Technical Multijet with new injection pump.

[DaveMcT]

After a torrid time trying to diagnose a low fuel pressure fault on my 1.3 70bhp Multijet, I have now got a PF Jones remanufactured pump, new fuel rail pressure control valve and new pressure sensor.

Issue.
Should I now reset the injector trims (which needs licenced MES) or will the ECU learn new trims to suit the new pump?

It’s doing a couple of obvious pulses just above tickover after you blip the revs. I’d expect a little pulse to prevent engine stalling but this is more obvious than anything before. Maybe the weak pump hid the feature but it does seem unusual.

So should I use it and wait or should I reset the injector parameters?

 

[koalar]

Inject trim are only changed when you swap injectors

Normally not that important anyhow

Does it run better with the MAF disconnect

Over 4 weeks and we still don't know what the pressure is at cranking, I figure we know

With meaningless data, all you will get is guess that may or may not be correct

 

[DaveMcT]

I’ve said before I refuse to spend £200 on common rail test gauge (and fittings) to use on one job. The MES values are all we have. Values for the new pump will follow when I’m back in UK.
That all said - the rail low pressure alarms were seemingly random. But became more consistent. Then engine starting got ever more difficult. Finally it started only with the rail pressure sensor unplugged but not when connected. New pump fitted - it started after around 1 second cranking and no more low pressure alarms.
The engine feels more aggressive. As if the accelerator pedal map has been hopped up.

 

[The Panda Nut]

I went through hell back in the early 90's with bad diesel running. I really feel for the problems you are having. There are so many things that can affect this. I have two thoughts on you current predicament.
FIrstly the pump. I know its been reconditioned but it wouldnt be the first time such a thing is still not quite right.
Secondly Bosch / CAV diesel injection centres. If you can get the car to them, although I expect they have vans out and about.

I spent months trying to get my BX running right. In the end I took it to our local centre and they were amazingly helpful. In the end they had the pump apart and also redid the injectors. Its too long ago to remember what issues they found but it solved my problems at a reasonable cost of around a 3 hours labour. Have you been back to PF Jones and asked if they can check that pump again or arrange for someone local to you to do it on an agreed charging basis. PFJ might just come up with some other ideas to check before hand to verify things or just do the pressure checks for a sensible fee?

I think Koalar is right in saying you need pressure readings as they must be revealing. I think my problems started after a cam belt change and were timing mainly but not entirely related. Fortunately that was before elctronic controls were involved.

 

[RalphM]

Have you done a leak off test on the injectors? lots of how to do it's on U-tube
the pumps generate very high pressure but not much volume so a leaking
injector means low fuel pressure, there's always some leak back but look for
any that are leaking more than the rest.
If your getting the pulse as you are driving as in coming to a stop check the
switches on the clutch and brake pedals

 

[DaveMcT]

The engine was rebuilt last year but original injection pump was leaking at a cylinder head. New seal failed to fix it. PF Jones did not have the type in stock so I bought from a supplier on eBay. It was a fair bit cheaper so I should have been more suspicious.
I clutched at straws thinking pump must be ok. But the decline in performance pretty much proved what was going on.
New remanufactured pump (this time PF Jones had one) and the problems are gone.
Injector leak off rates measured on old pump with 1 minute at idle speed are more than ideal (approx 15ml each).
I don’t want to spend £400 on a set of remanufactured injectors but no doubt one day they’ll be needed. The software trims are reasonable so they’ll do for now.

Edit. These are more reasonable but if my power and mpg are ok I’ll stay with what I have. https://www.ebay.co.uk/itm/22480433...-URlgaVRlK&var=&widget_ver=artemis&media=COPY

 

[koalar]

I’ve said before I refuse to spend £200 on common rail test gauge (and fittings) to use on one job. The MES values are all we have. Values for the new pump will follow when I’m back in UK.
That all said - the rail low pressure alarms were seemingly random. But became more consistent. Then engine starting got ever more difficult. Finally it started only with the rail pressure sensor unplugged but not when connected. New pump fitted - it started after around 1 second cranking and no more low pressure alarms.
The engine feels more aggressive. As if the accelerator pedal map has been hopped up.

Who said anything about ording £200

All I asked for is the cranking RPM and cranking pressure

All can be read with a simple scantool

I just disconnect the injectors but your telling me multiECUscan will not do this, strange as WoW, Delphi and examiner all work fine

As a work around it was suggested to disconnect the crank sensor instead, yes it's a little awkward, but I just had my hands on the crank sensors connector on a Fiat Fiorino 1.3 Multijet it wasn't that hard

 

[DaveMcT]

I can get live data using MES but I have not found a way to crank the engine without it starting. This leaves no time for a stable pressure reading to be recorded. Disconnected injectors leave the ECU "offline". I did not try unplugging the crank sensor as access is difficult and I expect the same outcome.

The graphs created during early tests are in the "No Start" thread.

Interestingly, the pump pressure fell off quite quickly but I have found no debris in the HP fuel pipes. As said before, when it started with pressure sensor disconnected but not with it connected I decided to call it a day and replace the pump. New pump fitted and it fired up immediately.

I'm now in Spain for most of August so won't have any new pump graphs until I return. It will be interesting to compare with the old ones.

 

[koalar]

I can get live data using MES but I have not found a way to crank the engine without it starting. This leaves no time for a stable pressure reading to be recorded.

Should be up to pressure in well under a second
Yours is showing at 200 rpm 150 bar, should be around 400 bar, although with multiECUscan you only appear to be getting about 3 samples per second so it could be way off

Disconnected injectors leave the ECU "offline".

I no idea why multiECUscan is struggling

I did not try unplugging the crank sensor as access is difficult and I expect the same outcome.

Why, the starter is purely electrical

It's not that difficult

The graphs created during early tests are in the "No Start" thread.

Link them, I haven't seen any posted except after the car started,

Interestingly, the pump pressure fell off quite quickly

I expect it's normal, if it didn't drop of quickly there would be a higher load on everything at the start of cranking

Some cars are different with check valves, this the pressure is kept in the system by the duty cycle at the valves on power no pressure, we a little by a small spring but not much

but I have found no debris in the HP fuel pipes. As said before, when it started with pressure sensor disconnected but not with it connected I decided to call it a day and replace the pump. New pump fitted and it fired up immediately.

I'm now in Spain for most of August so won't have any new pump graphs until I return. It will be interesting to compare with the old ones.

 

[DaveMcT]

I can crank the engine with injectors unplugged and probably with crank sensor unplugged. But ECU does not switch on so MES can’t see anything. It behaves like the engine is switched off.

 

[koalar]

You need some software that works

Here's fiats full instructions

Vagcom or whatever it calls itself now, Delphi and WoW all work

I very much doubt multiECUscan fails, after all it reads pressure at ignition on.

I working vauxhall is around 400 bar

Without known results we can only keep changing parts and hoping

We haven't even confirmed its s fuel pressure problem

Every time something is touched the symptoms change.

 

[DaveMcT]

This test makes a lot of sense but MES is not up to the job. When injectors were unplugged the software failed to connect - acting like the ignition was switched off. Very annoying and wasted a load of time.

There’s was a clue in the rate of fuel rail pressure rise and I now believe the pressure drop when trying to accelerate was a sign of pump problems.

Maybe the licensed version works on this, but I would want specific statements before buying.

 

[DaveMcT]

The engine starts and runs well with the new pump. Even with air in the system (d/t fitting the pump) it started with 1 second of cranking.

I have also fitted new glow plugs (three were down) and replaced the (very dirty) MAP sensor. Rude not to at £23.

Now the problems - The new pump seems to have exposed the injectors are also past their best. No surprise, the 1 minute leak-off rates were really not good around mid teens ml after 1 minute at idle. There is no way they could meet 25ml per injector over three minutes.

It black smokes on start-up and when revved from idle. Grey smoke when pulling uphill. It hunts under light throttle. It hunts quite badly below 2000 rpm when revved to 3000 and dropped back to idle. Worst of all, there's a significant diesel knock just as it injects fuel to catch the revs just above idle speed. I believe the pre-injection "puffs" are not puffs any more.

I believe there's at least one injector dribbling and the others won't be much better.

Maybe I am catastrophising, but I can see another parts bill incoming. Any ideas? >

 

[RalphM]

The high pressure pump can supply great pressure but little volume dribbling or high leak back and you will have bad starting as
the pump won’t be able to build the pressure needed at cranking speed

You do get diesel knock at low revs and light load but a dribbling injector will also cause it,

 

[A3jeroen]

Bad injectors in a common rail can cause holes in a piston. Don't wait too long.

gr J

 

[koalar]

The car was working up to the broken timing chain

As far as I remember

We have fitted a HP pump seal set
Fitted a new pump
Fitted another new pump

Fitted a new MAP
Fitted a different head
Fitted a new filter housing
Adjusted the turbo actuator
Fitted a new pressure regulator
Fitted the wrong injector washer
Fitted 3 glow plugs

Probably missed things out

It's been not starting
Starting no smoke
Starting black smoke

Every time we touch the car the symptoms are different

Done a leak off test said they are okay
Now there not

I have no idea the leak off test is dependent on the pressure there tested at, and the pulse duration if the leak off is identical on all four, I would find it suspicious for all four to fail at the same time

As of yet we have no meaningfully data

Pump duty cycle
Valve duty cycle
Fuel Pressure
Air flow
Ambient temperature
Inlet pressure engine off
Inlet pressure at idle

And so on

 

[DaveMcT]

Pump duty cycle - not available on MES
Valve duty cycle - valve is the pump cycle but not available on MES
Fuel Pressure. - I can get that but it clearly starts better than with the old pump.
Air flow. - not sure why this matter but MES has it.
Ambient temperature, Inlet pressure engine off - and Inlet pressure at idle. - Not sure why these matter but MES has it.

 

[DaveMcT]

The fuel system has had everything swapped EXCEPT injectors. I discounted HP pump because it was supposed to have been remanufactured. Experience shows it simply had new seals fitted. Fuel lift pump was making 4 to 6 bar IIRC.

Parts replaced -
New filter unit to resolve the fuel temperature error and fix cruddy pipe connections.
It’s had - New pressure sensor, new pressure regulator, new HP pump.
When the new regulator was fitted the engine would only run when unplugged. That triggered buying the new pump.
Ever since the engine rebuild last year, there has always been a lumpiness just above tick over. A minor annoyance. And it smoked but it did pass MOT.
The new pump has exaggerated the lot. I did not notice any hunting but it’s now obvious. The weak pump hid a lot.

MES figures to follow.

 

[DaveMcT]

Engine Speed 885 rpm
Fuel pressure 280 bar
Engine Speed 1818 rpm
Fuel pressure 412 bar
Intake air temperature 29.5 C

These ignore the transients. I’ve given up trying to get a graph from MES. The axes have changed (probably got caught with track pad mouse). I have no idea what the axes should be (Meaningless abbreviations) and there is no visible reset to default. I simply cannot be arsed doing trial and error against its convoluted programming.

And to top it all, the engine is now behaving itself. No black smoke and no significant hunting. Unplugged injectors with engine running created a knock when reconnecting. But afterwards it was running smoother.

This is yet more clouded information that had me chasing my tail last time.

Please note MES is not even close to professional software. So demands for professional level data get us nowhere.

 

[koalar]

It could be anything, we don't even have clear set of symptoms, as they keep altering

Doesn't start
No smoke
Black smoke
Now no smoke again

MAP, MAF, EGR, vacuum leak and so on, it could be anything

On a good good 1.3 multijet as used in Corsa using free software I have shown, Pump duty cycle, Valve duty cycle, Cranking pressure before injectors fire, pressure drop off and ramp up

Comparing this data to your car would tell us whether we were looking in the right area

Without some data, we are shooting blind

Everything in your last post looks correct

We are assuming it's a fuel problem and keep changing things without measuring anything and hoping

If multiECUscan isn't right for this application, find something that is

There just PIDs there's plenty of free stuff out there that work fine with an elm327