Blank Cheque

[cachaciero]

hi Jim

only thing that concerns me it must pump all the Eolys fluid into the tank which carnt be good .
so is there any cures out there .apart from pulling the fuse every time it starts running
cheers
dexy

Not necessarily, to put fluid into the tank the pump has to run AND the inlet valve /injector (to the fuel tank has to be open).

I am not sure if what I am about to say is correct but it is how I suspect that the system works.

The most accurate way of metering precisely what can be relatively small quantities of liquid is to charge the liquid to a defined pressure and then inject it through a fixed metering orfice the amount of fluid being injected being proportional to the time that the orfice is open. So in summary I believe the eolys injection system works not unlike the engine injection system.
If this system is engineered properly it follows from this that the pump will always run to maintain a fixed pressure in the Eolys supply line to the tank injector but why the pump runs when switching of the engine I can't expalin unless that is normal operation. The only time that the pump would need to run is after fuelling the car as detected by the filler cap being opened.

I believe that the logic would go something like this:-
Engine off and filler cap open check fuel quantity, run pump to pressure system, cap shut get new fuel quantity, go calculate fuel added, go calculate injector open time, open injector for required time, run pump to maintain pressure, close injector stop pump.
There is I believe a failure case for when the cap switch is not working, the system will inject an amount of eolys corresponding to a max fuel load during every engine start / run cycle but I don't know if that happens when the engine is started or when it is stopped, if the latter then that maybe is what you are hearing and there is a problem with the cap switch.

Please don't take any of the above as Gospel it is a little bit of knowledge and a lot of conjecture.

cachaciero

[myglaren]

I believe that there is a magnetically operated reed switch in the filler flap.
If the switch is defective - and I have heard a few people mention problems with them, then perhaps the pump is reacting to an assumed refill of diesel and preparing to inject, once the system identifies the volume of diesel added.
As no has been it may well not inject any.
Disconnecting the reed switch may verify this theory.

[cachaciero]

I believe that there is a magnetically operated reed switch in the filler flap.
If the switch is defective - and I have heard a few people mention problems with them, then perhaps the pump is reacting to an assumed refill of diesel and preparing to inject, once the system identifies the volume of diesel added.
As no has been it may well not inject any.
Disconnecting the reed switch may verify this theory.

Yes that's kind of where I had got to in my last post.

I wonder how the system decides that it has a faulty switch? switch state = open but car moving ?or more possibly switch state= open but filler flap lock indicates closed.

If you have a Lexia and Jim (leshed) has you can check the filler cap switch, also if it is faulty the additive ECU should post a fault.

cachaciero

[jimleshed]

My guess for the system deciding if the switch was faulty (could be open or closed state) would be any increase in fuel level without a change of switch state.
I seem to recall a faulty filler switch on mine when I cleared all faults. Cant remember if it did clear or not though.
But mine does not do the pump cycling, a battery removal seemed to cure that.
JLS

[cachaciero]

o.k To answer my own question and a theory.

To initiate an additive function with the engine off the system needs to see a fuel cap cycle. This is that the filer cap switch changes state for at least five seconds. i.e the time that the cap is removed to the time it is fitted and rotated to locked condition has to be more than five seconds. Now as regards the switch it has been stated that this is a reed switch however I think we should consider that it may be a hall effect sensor it is supplied with 5 Volts. and very specific on and off resistances are quoted for the device i.e on resistance is 15 ohms between the signal line and five volts if it was a reed switch the on resistance would be pretty close to zero and the off resistance is 150K again a reed switch would be close to infinity unless it has a 150K resistor across the contacts. The 150K provides a mechanism for the ECU to monitor the presence of the switch when the switch is open.

The interesting thing is that if the filler cap is removed and refitted without any addition of fuel then the system will inject additive equivalent to 7 litres of fuel which is the minimum resolution of the fuel gauging system.

Also the pump runs for five seconds every time the ignition is switched on this equates to 0.1 ltrs of fluid, now I reckon that is close to the dose of additive required for a full tank

Speculation:- If I was designing this system given the pump has to get up to 3bar pressure and has a leak off back to tank then I would for any quantity of additive run the pump for a fixed time, the actual amount of additive injected is controlled by the injector in the fuel tank and five seconds seems to be about the maximum time required for the pump to pump a tank full of additive.

Ergo I reckon if the pump runs it will always run for about five seconds irrespective.

Which still leaves the question why does it run when switched off? well I reckon that if the filler cap switch is good then the most likely reason is the Switched Ignition supply which comes from the BSI bouncing i.e goes off comes on and then goes off or maybe it doesn't. The ECU actually has three supplies including a permanent Battery Supply I suspect that the ignition supply from the BSI is as much an "on" signal as a power supply but I could be wrong. The pump I guess will derive it's power from the permanent Battery supply.

Enough for now!

cachaciero

[jimleshed]

Just checked. no faulty filler on mine, so the fault must have cleared.
Why do they have permanent faults?
I know I had permanent faults on mine (the Eolys ecu had one) so I assumed it could not reprogram.
Then I had the pump cycling.
Then I removed the battery. Pump cycling stopped.
Just rechecked all faults on Lex. NO faults in Fluid system. so the permanent ones have gone......???
JLS

[myglaren]

The reason I theorised a reed switch was that many of the complainants have said that the magnet had either fallen off or become ineffective.

A Hall effect sensor would be a bit of overkill I reckon - doesn't mean there isn't one, of course.

I was also under the impression that no Eolys dosing would occur unless >10L were added to the tank

[admiral51]

I have been reading and re reading this thread and have had trouble understanding whats been going on
Now i dont have anything that needs Eolys fluid to be added so cant add anything from personal experience but here goes anyway

As i understand it the additive is added to the main fuel tank on a given ratio to fuel added.This process is all controlled by ECUs.
One ECU will detect that fuel has been added (filler cap removed for x secs etc and intank guage).Another ECU will tell additive pump to start and run for a set time to enable additive to be pumped to main fuel tank.Another ECU will measure quantity of additive actually added.
Assuming that there is a type of one way valve only allowing additive into the fuel tank and that this is the point of reference for the ECU to confirm the amount of additive added is it possible that the problem is actually mechanical and not electrical.
My reasoning is logical but it takes a leap of faith .................................................................

If i try and print a document on my printer and there is a problem with the printer it gets put into a "still to do list".Until i "purge" the list it will still try and print the document.Its not a problem with the computer(ECU) its doing its job its just not getting the "Job Done" response.
So my logic is this
The additive pump has been told to run until it gets told the required quantity of additive has been delivered from the additive inlet(fuel tank) ECU.
If this ECU is saying i have not received the right amount of fluid then the pump will still be told to keep running until it is told job done or you disconnect its power supply long enough to "purge" its to do list.
By switching the ignition on/off may confuse the ECU enough to forget its to do list.
If the pump was running for the length of time described i would imagine the additive tank would be emptied very quickly so im guessing the additive is not getting to where it should be unless the delivery point is being opened intermittently,which may allow the additive to be added on a bulk to do list.
My apologies if this is not technically based,its just my random thoughts

Colin

[cachaciero]

I have been reading and re reading this thread and have had trouble understanding whats been going on
Now i dont have anything that needs Eolys fluid to be added so cant add anything from personal experience but here goes anyway

You should be so lucky

As i understand it the additive is added to the main fuel tank on a given ratio to fuel added.This process is all controlled by ECUs.

One ECU does the lot.

One ECU will detect that fuel has been added (filler cap removed for x secs etc and intank guage).Another ECU will tell additive pump to start and run for a set time to enable additive to be pumped to main fuel tank.Another ECU will measure quantity of additive actually added.
Assuming that there is a type of one way valve only allowing additive into the fuel tank and that this is the point of reference for the ECU to confirm the amount of additive added is it possible that the problem is actually mechanical and not electrical.
My reasoning is logical but it takes a leap of faith .................................................................

There are only three electro mechanical devices in the system, the fuel cap switch,the pump and the fuel tank injector.

The pump runs, for as long as it is needed I reckon about 5 seconds the injector valve into the fuel tank actually meters the additive into the tank and is open only long enough to admit the required amount in many cases only fractions of seconds and surplus additive returns to the additive tank.

If i try and print a document on my printer and there is a problem with the printer it gets put into a "still to do list".Until i "purge" the list it will still try and print the document.Its not a problem with the computer(ECU) its doing its job its just not getting the "Job Done" response.
So my logic is this
The additive pump has been told to run until it gets told the required quantity of additive has been delivered from the additive inlet(fuel tank) ECU.

No the additive pump just gets told to run for the amount of time required to inject the amount of additive required for a full fuel load there is no feedback back to the ECU of the amount of additive actually injected.

If this ECU is saying i have not received the right amount of fluid then the pump will still be told to keep running until it is told job done or you disconnect its power supply long enough to "purge" its to do list.

There is no feedback of the amount of additive added.

By switching the ignition on/off may confuse the ECU enough to forget its to do list.
If the pump was running for the length of time described i would imagine the additive tank would be emptied very quickly

Not unless the injector valve into the fuel tank was open, the fluid just gets pumped around the system. The injector valve is the actual control element in the system.

so im guessing the additive is not getting to where it should be unless the delivery point is being opened intermittently,which may allow the additive to be added on a bulk to do list.
My apologies if this is not technically based,its just my random thoughts

Colin

Don't apologise, trying to explain things to someone who knows a bit less than i do often forces me down though trains that I wouldn't have otherwise gone, forces re-evaluation and that is no bad thing.

cachaciero

[cachaciero]

The reason I theorised a reed switch was that many of the complainants have said that the magnet had either fallen off or become ineffective.

A Hall effect sensor would be a bit of overkill I reckon - doesn't mean there isn't one, of course.

I was also under the impression that no Eolys dosing would occur unless >10L were added to the tank

I suspect that a Hall effect devices might even be cheaper than a reed switch and far less susceptible to mechanical vibration but i don't really know I just suspect that it is an electronic sensor device because it's supplied with 5 volts not 12.

My understanding is that the minumum amount of fuel level change measurable is equivalent to 7 litres. If the cap has been removed and re-fitted but the system detects no fuel level change it assumes that fuel has been added maybe upto 7 litres and injects the appropriate amount.
I guess that the logic is if fuel does not get treated with the minimum amount of cerine then the carbon formed will never burn off in the FAP and it will choke up sooner rather than later. An excess of cerine is not so bad the cerine will ultimately choke the FAP but the amounts are so small it's not really an issue providing the carbon gets burnt off.

cachaciero

[cachaciero]

Just checked. no faulty filler on mine, so the fault must have cleared.
Why do they have permanent faults?
I know I had permanent faults on mine (the Eolys ecu had one) so I assumed it could not reprogram.
Then I had the pump cycling.
Then I removed the battery. Pump cycling stopped.
Just rechecked all faults on Lex. NO faults in Fluid system. so the permanent ones have gone......???
JLS

Interesting as regards permanent faults I guess it depends on what type of mechanism they use for storing the fault flags, if they use a flash memory type device then one would expect a permanent fault to be there until written to and cleared with a diagnostic tool. But if they are relying on a device which requires power to hold data then removal of battery power would lose everything and this ECU is fed directly from the battery.
I have to say that doesn't make a lot of sense as we know that the additive counter data is held in non volatile memory so why not the fault codes.
Maybe the additive ECU is another device that like the BSI occasionally ends up with it's running software corrupted and benefits from a total power off reset.

cachaciero

[wigelywoo]

A sudden thought wiggley.
I seem to recall a specific routine you can go through after fluid top up. I'm sure someone here will have a link.

Something like (but only from memory)
A Top up tank (more than 10 Litres)
B run engine for 2 mins
C switch off for 30 secs
D open/close filler cap
E restart engine.

NOTE - this is only from memory, I'm sure the correct routine varies by some important little detail or is more complex, so best to find out exactly what it is. Someone here may be able to cut and paste it in!

Now then, fingers crossed, it has stopped doing the malfunction..
I'm looking at the above post from Jim

The day after getting car back, end of April, I filled up the tank A
Run engine a number of times B
Switch off for indefinate times C
Last weekend, open filler cap, fill up, close filler cap D
Use car ie. run engine E

On Tuesday, whilst waiting in the car, switched to Auxilary, my better half told me the pump doesn't run on any more.I've tried and checked since and every time I sitch off I wait but the pump no longer starts up.
I've put letters to Jim's post to identify that I have followed the sequance although I didn't D, open/close the filler cap, until I filled up again

[admiral51]

Think im better off sticking with Xantias, atleast i have some knowledge of how they operate
Having read your post cachaciero,you have explained what i was thinking better than i put in mine

My trouble is putting my thoughts of non mechanical knowledge into mechanical terms

Hope the OP gets to the crux of the problem,ill retire gracefully....................

Colin