230SL Gas Gauge Goes Crazy after 30+- minutes of driving every time...see video

[AllenF]

Hi Folks,

So this is an odd one.  New gas tank, screen, sending unit a couple years back.   After about 30 minutes of driving (sometimes 40) hilly, curvy driving (it's all we have here) the gas gauge goes crazy for about a minute then shuts off.  After parking and sitting a bit, it's all back to normal.   I happened to catch it going crazy as we were getting on the freeway in this video.   

Maybe an intermittent connection in the sending unit?

https://youtu.be/k_0gDywzeAI

Thanks for any tips.

Allen

[Pawel66]

I do not see it on the video, but - is it a new gauge? Or it just started doing it one day recently?

It is a 230SL, so you have the float type sender, not tube.

[AllenF]

thanks, video is working for me.  youtube must have still been processing.   it's the newer tube type, and it was all new a couple years back. It's the later serial number so the tube is correct.  Works perfect until....

[dirkbalter]

I do not see it on the video, but - is it a new gauge? Or it just started doing it one day recently?

It is a 230SL, so you have the float type sender, not tube.

Pawel,
Only if it’s an early 230. Thr later 230s have the tube as well.

[Pawel66]

Thank you.

I watched the video on a bigger screen - I can see it now. I think you are driving too fast 

I am not a mechanic, I was just working with sender\gauge set in my car. Two things come to my mind:
1. Indeed, some lose connections: at sender, tail harness connector, under dash, instrument cluster. I would start with wiggling those easily accessible while watching the gauge.
2. Gauge itself. What I noticed with gauge bought in internet store - you had to work with the screws that were holding the gauge together, loosen some and tighten some. Then the hand was working properly. There was originally too much tension there that was blocking the hands movements. Also: in aftermarket gauge the fit of the orifice for the bulb was so so. I had o use Dremel to enlarge so that the frame does not push on the bulb (hand was moving, reserve light was coming on as bulb's base was shortening everything to ground). Do you recall any issues like this while installing the gauge? Do you drive with your lights on? Is it possible that something expands a bit because of heat and pushes on something? 

But these are just suppositions...

[AllenF]

I ran a second ground from  the sender plug, and it didn't display its crazy behavior on a 40 minute drive last night.   Could be that it was just a crappy connection at that ground.  We shall see.

And yes, I have a 3/75 Euro rear end, so indeed I was driving too fast!   

[Pawel66]

 

I do not remember now how this ground is hooked up there... will look at it when I can.

[AllenF]

Looks like there might be a ground stud behind the rear carpet area, so I probably will have to do a jumper.

[lpeterssen]

The ground for the fuel gauge circuitry is at the trunk next to the left side rear wheel arch. That grounding point is shared by the fuel sender unit and the fuel pump. It has a thick ground for the fuel pump and a thin one for the fuel sender.  Clean and sand the post o rings and use conductive grease (carbon grease, non dielectric) there to avoid future corrosion of the post.

I was going to send you today pictures of the other possible culprit which were faulty connections at the instrument cluster itself.

Later will do anyway just in the case that problem arises to some other MB enthusiast.

[lpeterssen]

The central instrument cluster where the fuel gauge resides, has on the back a connection board circuitry which contacts (circled in red on attached photo) need to be well polished in order to have a perfect connection to it (fuel gauge)

Take the time to clean and sand with grit 240 or similar abrasive paper those contacts.

Also check continuity between those three post and the male pin terminal at the 12 pin connector which interconnects this central cluster to main wiring harness.

The terminals involved on the fuel gauge operation are:

1. Blue/black (negative signal from sender) fuel level signal
2. Blue/green (negative signal from sender) low fuel signal
3. Black/blue (positive signal from fuse no.2)

It is always a good idea to polish each of the post of this male 12 pin connector to assure a perfect signal connection to all central cluster instruments.  Do the same thing on the female connector at the main wiring harness.  Make some sanding paper rolls very thin so that you can sand this contacts internally.

Finally when re-attaching the circuit board to the central instruments check that everything has its screws in place and is tight. 

Some carbon contact grease will not hurt on each of the connections, since it will improve signal passing.

Best regards
L.peterssen

[AllenF]

Thanks for the detailed guidance.   If it isn't the ground, it's surely this connection at the back of the cluster.  I don't think I disassembled that or cleaned it when I had it all apart, so it's suspect.

I will report back once the gremlin reappears.

Allen

[AllenF]

Fixed

Turns out it was a manufacturers defect.  Lug on one side of the resistance loop was not riveted (punched) onto the copper flange.   I forced it over the lug and properly mushroomed it (and its sister) and voila.   

[Pawel66]

To be honest, the internal issue inside the sender would be at the end of my list of suspects...

[AllenF]

Almost missed it really.  I was just checking the resistance at each limit, and noticed the contact bar was swinging around.  This is NOT an MB part...probably the OEM parts have better QC?

[Pawel66]

You will hear many voices on this, most of them much more knowledgeable than me, but I would say: a part made for Daimler by supplier X is usually better than the same part, not made for Daimler. It is based on my fairly narrow experience.

[waltklatt]

Allen,
Glad you fixed your problem.
Thank you for the clip with your driving, brought back memories of my SL.
Which has been stationary for the last 15 years in the garage.
Walter