How Does It Work?

[docc]

My low fuel light is not helping me out. Even after I drained the tank, then tipped it up and finally poured gas out the petcock hole. The light would still not come on. I can connect across the connector and the light burns brightly. Is the thing in the tank a potentiometer? Some kind of bimetel thing? What?

 

BTW, when I refilled the tank after literally turning on end it would still only take 5.3 USG (specification is 5.8 USG). It also became amazing how much gas was trapped in the right side and in the two forward panniers.

 

Also, I found a ton of crap stuck in the petcock screen. Perhaps from when it fell over, every scrap of junk dislodged to stick in the pick-up. I'm glad I cleaned it and perhaps it will give me a little more range. I intended to switch to the manual petcock but the one I have ( thanks to T.X.) has a pick up screen too large of diameter to fit in the hole.

 

Any ideas?

[callison]

I believe it is nothing more than a magnet on a float activating a reed switch. I have been known to have been wrong before though.

[al_roethlisberger]

I can't tell you anything about how the low-fuel switch works... but can offer a couple thoughts on your other observations.....

 

During my pump-relocation "phase III" ... where I was constructing the next Mars Lander fuel-tank "balance tube" ... I too noticed that upon removing the petcock, that it's screen was just ENCRUSTED with white grainy gunk

 

But I guess this is really a "good thing" as it didn't make it through to the pump, so .... I guess the screens are doing their job.

 

 

In regard to the screen not fitting through the hole, I found this to be the case at first as well, but all I had to do is loosen the base that the petcock attaches to, and move it around a bit(there's wiggle room) and the holes will align and give more room to allow the screen/pickup-tube to slide right into that tank.

 

I haven't had a chance to really test my balance-tube/dual-petcock setup for increased range yet, but I'm hoping it will address the issue you mention about additional fuel being trapped.

 

 

All I know so far is that it runs

 

al

[TX REDNECK (R.I.P.)]

Doc, I thought the screen looked a little oversized too. Can you roll it a little tighter & then push it thru the opening ?

 

P.S. I see Al beat me to it, i'm a little slow on the draw

[al_roethlisberger]

The screen is(should be) a black plastic frame about 3 inches long, with rectangular screened holes. One end is threaded to screw into the petcock body.

 

There isn't any way to "make is smaller" in diameter... but again, the hole in the tank is quite large, and petcock base can move around a bit if the two button-head hex bolts are loosened to realign the base and hole in the tank. Once you adjust the base, the screen/tube should slide right in.

 

I had this same issue on the fuel-regulator side when I installed a petcock in it's place.

 

al

[Kiwi Dave]

I posted a query regarding this issue on the forum just over a year ago. (see Low-fuel Indicator, Dull Low-fuel indicator c/w other lites, 28 Jan 2003).

 

Replaced the sender unit four times now, the last time was due to fuel leakage, the first three in an attempt to cure this problem. I agree shorting the wires will give a bright indicator. My 2003 EV Touring also suffers this same problem, but I haven't tried to do anything about it.

 

Suspect that MG have changed the design from a simple reed switch as Carl describes in the 1100i California models to something more sinister like a solid state unit. As all my replacements were warranty related, I did not have access to cut one up and confirm this.

 

The cure? I reckon the only fix is to retrofit the older type sender unit, if they are still available. Not sure of the practicality, tho'.

[docc]

You guys aren't giving me any real good news. Something like, " Oh, yes, you can rebuild the sending unit with a toothpick without removing it from the tank."

 

No luck, heh?

 

The screen-tube on the stock petcock is white and , maybe, 4mm diameter. The black one on the manual petcock is more like 6 or even 7 mm. It was really a 'no-go' fit. I'll look again when the tank comes off next time as I'll be interested to see how much gunk has built up around it next time or that was all debris from Mandello.

 

What could go wrong with a float-reed switch? Dirty contacts? Bent reed?

[Cliff]

I haven't looked into the sender, but on my Corsa I got the impression that is was some analog type of device as the lamp would slowly light up or fade out

[Kiwi Dave]

I haven't looked into the sender, but on my Corsa I got the impression that is was some analog type of device as the lamp would slowly light up or fade out

 

Yeah, others have posted their lamp behaved like that too, but I'm yet to see it on any of my machines. I thought that perhaps I had to travel a certain distance on the reserve before the light would glow brighter, but how much further past completely empty do I have to travel?

[al_roethlisberger]

... mine also lights dimly, then grows brighter as the fuel level drops. ... on a 2002.

 

al

[callison]

My ISP dropped the ball and I've had no net access for two days, but I did save the post I tried to make and couldn't.

 

The problem with dashing off a quick answer is... giving a half answer isn't much different than being "half-asked" (pun).

 

It isn't likely to be a reed switch. They're off and on with no in between. The Guzzi fuel light does vary in intensity ruling out a reed switch. The mechanism is likely the same though, a magnet on a float, but instead of a reed switch, a Hall sensor instead. That would give a varying signal as the magnetic field near it changed strength. It makes no sense to use a magnet glued to the float in the presence of a volatile solvent like gasoline, so the magnet would be inside or part of the float. To speculate wildly, it is more likely that the Hall sensor is acting as the only active element in the circuit and so must pass the entire current of the dash light through it and perhaps it isn't sufficiently robust to do so or mechanically it just fails because of its environment. This is all just speculation. I don't think they're using a electromechanical (float activating a potentiometer) setup like the ones in cars because of the limited space. That leaves the wires themselves where they exit the assembly. If they're like the wires of the electomechanical fuel petcock, then they are subject to failure if the leads are flexed too much or crimped against something. I can only say that the ones I on my bikes have not given any problems in 100,000 miles accumulated miles on 3 bikes. Everything else, sure, but not the fuel sensors. Weird. I would surely trade my various problems for bad fuel sensors though.

[docc]

It's pretty hard this time of year to sort anything on the bike. But i have run two tanks of fuel through the Sport now and think it's back to normal.

 

When I cleaned the petcock screen it was packed with lots of sealing and milling debris. I can only assume that much of it dislodged from the cavities in the tank during the tip-over.

 

I'm also assuming that some of the junk was interfering with the movement of the floating magnet limiting its proximity to the sensor.

 

At any rate, cleaning out the petcock screen and rinsing the tank seemed to do the trick!!

[al_roethlisberger]

Yeah, the first time I ever removed my petcocks, there was quite a bit of the same white tank material caught in the screens. I've since flushed out my tank several times due to the various projects, so I hope I've gotten all this chaff out.

 

Glad to hear your sensor is working again now that you've cleaned everything out. I'm sure that's a relief

 

al