NorthWet Posted November 27, 2006 Author Share Posted November 27, 2006 Next step is to put a test light on a fule injector and find out if the injectors are cutting out or the fuel pump is cutting out. nipper With green connection made, I can hear the pump cycling, so my (unconfirmed) belief is that the pump is running. Also as previously stated, I have not verified pressure or flow yet. First thing I plan on doing is replacing filter and checking pressure/flow. Hopefully will have the time and weather to do this tomorrow. Link to comment Share on other sites More sharing options...
NorthWet Posted November 29, 2006 Author Share Posted November 29, 2006 Since I had e code for the knock sensor, I went ahead and replaced that. Old one was cracked, like PO had overtorqued the mounting bolt or something. Put new (used) spare in, started it up and it was still throwing the "22" code for the KS. Oh well, probalby just need to clear/reset the ECU. We have been driving it around for a couple of days, never too far from home (until I took it to work last night). No probs yet, but as they say, " One robin does not a spring make." I'm going to be annoyed with Subaru if a faulty knock sensor keeps the engine from starting. Link to comment Share on other sites More sharing options...
Cougar Posted November 29, 2006 Share Posted November 29, 2006 I think the code light should go out immediately if the sensor circuit is ok. You may have gotten another bad sensor. The crack you describe is a common problem with them I believe. You should be able to trick the circuit by replacing the sensor with about a 530k ohm, 1/2 watt resistor. One end of the resistor goes to the lead and the other goes to ground. If the code doesn't go away then there is something wrong with that circuit and you will need to check that wire lead coming from the ECU. A bad sensor will not cause the engine to have any starting trouble. Link to comment Share on other sites More sharing options...
nipper Posted November 29, 2006 Share Posted November 29, 2006 aha USED is not new. Using a used knock sensor is like using a used o2 sensor, its just not done. The older knock sensors crack and that causes failure (they even had a recall onthem i think). Get a new one. Normally the knock sensor would not cause a no start condition, but it in conjunction with something elese being just on the edge of tolerance may (tolerance stacking) but it is very rare. Now if it is a combination of mistimed spark and a bad crank/cam sensor (im guessing crank as that has to do with fuel injectors) it can be an issue, a real long shot. Have you gotten yourself a Haynes manual yet? nipper Link to comment Share on other sites More sharing options...
NorthWet Posted November 29, 2006 Author Share Posted November 29, 2006 ...Have you gotten yourself a Haynes manual yet? nipper I have an FSM for it. I know that used is not new, but the "new" one didn't have a pie section of the annulus coming loose; physically better than what was there. Regarding the light not going off immediately, when I swapped it out I hadn't considered that it was a one-wire sensor and needed to round through its mount. Car was a salt-belt car, and the old sensor was rusty and in general cruddy looking, and I didn't clean the block mount pad. Maybe contributing to the ECU code. Link to comment Share on other sites More sharing options...
NorthWet Posted December 3, 2006 Author Share Posted December 3, 2006 Well, the other day while I had 15 minutes to spare and a helper (kind of) I connected a "noid", and it indicate injectors were firing nominally normal. Today, I realized that the fuel pump was accessible without crawling in the snow/ice/slush, being accessible from inside rather than under chassis like my OldGen cars. SO,... I checked out the pump and it is not working. Does not seem to have power at the connector, but with TEST connectors together and ignition on I can hear the cooling fans slow down rythmically and a clicking like a relay from near the dash. Temperature was dropping and light was failing, so I will follow up tomorrow. Link to comment Share on other sites More sharing options...
Manarius Posted December 3, 2006 Share Posted December 3, 2006 If you have a sedan, pull up the rear seat and check the connector there. They have a tendency to heat up and melt. Link to comment Share on other sites More sharing options...
ninjaben43 Posted December 3, 2006 Share Posted December 3, 2006 sounds like you nailed it, but fyi with gas as $#itty as it has been lately, i have gone thru 2 fuel filters in the last 2 years, and both times it has acted like my fuel pump has shut down. the crap in the gas literally clogged my f.f. Link to comment Share on other sites More sharing options...
NorthWet Posted December 3, 2006 Author Share Posted December 3, 2006 Manarius, I had already done that before my last post (yes, sedan). Connector looks fine, no significant power at the connector when in TEST mode. sounds like you nailed it, but fyi with gas as $#itty as it has been lately, i have gone thru 2 fuel filters in the last 2 years, and both times it has acted like my fuel pump has shut down. the crap in the gas literally clogged my f.f. Prior to posting, I had replaced the FF (which, BTW, was not clogged and flowing fairly freely) and installed a pressure guage. 'bout time to go outside and do more diagnosing. Link to comment Share on other sites More sharing options...
NorthWet Posted December 5, 2006 Author Share Posted December 5, 2006 Update: Fuel pump is not getting power to its connector, I have bypassed the pump relay with a wire jumper so it should be getting continuous power (yes, 12v at the relay connector). Anybody know if there is another connector between the relay and the pump? And/or where the wiring runs? Link to comment Share on other sites More sharing options...
nipper Posted December 5, 2006 Share Posted December 5, 2006 Only one way to find out. Disconnect the relay. Take an ohm meter and place one end on the relay terminal, the other on the fuel pump feed. This will check the wire (you will need extra wire to do this). SHake the harness at the tank end and see if you loose continuity. Its very very rare that there is a wiring problem. you may have a bad relay that is on its way out. nipper Link to comment Share on other sites More sharing options...
hankosolder2 Posted December 5, 2006 Share Posted December 5, 2006 Update: Fuel pump is not getting power to its connector, I have bypassed the pump relay with a wire jumper so it should be getting continuous power (yes, 12v at the relay connector). Anybody know if there is another connector between the relay and the pump? And/or where the wiring runs? Well, not to state the obvious, but you could go the hack route and just run a new wire from the switched side of the relay to the fuel pump...just under the carpet or something- then maybe do your real troubleshooting of the factory harness once the weather improves? Nathan Link to comment Share on other sites More sharing options...
NorthWet Posted December 5, 2006 Author Share Posted December 5, 2006 ... you may have a bad relay that is on its way out. nipper I bypassed the relay, running a jumper wire from hot-to-switched contacts on the connector and appeared to get no voltage at the pump connector. However... ...there seems to be some irregularities with my testing results, possibly due to the relay connector being near impossible for ham-fisted ETs to touch, pump connectors that use small contact pins, and environment that is both dark and freezing. I had thought about ohming from relay connector to pump connector, but my dark-and-frozen mind somehow didn't think to use the spool of wire that I was trying to use (unsuccessfully) to "ghetto" power straight back to the pump connector (hampered by the aforementioned small pins and a possibly non-working wire tap-splice). Take another stab at this Wednesday evening... Link to comment Share on other sites More sharing options...
Cougar Posted December 6, 2006 Share Posted December 6, 2006 Have you checked to see if you have 12 volts going the pump by placing the meter common lead to chassis ground? If you have voltage there then the trouble may be with the ECU. It supplies a ground to the pump. I have fixed a couple of ECU's with this problem. Link to comment Share on other sites More sharing options...
NorthWet Posted December 6, 2006 Author Share Posted December 6, 2006 Have you checked to see if you have 12 volts going the pump by placing the meter common lead to chassis ground? If you have voltage there then the trouble may be with the ECU. It supplies a ground to the pump. I have fixed a couple of ECU's with this problem. I don't think that the FSM shows the ECU grounding the pump, just the relay control circuit. I need to do some proper testing later today... last session was squeezed-in under less than ideal siituations. (This is wife's car, and she is getting a little peeved about having to drive the Aerostar around. "When Momma's not happy, NOBODY is happy!" ) Link to comment Share on other sites More sharing options...
Cougar Posted December 6, 2006 Share Posted December 6, 2006 You are probably correct. There is that design also. Either way, if you have 12 volts on both sides of the pump wires, with reference to chassis ground, when the pump should be on, then ground is not getting to the pump wiring. Link to comment Share on other sites More sharing options...
NorthWet Posted January 7, 2007 Author Share Posted January 7, 2007 Well, I got this problem resolved a week ago... I am not happy, since it leaves me with more questions than answers. Bad fuel pump. Should have swapped it out as second thing to do (after checking fuses), but weather discouraged me too much. Even after swapping it out I was unconvinced, because the car I swapped it into started and ran just fine. At least until the second time I tried to start it and it didn't. Regarding my new "questions", I was not getting any voltage to the pump with green connectors connected, even with the relay that I believed to be the pump's bypassed to provide constant power. As soon as I tried to start the car I got voltage to the pump. Maybe I bypassed wrong relay of the pair, or maybe there is something I do not understand. Thanks to all that responded. Momma's happy again. (Well except the moon roof seems to be leaking... :-\ ) Link to comment Share on other sites More sharing options...
Cougar Posted January 11, 2007 Share Posted January 11, 2007 Thanks for the update NorthWet. Glad you got it going again. I'm not sure if the pump is supposed to turn on with the green connectors connected or not. Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now