Fuel Level Sender Issues/Resistance

[ScottyB326]

Here's my situation:

Gas gauge doesn't work. Always reads full, and will once in a great while indicate actual fuel level (at least what I'm assuming is actual fuel level) but only once a month or something. Low fuel level light DOES seem to work.

I have an 1988 325is, it's got two fuel level senders. I pulled both of them out individually and checked the resistance, first right side up and then upside down to simulate full tank. The passenger side sender went from about 2 ohms to about 60 ohms. The driver's side went from about 50 ohms to about 150 ohms, although it seemed slightly sporradic. I looked up what they're supposed to be, and from what I could find, people say they should both be 0-60 ohms.

Now when I hook the brown/green wire that goes from the gauge to the drivers side sender straight to ground, it pins the gauge above full, and when I disconnect it, goes to below empty. Then I hook up a 1K resistor and it gives me 1/2 tank almost exactly. If a completely empty tank gives me a total 120 ish ohms resistance (according to the numbers I resarched) and completely empty off of my two measurements gives me about 210 ohms, I should get a similar result if I bypass the two senders and use about 220 ohms in resistors, which is what I did, and it BARELY moved the needle off of full.

Hope I haven't left out anything important.... but what I'm feeling like is that either the gauge is reading incorrectly, or there isn't enough resistance from one or both of the sending units when the fuel level gets low.

Anyone see some flaws in my testing or logic? Anyone know if the values I've found for the two senders are totally off?

Thanks in advance for any help!

[nando]

220ohms isn't something that the gauge will ever see.

the two sensors are wired in series. so their resistance is added together - at MAX, it will be 125 ohms (empty). at minimum, it will be about 6 ohms (full).

what comes to mind is the ground nut on the back of the gauge. the gauge itself acts like a second resistor (actually, it is wired in series with a 68 ohm resistor, and the gauge itself is also 68 ohms). if the gauge or the other 68 ohm resistor aren't connected well, the gauge of course won't function.

Do you have an OBC? what does it say?

[ScottyB326]

Yes I'm familiar with that 68 ohm resistor, saw that in the wiring diagrams.

OBC SEEEEMS to indicate true mileage left in the tank, because when i fill up it reads 300+ to go and when it gets low it'll say 22 or 19 miles or whatever it happens to be.

Yeah seeing the resistance in the drivers side sender at 150 makes me think somethings wrong with that, but then again I can't figure out why it took 1K to get to half tank, or 220 to even budge it off full. That seems to indicate the gauge is reading incorrectly or is faulty too?

[nando]

If the obc reads correctly, the senders must be working. check the gauge. Measure the resistance across the posts, should be 68 ohms as well.

[ScottyB326]

Ah yikes I didn't really think of that! Was too focused on the senders...

And yes i could see if there was a bad ground out of the gauge it would take a ton more resistance on the sender side to make the current choose the path of "least" resistance, in this case being a faulty ground.

I'll have a go at it after work tomorrow and report my findings....


.......not that anyone cares hahaha but in case someone has similar issues.

[ScottyB326]

Okay just for anyone who has a similar problem:

There's a threaded rod coming out of the back of both the fuel gauge and the water temp gauge that can be seen once the instrument cluster is removed from the car. They DO need the M4 x .7 mm nut on the back to pull them tight against the circuit board in order to ground. The previous owner didn't put the nut back on either side after the cracked dash was replaced. Now both the fuel gauge and the water temp gauge work.

Hope this helps someone else!