed. However, I used the ones that belonged together with the second hand pump, becaused I worried that the original ones maybe had been filled with steel chips same as the fuel filter. Likewise, I also changed the high pressure pipes. However, now I have cleaned the pipes and injectors thoroughly, really without finding any foreign matter in them. First issue is to check the opening pressure, I think I can borrow a rig from a friend, who made it from a cheap hydraulic jack, a manometer, some fittings and a wooden board. Next, I should find out where to obtain the proper shims needed for adjustment - but really, no haste.
Jim, yes, its true, I swapped the complete set of valves, including the two position sensors (not knowing which of the valves or sensors that had the intermittent, maybe temperature dependent, fault), and to save me the trouble of soldering. Regrettably, it means that I still do not know, which component failed. But I do not regret too much, since the repair was a success. And I re-used the old seals - I had no new set, and although they were a little flattened, they seemed to have some elasticity left, so I took the chance. But like the renewal of the leakoff pipes, a new set of seals is on the to-do list (for the next time trouble occurs - I usually do not make repairs to things that already work).

I did not pay sufficiently careful attention to the dashboard lamps before I changed the valve set, so I do not really know whether the Engine Management Light came on at engine stop. I know that the stops were quite immediate after turning the key to the off position, and I know that by now, all lamps are off immediately, when the key is turned.

The stop solenoid was certainly not sticking, actually the core with the viton tip tends to fall out, when the valve assembly is removed. The force of the closing spring is also very low. And I think the solenoid must be energised continously to keep the stop valve open, so a failing electric connection should occur at the moment of stopping the engine, and then only reestablish connection after cooling? I do not think this is very probable. The engine never stopped spontaneously while running, as I think it would have, if the electric connection to the stop solenoid failed when the pump reached normal working temperature. The only, although far fetched, possibility, I see, is that the pump will heat a little more after the engine is stopped, because the fuel flow cooling of the pump stops simultaneously. This extra heating should then cross the threshold of the temperature dependent bad connection? That is surely one for the Prince of Darkness.

Anyway, before the valve exchange, I tried to cool down the high pressure pump housing, using a small length of hose and about two gallons of water, allowing it to trickle down around the pump body in a couple of minutes - but that did not help, which is good, because I would look very silly, if I had to follow that procedure every time I would start my engine

[CitroJim]

I usually do not make repairs to things that already work.

If it ain't broke, don't fix it! Very good advice Neils!

That is surely one for the Prince of Darkness.

Joseph Lucas. He's well known as the Prince of Darkness

That's all most interesting and quite a mystery. One for Sherlock Holmes! So we can't accuse any ESOS valve of binding when hot but I do wonder if perhaps they have back-EMF suppression diodes built into their coils (as do Hydractive Electrovalves) and perhaps if such a diode is present it is failing when hot but that would be unusual. Diodes usually fail permanently and don't become heat-sensitive but we must remember this is one of Joe's EPICs we're speaking of here so anything could be happening