Some questions about fuel injection

This is the Forum for all your Citroen Technical Questions, Problems or Advice.

Moderator: RichardW

Post Reply
Rob_e (UK)
Posts: 13
Joined: 13 Nov 2006, 22:05
Location: Hertfordshire, UK
My Cars:

Some questions about fuel injection

Post by Rob_e (UK) »

I have got a 1.9 GTi 8v which isn't running too smoothly at the moment and I was looking at the electrical circuit diagram for the the fuel injection and there was a couple of things that didn't make sense to me. The circuit diagram I think is in the Haynes manual but is also on page 4 of this document, which looks like an extract of the Citroen manual.
http://www.rwbsmith.plus.com/citroen2/I ... ection.pdf
Firstly when is the supplimentary air valve used? I thought it was only used when the engine was cold to act like the choke on a non injection engine. But looking at the circuit diagram it looks like it is just on whenever the injection relay is on, which must be the whole time the enigine is running as that relay also supplies the fuel pump?
The other thing, I always assumed that there were four injectors and so each one injected only before/during the induction cycle for it's respective cylinder. But looking at the circuit diagram they all look like they are connected together with only one connection to the ECU, which would imply to me that they all inject simultaniously; is this the case?

The answer to those questions propbably won't make much difference to getting the car running better. But, I just like to understand how things work properly.

Having tryed to do some of the tests in the above document today; it seems like the throttle spindal switch is not working correctly. Both measurements where open ciruit regardless of the position of the throttle. I haven't looked at it too closely yet. But, does anyone know if
it is possible to take the switch apart and fix it?

Thanks,
Rob
1989 White BX 1.9GTi 8v (Now awaiting some repair)
User avatar
AndersDK
Posts: 6060
Joined: 21 Feb 2003, 04:56
Location: Denmark
My Cars:
x 1

Post by AndersDK »

Hi Rob -

Thats exactly the reflections I had years ago when I first saw an injection engine - and the electrics schematic.
The suppl air valve is used for maintaining the idle revs during all conditions.
At cold it will allow for more air bypassing the throttle butterfly flap to let the idle increase. This makes the system run lean, which then compensates by adding fuel, and consequently the idle revs will increase.
The idle revs may be as high as 1500rpm at really cold mornings.
Simply to prevent the unstable cold engine to stall.

During idle, the bypassed air is constantly regulated to maintain a hot idle at some 850rpm. This is important, because the hydraulic system keeps cutting in/out the hydraulic pump, which eats approx 1.5Bhp from the idling engine. It can easily make the engine stall in idle, if not compensated for.

The suppl air valve is rather easy to recognise. Its located separately on the engine somewhere, with 2 fingerthick rubber hoses connected. One hose comes from the filter housing. The other hose runs to the intake somewhere past the butterfly flap. Its got a 2lead electric feed from the ECU.
This valve is most often the cause to unstable idle, as oil fumes & dirt internally makes the valve stick and slow reacting.
The valve can be disassembled and cleaned. Much like a windscreen washer pump. Except its not quite as delicate. All internals are stainless, and must not be lubed, only cleaned.

As for the multiple injectors. Yes they are simply connected in parallel. The injectors are consequently of the so-called Hi-Ohm type, specified for parallel connection. This simply means the ohmic resistance of the injector coil is 16 ohms. As opposed to a monopoint single injector which is designated a Lo-Ohm type having 4 ohms in coil resistance.

Sometimes you can see on other car makes that multiple injectors have their own separate connections to the ECU. But still they are paralelled internally to the ECU. The injectors are then consequently not opened in direct relation to the engines working cycles.
Same is true for a monopoint single injector system.
Instead the injector(s) are opened frequent enough to resemble a flow - or constant mist - of fuel.

More on the subject here :
http://www.tramontana.co.hu/citroen/fuel/injection.php
Last edited by AndersDK on 07 Jul 2007, 22:33, edited 1 time in total.
Anders (DK) - '90 BX16Image
philhoward

Post by philhoward »

It's not the feed for the idle air valve you need to look at, but the return (earth) - either one of two systems:
1) Controlled idle - its a solenoid plunger which will oscillate to allow the right amount of air through to get a pre-determined idle speed (and ECU controlled) or..
2) It's a bimetallic strip type and is basically a hole which closes up as it heats up (or the engine heats it up); 12V from the fuel pump and earthed back to the battery.

The injectors are what's called "Batch fire", rather than sequential and it works fine. Sequential injection does apparently give some economy and smoothness benefits at very low speeds and throttle openings, but only when you're chasing the last couple of bhp and mpg..only really used in the last 15 years or so with the odd exception.
jeremy
Posts: 3959
Joined: 20 Oct 2002, 16:00
Location: Hampshire, UK
My Cars:
x 2

Post by jeremy »

At least the Jetronic / Motronic injectors pulse! - Its predecessor - the all mechanical 'K' jetronic simply sprayed all the time.

Some earlier systems used a pump more like a diesel injection pump to deliver the fuel. I think its difficult to draw comparisons as they measured even fewer parameters.
jeremy
User avatar
CitroJim
A very naughty boy
Posts: 49534
Joined: 30 Apr 2005, 23:33
Location: Paggers
My Cars: Bluebell the AX, Polly the C3 Picasso, Pix the Nissan Pixo, Propel the duathlon bike, TCR Pro the road bike and Fuji the TT bike...
x 6163
Contact:

Post by CitroJim »

Rough running on an 8V GTi with Bosch L-Jetrionic. Tell me about it :twisted: :twisted: I've been all over the L-Jet injection system very closely to get my 205 running nicely!

The valve you're looking at is the SAD (Supplementary Air Device) and is only in play when the engine is cold to allow some more air in to up the idle speed a bit. The electrical connection heats a bi-metal strip to begin closing it. The fianl closure is done via heat from the coolant, hence why it is mounted on the side of the block on a large lump of alloy! The electrical supply is maintained all the time the injection is active.

Test it by pinching off one of its air pipes when the engine is cold and the idel speed should fall dramatically. When the engine is hot the idle speed should not change, indicating it is fully closed.

Normal idle is simply set via the screw on the throttle body and is not regulated electronically.

Rob, tell more of the rough running symptoms you're experiencing and I'll give some suggestions. The four most popular causes are the inlet tract, TB and breather system being full of muck (these are very mucky engines :twisted: ), the AFM being worn/out of calibration/dirty, air leaks on the inlet tract anywhere south of the AFM and a duff coolant temperature sensor (CTS) causing the engine to run rich all the time.
Jim

Runner, cyclist, time triallist, duathlete, Citroen AX fan and the CCC Citroenian 'From A to Z' Columnist...
philhoward

Post by philhoward »

Ignore my suggestion no.1...just had another look at your sig and it's an 89 car so should be L-Jetronic (I had a GTi 8v, but a 91 which had Motronic...).

IN which case, i'm with Jim - personal favourites are the AFM can wear out its potentiometer track, or possibly even a small split in the "bellows" from the AFM to the throttle body can cause bad running. Off the top of my head, i can't remember if that version on L-Jetronic had the cold start injector or not (they are known to weep a bit and throw the mixture out).

A good (but not totally guaranteed) check for an air leak is to spray WD40 (or similar) around all parts of the intake system whilst its running - if the note of the engine changes, then there's usually an air leak somewhere.
User avatar
CitroJim
A very naughty boy
Posts: 49534
Joined: 30 Apr 2005, 23:33
Location: Paggers
My Cars: Bluebell the AX, Polly the C3 Picasso, Pix the Nissan Pixo, Propel the duathlon bike, TCR Pro the road bike and Fuji the TT bike...
x 6163
Contact:

Post by CitroJim »

philhoward wrote: Off the top of my head, i can't remember if that version on L-Jetronic had the cold start injector or not (they are known to weep a bit and throw the mixture out).
Thankfully not this version Phil! (or at least I've never met a GTi/L-Jet engine with one)

Another very rough method of checking for air leaks is to take off the bellows between the air filter and AFM and then put a board across the AFM air intake. The engine should stall. If it carries on running, you have a leak. Some, including me :roll: , use the palm of their hand for this but I'm not recommending it, these engines suck :lol: and they suck hard enough to hurt your hand if there are no leaks.

I once (and once only) did it on the intake of a diesel engine. Ouch!

A favourite air leak spot is the bolt that holds the oil filler/breater pot to the side of the inlet manifold. Many accidentially leave it out which results in a nice 6mm hole straight into the inlet manifold.
Jim

Runner, cyclist, time triallist, duathlete, Citroen AX fan and the CCC Citroenian 'From A to Z' Columnist...
philhoward

Post by philhoward »

citrojim wrote:
philhoward wrote: Off the top of my head, i can't remember if that version on L-Jetronic had the cold start injector or not (they are known to weep a bit and throw the mixture out).
Thankfully not this version Phil! (or at least I've never met a GTi/L-Jet engine with one).
I know some L-Jet versions did - CX GTi Turbo for one, but not sure whether or not the BX ones did - hence mentioning it!
lolingram
RIP 2010
Posts: 550
Joined: 27 Dec 2006, 07:59
Location: France
My Cars:
x 1
Contact:

Post by lolingram »

Thanks Phil for the confirmation of what I had deduced many years back, was the case with parallel injectors...
the injectors are what's called "Batch fire", rather than sequential and it works fine. Sequential injection does apparently give some economy and smoothness benefits at very low speeds and throttle openings, but only when you're chasing the last couple of bhp and mpg..only really used in the last 15 years or so with the odd exception.
R.I.P. January 2010.
XM 2.1 auto VSX 1996 - Bosch Inj, Xantia HDi 90 estate 1999, Xantia 1.9TD 1997
Previously...
GS 1970, Dyane 1974, Xantia 94 VSX TD, XM 94, 2.1 auto - Lucas Inj, XM 92 2.1 estate - Lucas Inj
Rob_e (UK)
Posts: 13
Joined: 13 Nov 2006, 22:05
Location: Hertfordshire, UK
My Cars:

Post by Rob_e (UK) »

Thanks, it makes sense now that the ECU isn't connected to the suppl air valve if it is just temperature that closes it. It seems I always assumed that the fuel injection was more complicated than it actually was, I thought it was controlling eaach injector separately. Oh well.

That actual problem seems to have been going on for a while but has also changed, more recently. To complicate things further a while back I was too lazy to disconnect the battery and killed the ECU when it touched the battery. So the ECU and AFM have been replaced with a second hand unit.

A while back it started holding a high tickover, 1500rpm, when warmed up, but not always, it seemed to be more after run at speed, e.g. if you came to the end of a dual carrigage way and stopped at a roundabout the revs would stay that bit higher. I thought it could be the flap in teh AFM sticking bit with the replacement ECU/AFM it did the same thing so I thought that probably wasn't the problem.
More recently when first started it ticks over correctly for a few seconds then the revs seem to drop to about 600, 500 or nearly stalling when the pump kicks in. If after starting it you open the throttle then as soon as you close it the revs drop to this lower level. When driving slowly or with low throttle it feels a bit juddery and isn't that smooth. It is better when warm and the tick over now seems to be about normal when warmed up. So I am thinking that perhaps there is still one problem raising the tick over when warm that is conteracting whatever is lowering it.
the other week I replaced the plugs, ht leads and fuel filter, distributer cap and rotor were changed not that long ago, and it has had a new ignition coil. The only thing that hasn't been checked on the ignition is the timing, doesn't look that easy to see the timing mark on this engine. I did notice yesterday that disconnecting the vacum advance when warmed up did not make any difference to the running, should I expect it to?

I checked the resistance of the coolant temp sensor and that seems to be okay according to the values in the cirroen manual.

I thought the throttle position switch was not working yesterday . But checking it again today it seems okay.

I have tryed pinching one of the pipe to the the supplimentary air valve when the engine was cold and it did slow it down further to the point where it nearly stalled.


Thanks for the help,
Rob
1989 White BX 1.9GTi 8v (Now awaiting some repair)
User avatar
CitroJim
A very naughty boy
Posts: 49534
Joined: 30 Apr 2005, 23:33
Location: Paggers
My Cars: Bluebell the AX, Polly the C3 Picasso, Pix the Nissan Pixo, Propel the duathlon bike, TCR Pro the road bike and Fuji the TT bike...
x 6163
Contact:

Post by CitroJim »

Hi Rob,

You have all the classic symptoms that come with age on these systems. Fortunately, they're not hard to overcome.

You ned to work through everything methodically in this rough order.

1. Dismantle and give the inlet tract a really good clean, especially the throttle body. Take the idle screw out and clean the passage and polish the end of the screw.

2. Dismantle and clean all breather hoses and the oil pot. These engines are terrible for collecting mayonnaise (emulsified oil). Clean out the hoses to the SAD and give the SAD itself a good clean out. The SAD and hoses get full of mayo as well. Check all hoses and bellows for integrity. Check the oil pot lid has its "O" rings in place. Use white spirit or carb cleaner for cleaning the hoses and throttle body.

3. Check the Throttle Position Switch (TPS) is working correctly. As you rotate the throttle butterfly from open to idle, you should hear a distinct click from it just before the butterfly closes completely. Check this click results in the switch closing with a multimeter. If necessary, adjust it by loosening off its mount screws and rotating it.

4. Give the AFM a good clean and make sure the idle passage is clean. Set the idle mixture screw about half-way. Don't touch the potentiometer part of it just yet. Use carb cleaner only on the AFM.

5. Reassemble and check for air leaks, paying particular attention to the breater hoses and oil pot, Any unmetered air that gets in can upset these systems. Fix any leaks before going further.

6. Check the vacuum advance is working. The hose from the throttle body to the Vacuum Capusle on the distributor often perishes and lets in air. The Vacuum Advance capsule itself often fails due to a punctured diaphragm. Try sucking down the vacuum pipe. If you can suck anything up it (usually a mouthful of something that tastes dire) it is duff. The advance plate in the distributor rarely gives problems as it gets enough oil past the sealing "O" ring in the head to keep it well lubricated. It sounds like you have a problem in this area.

The AFM can be reasonably recalibrated by following a procedure I have used quite succcessfully. The timing can be reasonably set at the same time. I'd replace your CTS as a matter of course before going further unless you know it is reasonably recent.

Start up and check the SAD is working and allow the engine to warm up fully. Keep the idle reasonably fast at about 1100rpm.

Slowly rotate the distributor for fastest and smoothest idle, reducing idle speed if necessary. Open the potentiometer lid of the AFM and mark where the cogged wheel which controls spring tension is presently set (so you can revert to it if need be). Now rotate the cogged wheel for fastest and smoothest idle, adjusting the timing a little at the same time. Keep adjusting the AFM and timing until no further improvement can be gained. Set the idle speed to 1000 rpm. You may need to rev the engine a few times doing this to prevent it bogging.

Now switch on all electrical loads (main beams, fogs, heated rear window etc.) and check the idle speed only drops a little. Fine-tune the timing and AFM for the smallest change in idle between electrical load on and off. With all loads on, finally set the idle to 1000rpm. When off-load the idle will now be around 1100 rpm but that work s well to avoid bog-downs when the engine is half-hot and the SAD is closed.

This is doing the job by "feel" and gets it approximately right but if you regularly use the car hard, the settings should be checked on a rolling road to ensure you're not running lean at full throttle/rpm. If you have a CO meter, set the idle mixture to around 1.5 - 2% CO using the AFM idle mixture screw and make a check that at 2500rpm the CO is very, very low.
Jim

Runner, cyclist, time triallist, duathlete, Citroen AX fan and the CCC Citroenian 'From A to Z' Columnist...
Rob_e (UK)
Posts: 13
Joined: 13 Nov 2006, 22:05
Location: Hertfordshire, UK
My Cars:

Post by Rob_e (UK) »

Thanks for the reply.

I made a start on some of the list. I have cleaned the inlet parts including the idle speed screw. It was a bit oily so the hole may have been partially blocked.

I'll have to have a look at the pipe on the sup. air vavle another day. Although that does seem to be working, it slows down the engine if the pipe is pinched when cold and does nothing when hot.

I think I have been quite lucky given what I have read on here about people having problem with Mayonase. I so far have encountered this on mine. Unless it is hiding in one of the pipes somewhere. :)

The oil filler cap is new and still clean at the moment, the old one was pretty clogged up with oil.

At the moment I am thinking the vaccum advance might be the problem. There is a bit of an oil leak from that end of the engine, jsut enough that from time to time I have to wipe a coating of oil off the parts in that area and it does tend to coat the outside of the vaccum module. I am thinking perhaps the oil has leaked into it and made it leak. You mention the "O" ring seal, I guess it is that which is letting a bit too much oil through on mine. Is that the only seal at that end of the cam shaft? The haynes manual describes changing a cam shaft seal at the pully end of the engine but not at the dizzy end.

I have bought a timing light on ebay, so should be able to see for sure if having the vaccum tube connected makes any difference.

Thanks for the help,
Rob
1989 White BX 1.9GTi 8v (Now awaiting some repair)
User avatar
AndersDK
Posts: 6060
Joined: 21 Feb 2003, 04:56
Location: Denmark
My Cars:
x 1

Post by AndersDK »

There is no cam seal on the dizzy end of a petrol.
Only the O-ring which seals against the bottom of the dizzy body.

This is because the dizzy drive axle is located higher than the O-ring, and you only have the oil fumes pressure from the cam axle cavity. Hence you should have no risc of oil leaks out through the dizzy body.
If you happen to remove the dizzy body some day, you will notice a drilled hole to let oil flow back from the dizzy axle in to the cam cavity.
This is to prevent the O-ring seal would fail from any small amount of oil collecting over it.

The mayonnaise will come, if you switch to short trips during wintertime, or the thermostat stops closing completely.
Anders (DK) - '90 BX16Image
User avatar
CitroJim
A very naughty boy
Posts: 49534
Joined: 30 Apr 2005, 23:33
Location: Paggers
My Cars: Bluebell the AX, Polly the C3 Picasso, Pix the Nissan Pixo, Propel the duathlon bike, TCR Pro the road bike and Fuji the TT bike...
x 6163
Contact:

Post by CitroJim »

It is a good idea to change this O ring as it is notorious for causing leaks and over time, filling the distributor with oil. This oil does not seem to cause many issue in the distributor though, it is just messy :)

I'd replace the vacuum capsule and pipe as matter of course Rob. They're not expensive but do make sure you get the right one.
Jim

Runner, cyclist, time triallist, duathlete, Citroen AX fan and the CCC Citroenian 'From A to Z' Columnist...
Post Reply