Fuel sender/fuel gauge system problem

[Rockfish]

Have had a strange problem with the fuel sender/gauge system for some time. As we are in the process of replacing the tank I was going to replace the sender unit, however as we had the tank out we had an ohm test done on the sender with it removed from the tank and it seems to be working just fine. The sender is not that old and is of the floating donut over a stainless rod type, not the lever type.

So the system works just fine much of the time and then for some reason will shut down and the needle on the gauge drops to indicate no fuel and will stay there. It never bounces around like it may if you had a fast intermittent bad connection; it just goes dead. There is a good chance that the next time you start up the boat that it will work just fine, especially if it has had a rest but not always and some times it will start on an immediate motor restart - but far less often and once it goes dead it is usually dead for a while. It never spontaneously starts up again with the boat running - it always needs a power off and on/motor restart, usually on a different day to start working again.

I hate intermittent problems. It acts almost as if an IC chip gets over heated and then shuts down and then works just fine at another time. It works enough of the time that it is still useful but still very frustrating. We don’t go that far and I know the boat very well so there is no chance we will ever run out of gas and in fact it is never allowed to drop below 60% full as we like the weight on the keel line for added stability and the security of a large fuel reserve. Don't really need a gauge, just top up the tank no less frequent than after every third fishing trip and the tank will stay above 60 % full.

Anyone got any ideas or suggestions ???

 

[Hourstonglassic]

Next time it happens if easy and accessible try grounding out the pink wire at the sending unit to see of the gauge pins, that will eliminate the gauge and wiring up to the gauge

 

[Rockfish]

Thanks HG. It would be possible to leave the tank cover not screwed down temporarily and hook the two sender wires up with temporary (non marine house type) marrette connectors and electrical tape for testing and even add a third testing wire out that we could use to ground out the wire to the main buss negative for a second when/if the system quits and see if the needle on the gauge spikes up to full fuel.

I assume that is what you mean by seeing if the gauge pins. If I understand it correctly that would be the same as having no resistance on the variable resistor in the sender – correct??? If it works and the gauge spikes, I assume that would indicate the intermittent problem is with the sender. If there is no continuation of the problem I assume that would suggest the previous sender connections were problematic and if the gauge does not spike that suggests as per your info. that the problem is with the gauge, wiring or ground.

Priced a new sender and it is around $60.00.

 

[Rockfish]

Hopefully my understanding of the testing options above is correct - ????

Am I understanding correctly that neither of the two wires that go to the sender unit is a ground but rather that they are a loop of a power circuit that the sender unit adds varying amounts of resistance dependent on the fuel level in the tank. If that is correct does it matter which of the two wires going to the sender that you ground out at the sender when the circuit intermittently fails in order to check the continuity of the circuit wiring and the gauge itself ???

Someone suggested to me that a poor ground on the gauge circuit is a fairly common problem with sender/gauge issues in marine systems and is worth redoing to be sure.

 

[bones]

Is your system on a network? Do you have a fluid level sensor network cable or is this an analog system? Mine does the same thing.

 

[Rockfish]

That is a very good question Bones. My understanding of small boat electrical systems is both limited and basic. I am assuming that my system is completely analog. I think the gauge itself is original 1983 boat vintage but the sender unit is more modern at about 8 to 10 years old. I assume a network cable system is rather modern/current technology that allows virtual gauges to be displayed on computer screens like state of the art chartplotters. I have no idea what is required with the new networked systems re: special sensors/senders or cables but assume I don't have them.

 

[TenMile]

Analog fuel gauge is a very simple setup. You have two wires -- the "sender" and "ground". Basically, when the gauge is powered on, it sends low voltage down the sender wire and measures the amount of resistance that drives the needle on the gauge. The sliding doughnut type are a magnetic resistor and in general should work long term as they are not prone to failure like the arm/float type. All fuel senders are designed so that if the connection to the sender fails, the tank reads empty (rather than full). Your issue is almost certainly in the wiring between the sender and the gauge. If possible, replace the wiring from the gauge to the dash with new. If you can't do that, redo all the terminations to the gauge and grounds.

It's possible and quite easy to convert analog gauges and senders to digital and read the info on a chartplotter. I did it recently and used a device from Noland called an RS11. You can leave the analog gauges in place, run a wire to the sender post on the gauge and convert the data to NMEA2000. I needed to do it to send fuel readings to the upper helm, and get accurate RPMs off both engines (gauges worked but not accurately). Got it working in an afternoon.

 

[Wild Bill]

All fuel senders are designed so that if the connection to the sender fails, the tank reads empty (rather than full).

While what you say would seem a logical goal in the real world it's not true. Different manufacturers use many different resistance ranges I won't bother posting however simply put some gauges will go to full open circuited and some will go to empty.

 

[gunnerlove]

Usually an screwed up sender/ circuit shows a gauge pegged at full full. The Reed types are known for rotting from the bottom up so the top half will still be reaping fine but you hit half a tank and they show empty.

 

[Wild Bill]

The more typical 0-90 ohm sender is the most common however gauge manufacturers can do whateve they want and then there are vehicle manufacturers. Troubleshooting at gauge or sending unit. Ensure there is a good ground at the gauge and good power, test light bright and or vom full battery voltage. power on with sending unit removed will either show a full tank or a empty tank. Ground the sending unit wire and the gauge should deflect the opposite direction, that proves gauge serviceability. Hook up gauge wires with wires unhooked from sending unit and insulated, not touching each other. The gauge should read the same way as power on and sending units removed from gauge. Connect the sending unit wires together and gauge should deflect the other direction. Sending unit will have a resistance typically less than 240 so you can remove it and see if there are any spots of the sender that don't work., there are wire units and thin film units and they can react differently. You should have 2 wires to the sending unit and not rely on a common ground for the ground circuit for the gauge.

 

[Rockfish]

Thanks to all of you. When we ran the sender on the Volt Ohm Meter you could watch the resistance change as we slid the doughnut up and down by hand, so most likely it is OK. Once the new tank and the sender are back in the boat, will start testing, and checking/reinstalling connections and the wire if necessary. If we are lucking, just replacing the connections we had to cut to remove the sender will solve the problem. The cover plate to access the wiring for the primary helm station and the secondary buss/fuse panel is on the wall in the head, so a least there is a comfortable seat while we work on it, lol.

 

[Rockfish]

Was at the boat and had a quick look into the helm station wiring compartment and there is a million wires going in every direction – looks like a road map of Victoria. The Fuel gauge appears to have three wires hooked up to it and what I assume is two positives using a common ground. The pink wire I understand is the positive for the sender, and assume the other would be the instrument cluster night light circuit or power in - not sure. The apparent ground (black) is daisy chained to other instruments and goes next to an old no longer used volt meter that was for the inboard, when the boat had an inboard. Now the volt meter function is in a smart gauge for the main outboard. Did the wiggle test and the apparent grounds at the old volt meter post could be moved. Perhaps that may be the problem.

Will remove, clean and reinstall the assumed ground wires on the old volt meter ground post. I am thinking we may also make up a short ground wire and ground the fuel gauge direct to the secondary negative buss 8 inches away. I assume that should be OK to do? Then I guess we cross our fingers. Could that black wire not be a ground at all but the night lighting circuit? - unlikely. Could their be a 4th wire (ground or positive) at the fuel gauge I did not see?

 

[profisher]

You are way better off to add proper neg buss bars and have each wire connect directly to a screw than to have multiple piggy backed wires. Just add a #10 or #8 size tin copper wire to the existing buss and add a new one. I have all mine labelled as well so if anything does go down I can find both the neg and pos quickly and either eliminate of confirm the problem is the wiring.

 

[Wild Bill]

The gauges only need 18 to 20 gauge however I like a double insulated to the sender direct from the gauge, not terminating at a neg bus bar with wipers, heater motors etc. I like this stuff, http://www.grote.com/products/82-55...plastic-wire/?industrycat=&prodcat=wire-cable
I know it says marine, not sure how if it meets some of the specs I know nothing about. Better than how most people wire. It's just every load weather it's a sending unit , light bulb, motor etc should have its own power and ground. I'll daisy chain gauge panel power or ground sometimes but stuff traveling across the boat should have their own ground and power imo.
So however I powered and grounded gauges, sender wires hooked up direct to the fuel gauge.
I typically wouldn't bother doing that with inboard, stern drive Temp, oil pressure etc, I expect the engine block to be a good ground.

 

[Bait Ban]

Proper marine grade wires should have tinned copper conductors to prevent corrosion (which increase resistance, causes heat and may lead to fire).