2.5 diesel warm starting - I hesitate to raise this issue again, but....

This site contains affiliate links for which LandyZone may be compensated if you make a purchase.
The fleabay bodge....fools the EDC into thinking the engine is cold and as such the EDC enriches the mixture and lights the glowplugs....

In normal operation the EDC doesn't light the glowplugs when the engine is warm/hot and doesn't enrich the mixture and due to wear in the FIP timing chain, the fuelling is out slightly and causes poor starts....as you have read above!!!

The best one to look into is the one with a timer circuit on, as this will take over from the EDC for 30 seconds or so, then hand back to the EDC for normal running.....the others fool the EDC permenantly so the EDC will constantly overfuel all the time so avoid this one....
 
jerrytlr said:
So the TB is basically saying (1) fuel pump, if not that try (2) leak off pipes, if not that then.... erm... no idea, give us a ring ;-)

Would be great to get hold of the questionnaire data, sort it so it only covers hot start issues, and see what the fixes were.

Wammers - any idea how much it would cost to get somebody with the correct tools (presumably a Bosch diesel injection specialist) to check and if necessary adjust timing? Spud's post suggests this could be a fix for some of us at least... the problem with this, though, is that we have no idea what his pump is now set to, other than whatever the setting is, it works and gives good readings on Faultmate! It is of course quite possible that the pump has to be set outside the recommended spec to achieve this.......

So, we need somebody with full fat Faultmate AND the injection timing tools, then we can get to the bottom of this. Any volunteers? :)


Cheers,

Jerry
Click to expand...

About two to three hours would be charged to check, and reset if needed satic timing, i would think.
 
jerrytlr said:
Also it wouldn't be that hard to find out if it is rich mixture, glowplugs, or both that are needed to treat the hot start symptoms - with the eBay bodge kit fitted, is there a relay that can be removed to disable the glowplugs? If so, anybody care to remove it and see if the car still starts fine when hot??

And the other way round - can the glowplugs be manually forced on easily? If so, I will happily try this as I don't have the eBay bodge fitted (yet) and see if it starts properly when hot just with glowplugs (no extra fuel)

Tests like this can sometimes help to get to the bottom of the issue.....


Cheers,

Jerry
Click to expand...

A hot engine should not need glow plugs. I know of one that is always started cold with Easy Start:eek: cos the owners is too tight/lazy to do the plugs, it starts fine when it's hot or just warm even in winter with no assistance from Easy Start.
 
I put the fix on from tigapiglets thread. It did the job but idles higher and is smoky.

So I read tigapiglets thread again and came to a new fix based on mention of possible temp sensor failure.

This involves eliminating the solenoid and diode from the above fix, by connecting wires to the ends of the grey/blue wire from the ecu box behind the battery.
These extend into the dash area and are fitted with a normally closed push button or a spring toggle.

When the engine is hot you press the switch before turning the key, which will give you the glow plugs, after starting release button and everything is normal again.
It works for me.

I know it is perhaps equivalent to fitting a timer to tigapiglets circuit which would be the auto version - like the ebay offerings

While this gets me by & I have done a lot of other research I am now inclined towards SpudH's pump advance solution.
 
SpudH said:
I completely cured my starting problems back in January by adjusting the static timing of the pump.

Mine took a lot of cranking when hot and if it was real cold needed two goes of the glow plugs to get it going. I mistakenly replaced the glow plugs with spurious ones and was blaming them for the cold start problems but they've been just fine since I sorted it. Now it starts a dream hot or cold. It'll even start withou a full go of the glow plugs from cold. Transformed!!


I've 144k on the clock, bought it with 70k so I've watched the hot (and cold for that matter) start issues develop over its life. I even bought the hot start 'fix' off ebay in preparation for the time it was going to be required but held out due to misgivings about the side effects. Also had the benefit of advice from a maestro called 'Chas' on the Blackbox-Solutions support forum who regularly chastises people to fix known faults before trying to deal with symptoms. Smart advice that has stood to me over 4 years of P38 ownership.

I've got faultmate so had the benefit of being able to observe a readout called 'timing modulation' from the EDC and observe that it was out of spec. From what I've gleaned this will be related to stretch in the timing chain or a pump that has been out and not timed exactly right on refitting.

Anyway, mine was out of spec and I sorted it by the simple procedure of physically rotating the whole pump where it sat just by loosening the retaining bolts. Doing this does not require any special tools (I thought it did, and had the tool for retaining the timing sprocket but this is not actually required). Getting at the bolts is awkward and you must remove the inlet manifold but that's about it.

Its an iterative process of looseneing the pump, rotating slightly, tightening back up, start it up, check the reading from the ECU with faultmate (read at a warm idle), shutting it off loosen bolts rotate another bit or back as required until the timing modulation reads in spec. Tighten everything up, put manifold back on (the p38 will start and idle just fine with the manifold removed so no need to put this back on everytime) and hopefully it will start as good as mine.

I don't know if this is a cure-all or if there is another type of hot start issue related to actual wear in the pump (the one that adding vegetable oil su[pposedly cures - low pump pressure) rather than the timing chain but this procedure has turned mine from a poor starter to great starter in one January afternoon.

HTH Spudh
Click to expand...

Well, I spent a very fruitful couple of hours today with Datatek and his Faultmate connected to my car.

First off - what an incredibly helpful person he is! Thank you Keith

We compared the timing modulation figure of his car with mine, and mine is much higher, and well out of spec. His car starts fine hot or cold, mine cranks for ages when hot....

So my next move is to adjust the timing using Spud's method and see if that improves things at all...

I pondered this a bit on the drive back home from Data's house.

If I understand it correctly, the electronic injection control on our diesel engines has a simple closed loop system - the ECU instructs the pump to advance by a certain amount, and then measures (using inj 4 and presumably crank position sensor) when the pump actually fires, and compares the two. On data's car, the figures are similar. On mine there is a considerable gap (this is at idle).

Here is some speculation now - I would expect the ECU to continue to advance the pump until the correct advance is measured at inj 4 - in this way compensating, for example, for wear in the timing chain/pump etc. It could be that this compensation work OK under load and at higher engine revs, so the effect of the timing error is negligible (most hot start problem engines seem to run fine when they are actually running...). However, perhaps when cranking, the compensation cannot be made by the ECU?

Anyway, whether this theory is correct or not, it seems the next thing to try is adjusting the pump to bring the modulation back into spec (if possible) and see what happens...

Thanks again data, and also Spud for sharing his fix with us on here....

Cheers,

Jerry
 
jerrytlr said:
Well, I spent a very fruitful couple of hours today with Datatek and his Faultmate connected to my car.

First off - what an incredibly helpful person he is! Thank you Keith

We compared the timing modulation figure of his car with mine, and mine is much higher, and well out of spec. His car starts fine hot or cold, mine cranks for ages when hot....

So my next move is to adjust the timing using Spud's method and see if that improves things at all...

I pondered this a bit on the drive back home from Data's house.

If I understand it correctly, the electronic injection control on our diesel engines has a simple closed loop system - the ECU instructs the pump to advance by a certain amount, and then measures (using inj 4 and presumably crank position sensor) when the pump actually fires, and compares the two. On data's car, the figures are similar. On mine there is a considerable gap (this is at idle).

Here is some speculation now - I would expect the ECU to continue to advance the pump until the correct advance is measured at inj 4 - in this way compensating, for example, for wear in the timing chain/pump etc. It could be that this compensation work OK under load and at higher engine revs, so the effect of the timing error is negligible (most hot start problem engines seem to run fine when they are actually running...). However, perhaps when cranking, the compensation cannot be made by the ECU?

Anyway, whether this theory is correct or not, it seems the next thing to try is adjusting the pump to bring the modulation back into spec (if possible) and see what happens...

Thanks again data, and also Spud for sharing his fix with us on here....

Cheers,

Jerry
Click to expand...

The pump has to be set to the correct start point static, with the engine at TDC and locked in that position. Depending on mileage it is set to either 0.95mm lift +- 0.02mm on engines with less than 20k. Or 0.90mm lift +- 0.02mm for engines with more than 20k. You cannot do that with a Faultmate.
 
wammers said:
The pump has to be set to the correct start point static, with the engine at TDC and locked in that position. Depending on mileage it is set to either 0.95mm lift +- 0.02mm on engines with less than 20k. Or 0.90mm lift +- 0.02mm for engines with more than 20k. You cannot do that with a Faultmate.
Click to expand...

Wammers, here is my thinking on this: the faultmate live data tells you what the injection timing is while the engine is running, it also (if I understand it correctly) tells you expected vs. actual timing, with actual measured from injector 4. Surely setting it up using this live 'engine running' data is as good as, or even better than, using the static value? The only downside of this method is that it takes time, as it is a trial-and-error approach.

If you think my assertion above is incorrect, please let us know why, as I fully admit my understanding of how the injection control system works on the P38 is not 100%. And I have a lot of respect for your knowledge of all things P38!

Cheers,

Jerry
 
meltontrees said:
Try changing your injector run off pipes, i did mine and it stopped the problem hot and cold straight away. cheap fix too.
Click to expand...

Yes it is about time I checked these - however I thought leaky pipes gave the starting problem both hot and cold? Mine starts fine when cold, so I have always assumed that the run off pipes must be OK...

I'll probably change them anyway - will be good preventative maintenance if nothing else!


Cheers,

Jerry
 
jerrytlr said:
Wammers, here is my thinking on this: the faultmate live data tells you what the injection timing is while the engine is running, it also (if I understand it correctly) tells you expected vs. actual timing, with actual measured from injector 4. Surely setting it up using this live 'engine running' data is as good as, or even better than, using the static value? The only downside of this method is that it takes time, as it is a trial-and-error approach.

If you think my assertion above is incorrect, please let us know why, as I fully admit my understanding of how the injection control system works on the P38 is not 100%. And I have a lot of respect for your knowledge of all things P38!

Cheers,

Jerry
Click to expand...


Because by static timing the pump you are setting a start point. By doing it from live data you maybe adjusting it to allow for a faulty advance solenoid, or a faulty number four injector. If you don't set it to static you will never know if you have a fault by reading live data.
 
Last edited:
wammers said:
Because by static timing the pump you are setting a start point. By doing it from live data you maybe adjusting it to allow for a faulty advance solenoid, or a faulty number four injector. If you don't set it to static you will never know if you have a fault by reading live data.
Click to expand...

Good point Wammers.

I was poring over a Bosh EDC manual last night trying to get a handle on pump timing and to be honest there were too many variables to get my head around it:doh: I'm beginning to think that rotating the pump might put more pressure on a tiring advance solenoid spring.

I'm a big believer in empirical proof in all forms of engineering and the burden of proof was that adjusting the timing in the way I did it worked exceedingly well for me as for others on the BBS forum who did it.

But..... your above post is spot on.

While wading through the EDC manual last night it was immediately apparent that electronic timing has far more variables than can be measured by one live reading from one sensor. The timing modulation is a combination of modified and combined signals from the crank and needle lift sensor. The expected timing is taken from the fuel map which introduces engine temp, fuel temp, revs, loading into that equation.

This is further complicated by the fact that during starting (until it detects a rev from the crank sensor) and on the over run the needle lift sensor will not return a value as fuel is shut off.

So a fault in any of the sensors is going to affect the timing map (it has failsafes and unbelieveably will drive on until the last sensor standing falls).

There are other factors which affect hot starting too, the initial fuel cut off is overcome by fuel pressure on the low pressure side of the pump so a bad lift pump can affect starting. Also as described above there is no needle lift signal during initial cranking so a faulty crank sensor (I've had 2 go intermittently faulty) will not tell the fuel shutoff solenoid to open sending fuel to the needle lift sensor in the no.4 injector.

So as you've said, the only true way to set static timing is the correct static procedure described in RAVE. This gives you a starting point and will help diagnose whats actually wrong with the engine.

I suppose its like the difference between conventional and holistic approaches to medicine. The Faultmate method treats the symptoms, the Wammer method will help find the cause:D



In the meantime I'm going to go out and listen to my symptom cured P38 start first touch of the key:p:p
 
Exactly Spud, static pump timing is the start point that everything else works from. Pump timing whilst the engine is running, is monitored and adjusted to an nominal best point, according to readings from numerous sensors. If you set static as it should be set. Then read live data and the reading is out, you know you have a problem. Setting it static is much more accurate than doing it using the Faultmate even if everything is ok. Where in the acceptable modulation range is either 0.95mm or 0.90mm +- 0.02mm. You are setting cam lift to one thousandth of an inch. A much more accurate way to do it.
 
I Know noting about all this most of it is gobbledygook to me, but if its other things (i have been following this post honestly) then how is it the fleabay bodge fix work then? can someone run it by me (not the glow plugs bit, that I get)
 
The ECU measures the engine temperature by measuring the resistance of a 'piece of wire' in the temp sensor. The resistance drops as temperature rises.

The bodge uses a relay to switch in an approriate resistance to fool the ECU into thinking that a hot engine is cold. The ECU sees a cold engine so switches to a cold fuel map and pump timing and turns on the glowplugs.

The first versions out did this on a permanent basis so the ECU always saw a cold engine. This was a recipe for things to go wrong as the engine constantly ran on cold engine timing and fueling leading to poor performance and increased fuel consumption.

The later versions used a timed relay to swith the resistor back out again after a few seconds. Good idea but I'd hazard a guess that the relay itself and additional wiring intoduced additional resistance into the temp sensor circuit so the ECU will never see the correct temp reading.

If I'm right then this means that if you fit the bodge, your engine will never run on the correct fuel map meaning reduced performance.

Thats why my 'timed relay hot start fix' is still sitting in the jiffy bag it was posted in 18 months ago.
 
SpudH said:
The ECU measures the engine temperature by measuring the resistance of a 'piece of wire' in the temp sensor. The resistance drops as temperature rises.

The bodge uses a relay to switch in an approriate resistance to fool the ECU into thinking that a hot engine is cold. The ECU sees a cold engine so switches to a cold fuel map and pump timing and turns on the glowplugs.

The first versions out did this on a permanent basis so the ECU always saw a cold engine. This was a recipe for things to go wrong as the engine constantly ran on cold engine timing and fueling leading to poor performance and increased fuel consumption.

The later versions used a timed relay to swith the resistor back out again after a few seconds. Good idea but I'd hazard a guess that the relay itself and additional wiring intoduced additional resistance into the temp sensor circuit so the ECU will never see the correct temp reading.

If I'm right then this means that if you fit the bodge, your engine will never run on the correct fuel map meaning reduced performance.

Thats why my 'timed relay hot start fix' is still sitting in the jiffy bag it was posted in 18 months ago.
Click to expand...


Naah nothing like that Spud. Hot wire in MAF is to measure air flow. Hot fix tells ECU engine is cold when it is hot, overriding the ECU engine temp sensor. ECU then enables glow relay. In otherwords sets up for a cold start. Then switches off after a few seconds. Never ever fit one without a timer in it.
 
Spud I did have some coms a while back with the guy that sold them on flabay, and he copied the thread (you'll see I did contribute to it quite a lot) I got a guy who i know from the Series II club to figure it out with diagrams etc, and the flabay guy basically copied what was put on this forum (and more power to him for doing it and then making £££ out of it) he then improved on it a few time as well

But thanks for clearing that up for me I understand it a little better now
 
wammers said:
Naah nothing like that Spud. Hot wire in MAF is to measure air flow. Hot fix tells ECU engine is cold when it is hot, overriding the ECU engine temp sensor. ECU then enables glow relay. In otherwords sets up for a cold start. Then switches off after a few seconds. Never ever fit one without a timer in it.
Click to expand...

Ahh Wammers, read my post again:p

I said the resistance of a 'piece of wire' in the temp sensor, never mentioned the MAF. I didn't want to say thermistor as not everyone knows what that is. The engine temperature sensor is a thermistor. The resistance of this changes in relation to a change in temperature. The EDC gets the temp by measuring the resistance of the sensor

The fixes I've seen don't override the engine temp sensor they just add another resistor to the circuit so as to measure the same resistance as a cold engine.
 
SpudH said:
Ahh Wammers, read my post again:p

I said the resistance of a 'piece of wire' in the temp sensor, never mentioned the MAF. I didn't want to say thermistor as not everyone knows what that is. The engine temperature sensor is a thermistor. The resistance of this changes in relation to a change in temperature. The EDC gets the temp by measuring the resistance of the sensor

The fixes I've seen don't override the engine temp sensor they just add another resistor to the circuit so as to measure the same resistance as a cold engine.
Click to expand...

everything is always pointing at the fip,this is not always the case,however in high mileage vehicles 200,000 miles + i would expect fip wear,but around 100,000 miles that is not really acceptable,however any diesel specialist worth his salt should be able to time the pump up or adjust to take up wear,shouldn't always be the case of going for a new fip, however my p38 was one of the first diesels to go "fly by wire" with no cable or moving parts on pump,it was mainly done electronically,so maybe the electronics are failing and not so much the pump,but because of the way these are made it would take more time to fix? well maybe but it makes it so bloody expensive,bosch won't comment (no surprise there),just refer you to landrover same as luk do,so you get caught in a loop,of everyone blaming each other. or it could be low sulpher diesel.
somebody has the answer,unfortunately it is not me.
 
SpudH said:
Ahh Wammers, read my post again:p

I said the resistance of a 'piece of wire' in the temp sensor, never mentioned the MAF. I didn't want to say thermistor as not everyone knows what that is. The engine temperature sensor is a thermistor. The resistance of this changes in relation to a change in temperature. The EDC gets the temp by measuring the resistance of the sensor

The fixes I've seen don't override the engine temp sensor they just add another resistor to the circuit so as to measure the same resistance as a cold engine.
Click to expand...

Ok Spud hand up. :):):)
 
You must log in or register to reply here.

Similar threads

steveshone
p38 fuse box following conversion from petrol 4.6 to diesel 2.5
Replies
7
Views
1K
mystery
mystery
O
L322 4.2 SC AJV8 LPG - Engine Timing kit
Replies
0
Views
559
onebounce
O
tarphenry
p38 diesel
Replies
15
Views
592
tarphenry
tarphenry
T
P38 4.0 Infamous 4HP22
2
Replies
32
Views
2K
Rubberknees
Rubberknees
G
2.5 NA won’t start after injection pump swap.
Replies
2
Views
176
Rougharse Racing
Rougharse Racing
Back
Top