Fuel Sending Unit...

[motitties]

OK.. I know this isn't exactly an engine problem but I didn't see a topic regarding "fuel delivery system" even though there were subjects like turbos and intercooling... Great stuff but I'm just trying to keep this old soobie roadworthy.

One day my fuel gauge just stopped working. Like BAM! Says no fuel in the tank. So my first inclination is a broken or loose electrical connection somewhere. Obvious. So I pull the fuel sending unit/pump (this is a '91 FWD so passenger side only) and everything looks pretty good. Cleaned the "wiper node things" with acetone and put 'er back in. With a low tank I measured a resistance reading of 80 ohms... poured in a gallon of fuel and got 76 ohms. So the sending unit seems to be functioning correctly, but could there be something else in the unit that would cause the fuel gauge not to work?

The gauge does move very slowly up from "super empty" once the engine is running but never gets to where it should be. Any ideas?

Thanks!

[lucas]

dude could just be your clusters old and decided to take a crap?

[motitties]

Possibly... is there anything between the sending unit and the gauge that could cause problems? I really don't want to take apart the dash... ='(

[turboleg]

Check out section 6.
http://www.main.experiencetherave.com/s ... SM_-_FULL/
Page number 52 of section 6-2 is where the good stuff is.

It appears fuse 15 is in line with what your discussing. Suggest checking that first.

[motitties]

I checked the fuse box... no blown fuses. I'm thinking its the gauge itself. If a connection is loose but not broken it would follow that the resistance would be much higher than it should be causing the bimetal coil to move much less than it should. Like I said, it does move... just barely. Any ideas on how to test this gauge? If I applied 12v to the harness from the fuel tank shouldn't that make the coil operate?

[turboleg]

If the fuse is good then take measurements at the gage. The schematic shows three wires. One will be nearly equal to the battery voltage with the key in the run position (this is fed by the fuse discussed). One will be ground which can be determined with the ohm meter (nearly zero resistance), and one will be a ratio of the battery voltage based on the level of fuel in the tank. All voltages reference ground. If any of the above are not true take a deeper look as to why. If all is true then its likely your gage.

Sent from my SCH-I535 using Tapatalk 2

[motitties]

OK so I think I might have just really screwed things up... I pulled the cover off the instrument cluster and realized that the gauges don't look like replaceable parts by themselves.. Do I need a new instrument cluster now or can those needles (engine temp and fuel) go back in at any point... they seem really small and don't appear to have a keyway of any kind. Did I just frack things up even more? I seem to do that a lot.

[turboleg]

I would replace the entire gage cluster regardless if one is broken. I'm not sure anyone here will really want to disassemble one to get just one part out. Did you verify that it is indeed your gage?

Sent from my SCH-I535 using Tapatalk 2

[sqc151]

ive split apart quite a few clusters. id recommend just getting another.
i might have an extra cluster in my stash if you need it.

[Legacy777]

What year legacy is this for?

Honestly, I don't think it's the cluster, they don't typically go bad. However you can check the resistance at the cluster connectors to see if the resistance you measure there is the same as what you measured at the tank sending unit. If it is, then the wiring is ok, and the cluster is probably the problem. However if the resistance is different, then the problem is either in the wiring or most likely the plastic connector under the seat.

The pins you want to test the resistance across at the gauge cluster are pins 9 and 12 on the i16 connector. This diagram should help.

http://www.main.experiencetherave.com/s ... iagram.jpg

[motitties]

It's a '90 FWD. I haven't been able to figure out where the connector for the cluster is. Is it directly behind the cluster? I took the cover off it but I didn't pull the cluster out. I measured the voltages on the solder joints of the gauge itself. I got 11v to one and something like 5 between the other two? Does this work the same as testing at the connector?

[Legacy777]

The connector is directly behind the cluster. You have to pull the cluster out to get at the connectors.

No, the testing you did is not the same testing you would do at the connector. You want to measure just the resistance of the sending unit and not any of the voltages.

[turboleg]

No, the testing you did is not the same testing you would do at the connector. You want to measure just the resistance of the sending unit and not any of the voltages.

I second that. The measurement has to be completed with the cluster seperated from the connector. The gage itself has resistance. So if its tested together with the fuel sending unit you will get skewed data. Also, if tested together, there is no guarantee which of the two (gage or sending unit) is defective.

[motitties]

OK so I finally got back to this little project.. Looks like there is no resistance (open circuit) between pins 9 & 12 Brown yellow and Red white respectively. If I remember correctly I did read resistance through the connector at the fuel sending unit which could mean that one of the wires is disconnected and grouding out. Anyone know exactly where this connector (R4) is located? Under one of the seats?

[Legacy777]

The i32/R4 connector is on the passenger side right where the dash meets the carpet area. It's the bottom left connector in this picture.

http://www.main.experiencetherave.com/s ... G_8421.JPG