Fuel Gauge Issues on my 2002 21i type R

[Rich M]

Bought the boat 2 years ago. Love it but the fuel gauge has never worked since I got it. The gauge bounces like a Ping pong ball continuously. I've replaced the gauge with new. No change to the problem. I've checked the sender and it's working fine. Had a mechanic look at it and he tells me that since the fuel tank is narrow and that the sender is on the port side, the baffles have no effect and if the boat rocks a bit, the fuel slops around too much and the sender can't 'send' an accurate signal to the gauge.....which is why the gauge bounces. Really? Hard to believe that this is a design flaw on a decent boat. Anyone else heard of this issue?

Rich M

[chpthril]

No change to the problem. I've checked the sender and it's working fine.

Welcome to the site, Rich.

Tell me how you tested the sender? These systems are really easy to test if you have a good DVOM. Is the bouncing less with the tank full, seems it would be if your mechanic's theory was accurate.

[Rich M]

Thanks for the welcome note and quick reply. Looks like lots on valuable resources on this site.

I threw a meter on the sender. No issues with it sending and the float is moving up and down just fine. When the tank is full the fuel gauge is pegged at full and doesn't move. So yes, after I burn a bit of fuel, that's when the dancing gauge starts.

[IDBoating]

Presumably this is an analog sender and gauge. You could put a cap from the signal to ground to act as a filter and get rid of the variations, leaving only the average which is what you really want anyway. Are you comfortable with electronic components? This would be a very easy thing to test.

[Rich M]

so how would I do that?

[IDBoating]

It will take some experimentation, since we don't know the impedence of the circuit. A local Radio Shack should carry discrete caps. I'd start with something like a 0.1uF tantalum. This is a polarized component with two leads. On the meter, put the negative lead to ground and the positive lead to the signal coming from the sending unit. You will see that the variations slow down a lot. Increasing the capacitance value will slow them even more. In most applications it's possible to go "too far" and lose data, but gasoline consumption is so slow that you probably won't run into that here.

In case anyone cares: The reason this works is because a capacitor's impedence is inversely proportional to frequency (the formula is 1/2PiFC). That means the higher the frequency, the lower its impedance (think: resistance). A cap to ground thus blocks DC (zero frequency) signals, but lets AC (nonzero frequency) signals pass through it to a degree based on their frequency, in this case to ground. Therefore the short-term variations caused by the fuel sloshing will be passed to ground, and the cap will average their values into a long-term value that your meter will display.

Another way to think of it is that the cap acts as a fast battery. When a slosh causes a "peak" voltage, the cap starts to charge up to that voltage but long before the cap can get there, the slosh and its peak voltage are gone. Likewise, a slosh in the other direction causes a "dip" voltage, and the cap starts to discharge down to it, but never gets there because the next "peak" slosh comes along and causes the cap to start charging again. The bigger the cap's value, the longer it takes to charge/discharge, and the more filtering you get on the signal. Thus the cap acts like a simple averaging circuit, which is perfect for a fuel gauge because the actual fuel level is likely a rough average somewhere between the peaks and dips.

Hope that's not too much information. You can try this experiment for a couple of dollars at the local Radio Shack. It's a heck of a lot easier than other solutions I can think of!

[IDBoating]

Also: Since you are comfortable putting a meter on the sender, you could do that again with the cap on the leads of the meter and you'll see the variations slow down as I described. Disconnect one lead of the cap and the readings will swing around as "normal"... connect both leads and they'll slow down. The bigger the cap, the slower they'll move.

[boatwakes]

If you remove the sending wire from the sender (pink) and tap it to the tank, the gauge should peg over full, if it doesn't, still have a gauge issue. If it does, the sending unit is bad. Get a WEMA unit for your tank which should be about 7-8" deep and that should solve the problem. ONE VERY IMPORTANT THING TO NOTE: all tank sending units have an offset 5 screw hole pattern so you have a 20% chance of placing it correct the first time unless you know this. Use the new gasket to line up the holes on the new sender then index(mark) the gasket and sending unit together. Then place the gasket on the tank and find the correct pattern and mark the tank off the original index on the gasket, makes install that much simpler.

[Rich M]

Excellent. Thanks Guys for the input. Looks like I've got some homework to fiddle with, but am sure I can muddle through it. The bigger problem I have right now is that I just winterized the boat. Will have to save the link to the thread and give it a go in the spring. the input here is very much appreciated, and I'll reply to the thread when I've done some testing.
Rich M