OBC Fuel Test #6 Question

[DennisG01]

Problem: I just filled up as the yellow, low fuel light had very recently come on and the gauge needle was now at approximately 1/16 to 1/8 when I filled up. After filling up (took the expected amount of fuel... 14.6g), the needle STAYED at that 1/16 to 1/8 position with the yellow light still on. The only thing I did differently during this fillup is that I reset the trip computer/mpg avg/mph avg during the process of filling up. I drove 7 miles to get home.

I have a "Low OBC".

I searched on this site for help and came across the ability to access the OBC and see specific info. I did the "fun" gauge test first and the gas needle moved normally through it's sweep. I then "unlocked" the OBC through Test 19, and proceeded to Test 6. I got these results:

340000 _ _ 6.0
_ _ 0340 _ _ 6.1
_ _ 0055 2 _ 6.2


I "think" this is telling me that one of the level sensors is reading properly and the other is not. I also "think" that the "2 _ 6.2" tells me which side is wrong... possibly the "340000" also tells me which side is wrong?

Can someone please confirm if I'm on the right track or what exactly those numbers mean? I read through the info I found on other threads, but am a little confused.

I have turned the car off/on a few times to see if I would get lucky and it would "fix itself"... but no luck

[DennisG01]

Just a shameless bump as there are a lot of posts in this forum and it probably got lost after a day or so.

[pleiades]

Just a shameless bump as there are a lot of posts in this forum and it probably got lost after a day or so.

Maybe this post from Qsilver7 will help with the low cluster. Looks like your right sensor is faulty, probably the float lever is stuck and submerged in fuel after being allowed to dry out on an empty tank....

https://www.bimmerforums.com/forum/s...5#post29141145

[edjack]

They are L/R/Total. The last number s/b 12.2. I have no idea. Might try a bottle of Techron; the rheostat may be dirty.

http://www.ideaphile.com/bmw/540-obc.html

[DennisG01]

Excellent - thank you for the link and the explanation. That's exactly what I needed to know. I'll pull the right sensor and take a look.

[pleiades]

Excellent - thank you for the link and the explanation. That's exactly what I needed to know. I'll pull the right sensor and take a look.

Another slight possibility is that the float on the right side sensor lever somehow detached itself and is bobbing around in the tank.

[DennisG01]

I must have missed your post when I was replying to the Seven Sisters (I recognized it, but have to admit that I had to google to get the exact constellation). Thank you, Ed.

Obviously I'll no more when I pull the float assembly out - but I don't think the float came off since the gauge continued to work over the last few days. It's just weird since when I filled up, the needle stayed put right where it was immediately before I filled up. Over the next couple days, it went down normally as one would expect with fuel usage - it seemed to move in direct correlation to the amount of fuel being used... Until it eventually got to completely empty with "---" showing for "miles remaining" on the dash readout.

[DennisG01]

Small update... I pulled the fuel level sender and tested it at the pins where the outside electrical connection attaches. The rheostat didn't "appear" to be excessively worn. As I moved the arm, my meter didn't change - showed infinity ohms. Then I tested at the little pins on the backside of the rheostat... functioned normally/as expected! So, thinking that the issue must be with the wires between the rheostat and the electrical connector, I connected back to the connector pins and played with the wires... yup! Got it work a little bit. And then the wire actually broke apart, completely (see pic of underside/fuel side of the electrical connector). So far, I have stripped the wire back and soldered it. I don't know if it's needed, but I'm going to get some marine, fuel-resistance sealant and coat the soldered area.

FYI... before anyone asks... and I KNOW that some would (heck, so would I!)... I did NOT do the soldering right there by a gaping hole in a 3/4 filled gas tank. I don't "think" a soldering iron would cause a ka-boom... but I wasn't taking any chances! I removed the sender/pump assembly and let it air out for a while, then did the soldering on the workbench.

[cnn]

This could have been fixed with "butt connector".
Simple, no need to solder.

[DennisG01]

I guess I was just overly concerned with the part about it being submerged in fuel a good part of the time. But you're saying I'm overthinking it? No reason then to coat the solder/exposed wiring with a sealant? Butt connector may have been difficult as the wire is super thin.

[edjack]

Good solution! Congrats.

The insulation appears to be brittle.

[DennisG01]

The insulation definitely has the "look and feel" of being on the brittle side. I played around with all of the wires and they appear to be at least "OK". I didn't notice any other wire showing obvious signs of splitting or feeling like the wire on the inside was compromised. But it does have a 160K+ of miles on it, so I'm not expecting miracles here.

EDIT: Woke up this morning and realized there are other, soldered spots (rheostat board) that aren't protected in any way. There are also push-on electrical connectors (fuel pump). So I think I'm good to go with the way it is. I'm going to make sure that the soldered section stays "straight" and doesn't have stress on the joint section as vibration, over time, can weaken or break the joint.

[DennisG01]

Final update (for now)... got everything back together and the gauge went right back to where it should be (currently, approx 3/4 tank). For now, everything is working normally... until another wire decides to break...

[DennisG01]

So much for the "final" update, previously...

The gauge actually ended up doing some wonky things for a few days, which I attributed to the two sides of the tank trying to level themselves out again. But after a few days it began acting totally normal. So, when the tank got low (needle at about 1/8th) I filled up. Started the engine and the needle didn't move - same as last time! Drove home and did that test again... sure enough, getting a bad reading from the right side level sender. So I continued to drive it and the needle dropped to completely empty over the course of a couple days, as you would expect. That's where I'm at now... yellow light staring me in the face, even though I have about 3/4 tank. I'll drive it till about 200 or 250 miles on the odometer and then pull the sender out again and take a looksie.

[JimLev]

You should try dumping in TWO bottles of Chevron Techron Concentrate, the big bottles that treat 20 gal of fuel each.
If the slider or pot windings are at fault that will clean them up.
I had a similar problem years ago. After a few days of driving my gas light came on. I pulled into a gas station and it only took 6 gal. Dumped some Techron in the tank, it's been fine. I now use a bottle of it every 4 months.

[Qsilver7]

So much for the "final" update, previously...

The gauge actually ended up doing some wonky things for a few days, which I attributed to the two sides of the tank trying to level themselves out again. But after a few days it began acting totally normal. So, when the tank got low (needle at about 1/8th) I filled up. Started the engine and the needle didn't move - same as last time! Drove home and did that test again... sure enough, getting a bad reading from the right side level sender. So I continued to drive it and the needle dropped to completely empty over the course of a couple days, as you would expect. That's where I'm at now... yellow light staring me in the face, even though I have about 3/4 tank. I'll drive it till about 200 or 250 miles on the odometer and then pull the sender out again and take a looksie.

When doing repair or replacment of the fuel level sensors...after finishing the work...this is the time you should actuall do TEST 21 (it resets the software). The fuel level sensors are buffered...so it takes a while for things to settle...this is to prevent wild swings of the fuel gauge. But doing a reset will start you off with a new "base" of data.

See the explanation of TEST 21 in info below:

[cnn]

You have 3 options:

1. Ignore it and use the odometer as a guide, for ex, refueling every 200-300 miles etc. etc. but this is somewhat inaccurate and can leave you stranded.

2. Look carefully at the sending unit, there are 2 curves and the level sensor prongs gently touch them and go up and down as the fuel level changes. Sometimes impurities collect on these 2 curve tracks. This is why Techron fuel cleaner helps.

3. There is another situation: the electronics (basically rheostat) is screwed up and no matter what you do, it shows a fixed level on the gauge. In this case, you need a new level sending unit.

Stick to BMW sending unit, I understand the car is old and people don't want to spend much on a 20-year-old car, but this is an important device b/c it can leave you stranded (aftermarket units are virtually all Chinese).

[DennisG01]

Thank you guys, all three responses are very informative and helpful.

I'll definitely use that "reset" test next time!

I've recently used some SeaFoam (I do that every so often), but that's a good idea to use the Techron - especially at the double/triple dose.

I'm still going to pull it out when the fuel level gets a little lower - just to see if there's something obvious like there was last time. I'll just some of that Techron directly on the rheostat with a toothbrush, too. Maybe I'll soak it overnight in Techron, first - that seems to work well with the O2 sensor in my Yamaha outboard - I use Yamaha RingFree for that, but it's the same idea (it's a 250HP, 2-stroke). Actually, because I use RingFree with every fill-up, the O2 sensor is pretty much totally clean every time I pull it out, anyways.

But if I do end up replacing, I'll definitely use a quality sender - I don't mind spending the money on parts as long as they're a quality unit. At least it's a pretty easy job to do.

[DennisG01]

Took the sender out. Same wire is broken as before (one of the leads coming from the rheostat). I initially thought I used too much solder by accident and it failed where the solder stopped, creating a pivot point. But on closer inspection, it appears "cut". I think when I initially stripped the sheathing off to solder it, I probably cut slightly into the wire (it's very thin. I'm going to resolder it, but this time I'll burn the sheathing off. If I can find some small enough shrink tube, I'll use that to reinforce the joint, as well.

[DennisG01]

OK, got the job done... again

It'll be a day or two till I fill up and then do that reset. But the gauge needle is back to working.

If I do that Test 21/reset... will I lose anything besides avg mpg and avg mph?