Intermittent fuel level sender

BR-549

Joined Sep 22, 2013
4,928
I don't know how old your tractor is, but depending on the time of year, if you leave the same gas in a tank for ~ months, it can sour, I call it turning to varnish, cause that's what is smells like.

Maybe it has set in the past.
 

Thread Starter

wml52

Joined May 8, 2014
10
So those are reed switches. We are making physical electric contact in a gas tank.

Shocking.
I thought they might have been reed switches too, but when I pass the magnet over the locations I can't see any physical movement, even under a 10x eye loupe?
 

Sensacell

Joined Jun 19, 2012
3,420
Very interesting system.

Just wondering about the TWO magnets? that's got me curious, is it doing some logical magic to create more measurement steps by using the two magnets?
 
Last edited:

BR-549

Joined Sep 22, 2013
4,928
After the board is repaired or replaced...... you'll find out the output.

Most interesting.

Maybe a bad reed switch?

Maybe its not a reed switch. If two wires are barely touching, will a magnetic field vary the resistance?

Puzzling. Maybe a call or visit to JD.
 

wayneh

Joined Sep 9, 2010
17,495
Very interesting system.

Just wondering about the TWO magnets? that's got me curious, is it doing some logical magic to create more measurement steps by using the two magnets?
That was my thinking. Pretty clever, and a solution to the old wire rheostat issues. But maybe it has its own issues.
 
Interesting.

Definitely looks like reed switches. In a 6 position sender, it looks like it "selects" a tap. So picture a set of resistors in series with the reed switch picking off a position, so you get 6 discrete fuel levels.

I'll bet the field is arranged such that the two magnets they will always activate one switch even when "in between" positions. It's really the only way it can work.

An ohmmeter would be your friend here.

I suspect that something is causing the float to hang up.

If I did my math right http://www.roberts-roberts.com/documents/Using Properties to Manage Flammable Liquid Hazards.pdf this says about 2K ohms per meter.

Watter is not very conductive. Pure water about 18 M ohm-cm for a DI water system.

Ethanol may mess it up more.

So, it;s possible that the conformal coating has worn away and the gasoline is affecting the readings.

You should be able to test that. e.g. A coffee can of gasoline or say a PVC pipe capped at one end. Compare resistance readings with water and gas.
 
Last edited:

Lestraveled

Joined May 19, 2014
1,946
KISS may have hit on something. The first fuel level sensor I used stopped reading the fuel level, the level was stuck at 3/4 tank. Long story short, the float that carried the magnet swelled and got stuck. I disassembled it and sanded down the float. I now have a spare fuel level sensor.

@wml52 How freely does the float move in the cylinder??
 

Alec_t

Joined Sep 17, 2013
14,264
Looks like an 8-position sensor to me. The reeds are numbered up to 8. The spacing between reeds 4-6 is closer than for the other reeds; presumably to correct for the non-linear gas tank geometry.
 

cmartinez

Joined Jan 17, 2007
8,212
Looks like an 8-position sensor to me. The reeds are numbered up to 8. The spacing between reeds 4-6 is closer than for the other reeds; presumably to correct for the non-linear gas tank geometry.
Simple, and probably reliable... I wonder why they didn't use hall sensors instead of reed switches. Some reed switches can handle only very little current (I've seen some that can handle only less than 1/2 a milliamp), maybe that's why there are resistors connected in series with them.
 

bwilliams60

Joined Nov 18, 2012
1,442
I did a video on one of these for my students at school. We use these on our Class 8 trucks. What you have is a laddered Reed switch setup. As the float moves past a switch, the switch will close and complete a circuit which adds or subtracts resistances depending on which way it is moving. Under a USB microscope, I was able to see this happening and created a video on it. I will attach it here along with the video if that is possible. As soon as I find them, lol. As for your problem, thinking you may have a float composition problem. Nothing else can really go wrong in this unit.

What video file extensions can be used on here and how large can the file be?
 

Attachments

Last edited:

Thread Starter

wml52

Joined May 8, 2014
10
KISS may have hit on something. The first fuel level sensor I used stopped reading the fuel level, the level was stuck at 3/4 tank. Long story short, the float that carried the magnet swelled and got stuck. I disassembled it and sanded down the float. I now have a spare fuel level sensor.

@wml52 How freely does the float move in the cylinder??
The float has plenty of free play, no hint of sticking or rubbing of any kind that I could observe.
 

BR-549

Joined Sep 22, 2013
4,928
wml52, what's your plan? Repair or buy board? I would suggest putting float in fuel for a while....to see if it retains buoyancy. Have you priced a board and float yet? That will probably knock you over.

les, who was the original manufacturer of your dune buggy sensor? And have you seen this before?

cmartinez, it will probably turn out that this kind of sensor, has been used since they started using reed switches.

I was expecting some kind of analog resistance. Really surprised me.

Staggering the switches to compensate for tank shape is pretty neat too.

bwilliams, that float could very well be bad. But I certainly would not put that board back in there.
That board is in terrible shape. From the look of it any way.
 

BR-549

Joined Sep 22, 2013
4,928
I don't think that's the same thing. I assume your example is analog out 0-90 ohms, resistive wire.

I believe this sensor has an output similar to this.




I was thinking it would be something like les has.
 

BR-549

Joined Sep 22, 2013
4,928
Roger on that. This is what your link said.

"VDO tube type fuel level senders use the least amount of space within a tank, as the float is contained within the aluminum cylindrical housing and rides on two resistive wires. You don't have to worry about the float arm direction or extreme fuel sloshing found in many tanks. The fuel level senders are available in nearly three dozen standard lengths, from 150mm to 800mm, and feature a standard SAE 5-hole aluminum mounting flange suitable for gasoline and diesel fuels. A neoprene mounting gasket and M4 stud and .25 in. spade connections are provided with each sender."

Well, that's good to know. Then it should be reasonable to replace. I didn't know it was common. I figured JD would charge a fortune.

Edit...Is that crud on yours, part of the original coating?
 
Last edited:

Lestraveled

Joined May 19, 2014
1,946
........Is that crud on yours, part of the original coating?.......
Upon close inspection, it looks like a conformal coating. The sensor was removed after about 1 1/2 years in the tank. A little hard to tell. The opposite side, where the float slides, looks to have been masked off.
 

BR-549

Joined Sep 22, 2013
4,928
The internal coatings don't have to be smooth and pretty I reckon. It very well could be dobbed up.

I recall seeing sensors like that at tank farms and old boilers. They were external. But I never thought of fuel tanks.
 
Top