Vacuum sensor values?:
Johan G 2001-02-25:
I was ready for the MOT-inspection. Mr Murphy awoke and that morning it didn't start. Coughed a
couple of times but no start. I have checked all the wirings, I have measured every sensor involved with both
the AEI and EFI, I have checked the ignition (with a strobe light as well as the actual spark check), I have
sprayed it with emergency start gas and found that the coughing can be increased (i.e. it has a spark), I have
pressed the anchor in the injection relay, I have replaced that relay just in case.
When I crank the engine the injection relay cuts in and activates the pump (as it should). After a 30 s crank
the smell and dripping of gas is noticed. From the exhaust, just beneath
the manifold, it is dripping with gas. Completely soaked! OK, out with the spark plugs, wet! Dry out, try
again, same result. Disconnect the vacuum hose to the pressure control valve on the injection valves. Of course
nothing comes through. No dripping. But what if the membrane in the valve has cracked? The car was parked with
a broken main LHM-feed for almost a year. Through -25 to +30 degrees. And then started up. It is possible that
this was too much for the membrane. Does anyone have the correct values (ohm) measured from the vacuum
sensor on a CX25 GTi? It turned out the whole sensor seems to be missing on my spare-part-car. Strange since
I haven't had any problem driving the car. It really ran like a clock... Does anyone have a clue on how many
different versions of the ignition/fuel injection-system there are on the CX25?
This is certainly something that Haynes has left out. The supplementary chapter 13 does not describe the latest
CX good enough. As an example, one difference is the missing slug sensor (actually mentioned in Haynes).
Chris S (AUSS):
For a Naturally aspirated CX2500, 240 mBar depression gives 13 to 17 deg advance. No resistance readings in the
Vacuum modules sensors:
Ducellier 527 005 A (1984 GB. & (?) others)
Ducellier 527 009 A (Swedish/Auss/Swiss(1984 to 1987) & ?US anti polution)
Ducellier 527 005 A (Germany, Austria, Holland, Swiss to end of production)
AEI modules, 2 sensor units:
Non anti air pollution: LA8-LD4 (20-165-646) 1984/85
Non anti air pollution: LA8-LD4 (20-165-970) 1985/87
AP models: LA5-ED001
AEI modules, Single sensor units:
Non anti air pollution: LA8-LD4 (85-102) 1987 onwards
AP models: EC004-ED005 (85 099) all AP's the same.
I suspect US-Canadian cars are to Australian-Swiss-German-Swedish-Holland specs, but we only got the Series 1 car here. I changed my 2 sensor ignition module & manifold pressure sensor from the Swiss AP to UK (GB) specs, to give a significant increase in power & a major increase in smoothness. It still beats local pollution specs required for a vehicle of that build time.
Marc B (NL):
Try the pressure on the fuel rail, it should be around 2.5 bar (see Haynes). If that is correct, the membrane
is OK. Otherwise the pressure is lost as a large amount of fuel is pressed through the vacuum line into
the intake manifold.
Magnus Å 2004-04-05:
Vacuumsensorn är en vit plastcylinder, ca 10 cm lång och 5 cm tjock, som sitter antingen vid höger fjädringsklocka fram eller längre bak vid expansionskärlet för kylarvätska. Det går en tunn slang till den från insuget, alldeles vid trottelventilen. Sensorn är viktig för tändningsförställningen -- särskilt avgörande vid hårt pådrag.
Plocka ut sensorn. Sensorn har en trepolig elkontakt, mät resistansen mellan pinne 1 och 3. Om jag minns rätt ska det vara runt 100-200 Ohm utan vacuum för att gå upp till 2 kilo ohm eller högre när man suger hårt i den. Nån på internationella CX-listan sa att den tenderar att "fastna" vid runt 500 Ohm, och då ska den bytas
Drar mycket bensin:
Som många andra CX gillar min bensin, och mycket sådan.
Om du har 2,5 l-motor (och alltså AEI-tändsystem) så är det vanligt att tändsystemets tryckgivare kärvar ihop eller går sönder. Det resulterar i att motorn går med för sen tändning vid dellast och hög bränsleförbrukning som följd. Givaren har tre anslutningar: Jord, 4,75 V matning från tändstyrdonet, utsignal. Utsignalen ska vara i stort sett 0 V vid tomgång och fullgas, vid måttligt gaspådrag ska den stiga till 1,5 - 2 V. Man kan för övrigt ändra tryckgivarens inkoppling (matningsspänningen får gå via kontakten i ett relä som styrs från tomgångskontakten vid gasspjället, vakuumet kopplas direkt till insugningsröret i stället för till vakuumuttaget vid gasspjället). Fast det är egentligen överkurs. Resultatet blir att motorn känns "piggare" vid lätt gaspådrag och man kan åtminstone inbilla sig att bränsleförbrukningen sjunker en aning.
John H (UK) 2002-07-30:
Ever in search of that elusive, very slight, but bloody irksome, 'hesitation' that my car suffers from. I thought I would check the vacuum sensor again. Multitester connected to middle and one outer pin, and sucking on vacuum pipe, I found 'open circuit' (o/l) upon occasional sucks... AHA! I have a spare second hand one which seemed to give much more reliable readings. Ran car tonight and she is now MUCH smoother through the revs and seems much more willing. I still have an idea the replacement one is not quite right, (whilst sucking I could feel a constant air flow through the diaphragm)? Anyone got a new one spare, or can they be 'fiddled with' (Alan D?)
Regarding repairability, since I was called out, I have no experience, but I would try if you send one to me.
Chris S (Auss):
I had quite a lot of trouble with my car doing this jittery trick a few years back and no Bosch agent here could cure it. Eventually I found a very practical speed shop who found a problem was in the vacuum sensing system. The vacuum module did not leak though. Using a simple strobe, the vacuum advance was found to be overshooting as soon as the throttle opened. They cured the fault using a very small diameter hosing from the inlet manifold to the sensor, simple elegant & effective!
Hi Chris, I don't suppose there is some originally fitted restrictor that has 'got lost' over the years?
Chris S (Auss):
No sign of one in the parts listings, more than likely someone changed the hose to a larger internal diameter one.
John H 2002-08-01:
So I got this sensor apart using a haidryer to soften the plastic and a small screwdriver (easy job). I found the spindle which actuates two sliding contacts mounted on a small plastic block along a tiny PCB had come detached from where it should screw into the small boss on the diaphragm. This tiny boss is aluminium and the threads had stripped. A blob of Araldite should repair it well enough. This detachment of the actuator spindle from the diaphragm will 'play havoc' with advance and retard wont it?
Chris S (AUSS):
Interesting, your vacuum advance capsule is plastic? My original one and its replacement were pressed steel, which would be a bit difficult to dissemble & reassemble.
John H 2002-10-29:
The vacuum sensor assists in advance/retard, yes? So if you have a car which when you prod the throttle it revs up but then takes a while for the revs to die down again (ie. a slow deceleration) - am I right in suspecting this unit in being slow in it's response? Or am I 'way off'?
I found my original sensor's actuating spindle had separated from the diaphragm. This was not a break-down situation but just an evenings "entertainment", sat at the kitchen table, sussing this gadget out, and fitted a spare one 'cos it was oviously bust.
What's made me curious is that after having finally Araldite-repairing the original sensor and trying it out tonight (merely out of curiosity), I found the acceleration / deceleration does seem more responsive. So now I feel the replacement unit I used maybe similarly duff. Might this be a common but generally unnoticed CX fault? Is this vac sensor a serious component regarding fuel consumption, apart from it's normal duties?
As far as I know, the usual use for vacuum widgets on the distributor is to advance the timing under conditions of light load to improve the fuel economy. Their usual failure mode is holes in the diaphragm causing the thing to stop working altogether and creating a vacuum leak. (In some Ford cars the automatic transmission got its engine throttle position information from such a vacuum widget, and when it holed its diaphragm the engine would burn the transmission fluid at an alarming rate.)
To test your vacuum unit, apply vacuum by connecting a long length of transparent tube to the hole. Fill the tube with water and apply vacuum by dropping the end of the tube till the two meniscuse (menisci??) are the required vertical distance apart. If the levels will not separate you have a leak, if they do separate OK you can see or measure the motion of the mechanism.
In my mind this would require further explanation.
Q: What would be the required distance apart.......not what is the distance but how would you know it was the "right"distance when you got there?
Q: How does "dropping" the end create a vacuum?
Q: What do you mean by "the levels separating"?
Q: If there is a leak will this put water in the distributor?
Just to show you how dump this is: I don't even have a distributor in my CX!
I should have said that the tube needs to be a length of flexible tube at least a metre long, in a "U" shape. Half fill the tube with water. Attach one end of the tube to the vacuum sensor, leave the other end open to the air. As the free end of the tube is raised and lowered the levels of the water in the two tubes will not be the same. For instance, if the level of the water near the open end of the tube is 10 inches lower than the level in the sensor end there will be 10inches of vacuum applied to the sensor. If the levels stay the same when the free end is raised and lowered, there is a leak in the diaphragm of the sensor (or in your tubing connections).