Non-turbo car, Regina.

Have had an intermittent problem with a CEL and a 1-4-3 code. Sensor has been replaced and is properly torqued, electrical connection has been cleaned, wire is routed away from alternator and has a ferrite core mounted on it to help mitigate RF interference. Wire routing and the ferrite helped for a long time.

Today, this got more severe. It comes back promptly after being cleared. There are NO other codes.

I am starting to suspect the fuel supply may be compromised, causing it to run lean and leading to detonation under certain conditions. At 120K, it's on the early side for the pump to fail, but the fuel filter is old, probably original. The car doesn't stall, doesn't hesitate, but does not have much power and acceleration is lousy (even before CEL comes on).

Thoughts?

John

I appear to have cured my CEL problem with 1-4-3 knock sensor
by simply giving the large connector behind the drivers side strut tower a thorough cleaning with de-ox-it. This conenctor is where the harness that runs into the cabin mates with the harness that runs under the hood. It was not visually dirty or corroded inside, but separating the halves, polishing the male ends and cleaning both sides of the connection with De-Ox-It does seem to have cured the code problem. I lubed up the outside mating surfaces of the plastic connectors with dielectric grease, since the things were a royal pain to get separated.

FYI, there are two large connectors there. I cleaned BOTH, since I figured they'd both probably benefit. But if you want to deal with only one, you can easily tell which one carries the knock sensor wire since that wire is green is fatter than most of the other wires.

However, the car still does not run with as much pep as it should. It feels anemic compared to my 240. So what I thought was the computer reacting to a bad signal from the knock sensor and causing poor performance thru retarded timing may well be something different. :-(

Tom--thanks for reminding me to post this answer in this thread, too!

John

Okay, this morning I replaced the filter. Everything's back together, no sign of any leaks. It was not a bad job at all.

A few differences I ran into on my 1993 945 vs the FAQ. Whether this is a Regina vs Bosch issue or a 1980's vs 1990's issue, I don't know.

-The banjo bolt on the front is 17MM, different from either end of the Bosch, and there is no banjo bolt on the rear.
-The cradle on my car was held up with 13mm nuts, not 10mm.
-The bolt for the clamp holding the filter to the cradle was 10mm.
-you only need one of the larger pair of washers for the rear fitting. The washer goes between the brass fitting and the filter, then the coupling attached to the fuel line goes onto the fitting plain. It has some sort of rubber nipple down inside it to give the seal.

The installation procedure I used:
-install brass fitting into rear of filter, using 27MM wrench as counterhold on the filter.
-under the car, assemble the banjo bolt/fuel line/2 washer setup, then carefully thread it into the front of the filter.
-Be sure the fitting is aligned so that it won't hit or rub stuff BEFORE tightening down the banjo bolt (use the 22mm wrench to counterhold the filter)
Thread the rear fuel line coupling onto the brass fitting previously installed onto the rear of the filter. Use a 19MM wrench as counterhold on the brass fitting while you tighten down the fuel line connection.
-reinstall the clamp bolt and tighten the filter into the cradle
reinstall the 13mm nut to secure the cradle, and tighten it down.
-clean up tools, put the car on the ground, reconnect any electrical things undone at the beginning, and start the car. I cranked mine for 5 seconds to bring the pressure back up, then it started on the second try. Check for leaks.

Not a bad use of my morning. Will be curious to see if this makes any difference with my knock sensor issue (it rained yesterday evening, so I did not get to play with the multimeter per Tom's suggestion).

john

John, here is the disclaimer: I am not an authority on Volvo electronics and hate to diagnose stuff I have not had problems with in a car other than my own but maybe this will help, without me helping you make it worse.

TP31903/1 1988, Rex/Regina. This was modified to include 1991 updates but I do not think there were any changes in the two wire knock sensor circuit.

"Conditions for fault code: With a hot engine and engine speeds above 2000 RPM a test is carried out of the signal from the knock sensor. If the signal is below a certain level, ***the ignition is retarded***(car runs like shit) and fault code 1 4 3 is set.

Since the signal from the knock sensor cannot be measured without special equipment, only the wires are measured.

Caused of Fault: Break or short circuit to ground in the wires of knock sensor. (or) Connection resistance in contacts." (If you trust OBD1.)

That being said, all it says is the signal is faulty or absent during it's internal testing at stated above. There is no indication that it is caused by a bad pump, dirty air filter or bad plugs, ah, but again this is OBD1.

Two wires, black is grounded inside the Rex ign control module and the green is the signal wire, screened (shielded by the ground?). My interpretations of the test that can be done with an ohm meter, unpluged at the Rex control module and the knock sensor the ctrl module connection points to 12 black(ground) and 13 red (signal) should be an open circuit (infinite resistance) meaning no short to ground.

I'd check continuity for each wire from the knock censor connector to the control module connector. These two wires go through a six pin connector located near the firewall on the drivers side. Pull this apart and check for corrosion. Worth a check too, corrosion at the Ign control module connection.

Summery, it should be a shorted wire, a broken wire, bad connection or another bad knock sensor and hopefully not a bad control module. You may have several unrelated problems going on at the same time if you had clear the code and the car runs bad when cold and before 2000RPM.

Before unplugging anything take what ever precautions required by the FAQ, like disconnecting the battery.

Tom

Edit: make sure the two wires are not shorted together too, Check while unplugged on both ends.

Tom,
Interesting, I did not realize the light was in response to a test function rather than out-of-spec readings during operation.

I would assume that given the nature of this wire, it would be easy for mechanical damage to cause serious problems. One stranded conductor runs down the center, surrounded by a layer of insulation, and the ground conductor is wrapped, in a _very_ thin sheath, around that. An intermittent connection is certainly possible.

I will check things with the multimeter today and report.

thanks,
John

Correction in my post. Where I said red wire I meant to say green.

Contrary to the Volvo publication, the FAQ says other issues will cause the code but that may be the differences in Rex/Regina and LH2.4 or the publication shortcuts potential other cause for the code. I do not trust the publications to the letter. It is is possible but hard to believe the code is the result of its self testing only.

Per the FAQ:
Cheap gas and heavily carboned engines may cause misfiring to the point that the knock sensor signal is not believed by the ECU. Wiring and the knock sensor itself are by far the most common sources of knock sensor problems. [Les Daniels] If this piezo detector fails, it goes open circuit and the ECU 'thinks' that a massive retarding of ignition called for - hence the drop off in performance. On carb models there can be backfiring as well.


Check the wiring and the connections and as both you and Spook mentioned, chase down anything that would cause the car to run bad.

Tom

Dear John,

May this find You well. Fuel filters are not costly and should be changed at 50,000 mile (80,000 Km) intervals. If the one on your car is original, it doesn't owe you a penny. I'd change it, simply because it is long past time to do so.

Do you recall an instance of the car running completely out of fuel? Running dry stresses the pump - it over-heats - and may account for a premature failure. The in-tank pumps on Rex-Regina-equipped cars seem to last at least 150,000 miles, based on Brickboard posts I've read.

Knock sensors and engine computers only rarely fail on B230 engines.

Hope this helps.

Yours faithfully,

Spook

Spook,

I've had the car 18 months, and never ran it out of gas. But who knows what happened before that. I'm hoping to get a few more miles out of the pump before replacing it preventively--at least wait until cooler weather to replace.

I will renew the filter tomorrow morning--still need to track down a 27MM wrench for one end.

Thanks,
John

Dear John,

May this find You well. You'll thank yourself for having looked at the FAQs (see pull-down menu at extreme upper right corner of this page) before tackling the fuel filter change. There are many useful hints as to how to get this done with the least amount of blood, sweat and tears!!

Best of luck with it.

Yours faithfully,

spook