The Stovebolt Site Forums The Shop Area The Electrical Bay Headline: Gas Gauge Sender Causes Explosion!

Ok, I’m looking for an answer from a FIRE expert here. I don’t want to be listed under the heading above as another flight 800 type incident.
The fuel gauge was giving me some screwed up indications so I pulled it last weekend to see if I could find the problems, I did! First off, I’ve never seen an original 1950 AD sender so I can’t be sure but I think this is a replacement sender. I would like to see a couple of pics of one if any one has one for comparison.

The first thing that came to mind when I pulled this thing out is that the wire wound variable resistor sits right down in or out of the fuel depending on the level. I expected a sealed case of some sort. In other words you have a hot wire right down in the fuel tank. As long as the liquid fuel is covering the element it isn’t going to start a fire. When the level is down and the element is exposed and in the fumes, if there is a spark you could have an explosion. You say how could you get a spark? Well, the wiper is probably going to be in contact with the element at all times so it would be rather unusual for it to spark.

But if something fails and the circuit is broken there could be a spark. Now in my case there is a 3” long .004” band that connects the wiper to ground and allows the float lever assembly to move over its entire range. This flexible band on mine broke into four pieces and they were lying loose in side the unit and could easily have shorted the hot side to ground causing a spark. And it would be my guess that this thing had sparked quite a few times before I discovered it. If there was enough oxygen present it seems to me that this could have been a disaster.
Are there any opinions out there??

Denny Graham
Sandwich, IL
p.s. I’m attempting to rebuild the sender, I’ll let ya know if I’m successful if ya care.

newer cars have electric pumps that have little electric motors running inside the tank, I have pulled more than a few that were burnt looking. But I never hear of any blowing up, & I have seen several cars burn to the ground, no explosions whatsoever.

Denny, I am no electrical guy, but I do not believe there is any "hot" lead to the gas tank. The wire from the gauge to the tank is a ground wire with the hot side at the gauge. There is no current flowing to the tank.

Denny Graham,

This is what I think. The resistance of the coils in the fuel gauge itself allows only milliamps to pass to the sender/ground.

Gasoline will explode in a VERY narrow fuel to air range..around 15:1. I'm not a mathematician, so I don't know when this ratio is developed inside the tank as the fuel drains and the atmosphere enters..or if it ever does.

The very bottom line would be to ask yourself how many threads have you seen here describing the horrible carnage caused by the pesky sender igniting the AD/IED.

Stuart

when you have the sender out, hook it up and fiddle the ground connection, see if you can cause a noticeable spark [assuming your gauge is OK and wired properly] .... it ain't gonna happen - the gauge passes all current to a ground on the cluster when the tank is empty or the sender ground is absent - current to the sender ground is miniscule to activate the coil that draws the needle up, and I'd doubt it could cause a spark under any but very unlikely circumstances .... the system has been in use for ages and I've never heard of a tank exploding

Bill

I,ve wondered about this myself for many years. The fuel gauge wire & other wires like the dome light run together under the rusty floor. What if they shorted out. Same with electric fuel pumps in the bottom of the tank. Must be safe though or you would hear of vehicles blowing up all the time.

I might have told this tale before...I have forgotten. I worked for Standard Oil at a marine terminal. The company put on a seminar on the safe handling of fuels, tank truck loading and static electricity. A large glass bell jar affair was filled with a atmosphere of fuel/air in the 15:1 ratio and a hefty static charge was fired across two electrodes to ignite the charge.

Much dinking around with the fuel/air mix was required before the whole shebang went off.

It was much more difficult to cause the explosion than the average person would have thought.

Stuart

My sending unit has the rheostat completely out of the gas all the time and it's not sealed away from the fumes. The tank would have a vapor rich atmosphere that would make an explosion unlikely(kind of like trying to start a flooded engine).

Keep it comin' guys, I'm havein' fun with this one.
Denny G

Now, back to the original thread where I was asking for opinions on the bouncing ball trouble.
I got her all restored and I'll discribe it here: http://www.stovebolt.com/bboard/cgi-bin//ultimatebb.cgi?ubb=get_topic;f=3;t=002823
Denny Graham
Sandwich, IL
Oh, and it dint splode when I turned on the switch!

Iside a gasoline tank there's very little air because of the vapor pressure of the gasoline. It's nearly always too fuel rich to ignite. If you take out something like a level sending unit, that little area around the hole you open up will get enough air diffused in to ignite but you'd have to force air in to get the right mixture in the rest of the tank above the liquid.

Never going to happen, that was the point to my original reference to flight 800. That’s what the engineers at Boeing calculated also for the 747. Millions of miles without incident, and it was the only that one time when the conditions were just right when it happened. The potential was always there but since no one ever recorded a problem they figured there wasn’t one.

15:1 is the ideal mixture for combustion under pressure in a combustion chamber, but if your saying that an open bucket of gasoline will not ignite unless there is a 15:1 ratio or that fuel spilled on the ground isn’t capable of ignition until it reaches a 15:1 ratio I would have to see more proof before I would believe that. Fuel vapors and air in my book will ignite if given enough heat for combustion over abroad range of ratios. Now I’m open to new facts if they are presented to me.
I certainly do understand that the conditions have to be right for combustion; and I don’t claim to be able to explain why there have (never?) not been any accidents associated with this. But when the filler cap is off or the sending unit out there is certainly an opportunity for air to mix with the fumes in the tank, in what ratio this would be is anybody’s guess.
And please let me make it clear as I stated in the initial post that the only time that your going to create a spark is when the circuit is broken and that would only happen in a case like I found in the sender that I just serviced. Or, if for some reason the spring tension allowed the wiper to lift from the element.
Nifty, yes my friend, there is current running through a completed circuit, break the circuit at any point and one side will be positive and the other negative.
And Bill I’m trying to dissect what you said about the gauge passing current to ground at the gauge cluster?? This is a simple series circuit with two resistance components in it, one of them (the sender) being variable. Any current that passes thru the circuit passes thru both and arrives at ground potential through a ground lead from the sender bracket to the body or frame. With out measuring it I have no idea what the resistance of the meter is nor how much current is allowed to pass thru the circuit, I suspect that your correct in your guess that it is miniscule, or probably in the milliamp range. At 6v this will spark, maybe not a long spark but it will spark.
Denny Graham
Sandwich, IL

Denny Graham,

You need to 'dissect' the inner workings of the fuel gauge and you will understand what Bill was referring too. This system doesn't just pass current to the variable ground (sending unit), it compares grounds...one inside the gauge, and one inside the tank, and positions the gauge accordingly...but you probably already knew that!

Stuart

I've seen the same deal on many makes and models having the sender rheostat inside the tank in the fumes & sometimes covered in fuel. Never seen an older vehicle with an enclosed unit, pretty much all open framed like what you have there.
I've also pulled them out & seen missing windings, burnt surfaces, even pulled one out where the wire was melted away.
One would think that would cause ignition, but have yet to see a fire started due to it. More often see cars burn to the ground because of poor fuel line connections or old rotting fuel lines leaking into the engine bay.

I,m thinking they use a material that melts below the flash point of the gas.

This talk of the "ideal" ratio for combustion reminds me of a program I watched on one of the satellite chanels (TLC?, Discovery?,) that showed a group of oil company emergency responders learning how to deal with the natural gas or propane pipeline rupture. They were standing about 100 yards away from a pipeline spewing propane into the air. Each person had a flare gun and took turns trying to hit the sweet spot in the plume to set it on fire. The very point being discussed here was discussed there, about the air to fuel ratio needing to be correct for combustion to take place. There were several burning flares that went directly through the plume without lighting it on fire. I don't recall them giving an exact air to fuel ratio required for ignition. Undoubtedly there are sparks inside the tank, especially when using an in-tank electric pump, but the level of oxygen is too low to support combustion. I suppose if your point is there is a chance an explosion could occur you're right. There's also a chance you could win the lottery and get hit by a meteor too. Not worth fretting over IMHO.

There's also this fun little fact: Way back when my father was known to weld closed leaks on diesel fuel tanks, while they were full. Sounds idiotic, but top off the tank to minimise fume buildup, seal the filler port, and keep the welds short. Works good, lasts long time, however retarded it may sound.

I have often wondered if the red tag on the gauge unit is more to prevent a disaster then just blowing the sender. I have thought about these things myself and wondered with all the millions of cars and trucks why I have not heard of it exploding.

Has anyone else heard of the demonstration where the guy lights a match and then snuffs it out by dipping it into an open container of gas?

I DO NOT ADVISE ANYONE TO TRY THIS!

In theory though it can work.

Let us know what your research tells you on the match in the open gas container thingy.

Think you got the right idea Tiny. Just talking what if here. Just like flight 800, it takes a freak set of circumstances. But I can still see the potential there.
I couldn’t tell ya how may hundreds of gallons of regular I went thru washing parts when I was a kid. Static or steel to steel sparking never crossed my mind. I worked in a Sinclair gas station for a while and we had the typical five-gallon bucket of gas for washing parts in the corner. The mechanic smoked Luckys and he used to flip his butts into the gas bucket to put them out, he got a big kick out of watching everyone pucker up.

Yahoo AgriHawg, why don’t ya test that theory, you can let us know from the burn ward how it came out. Now Steve, don't ya go and "triple dog dare" him, ya here!
As far as the match trick, that open flame is a REAL accident waiting to happen, that's one I'd really hate to be around. Pass an open flame through the fumes on the surface of the gas and I would bet my SS check you would have an explosion. The cig butt embers taint quite hot enough.

“WARNING if hot wire touches, sender will be damaged” That one?
Fred, you got the answer to that one, E/R=I. If ya touch the red tag with the hot lead with the tank full ya only have a 2 ohm wire wound in the circuit, that’s gonna have to pass around 3amps, so I’m sure you’d smoke the resistance.

Stu and Bill I don't have a spare fuel meter to dissect so I can't make an educated comment on that. It would just seem logical to me that the gauge is just a standard milliammeter with a fixed resistance. I’ve not seen anything to the contrary, if you have any more documentation about them I’d sure like to add it to my reference library. .

That’s so true, about fires under the hood. Every single carb that I have sitting on the shelf (approaching a dozen now) has signs of fire inside, that’s the only way to get soot in them. I personally have had fire from a blown fuel pump diaphragm on a ’60, 235, and I was in the middle of a Nevada desert when it blew.
Fun discussion guys!
Denny Graham
Sandwich, IL

Denny Graham,

I have enclosed a copy & paste description of how the fuel gauge functions. Stu and Bill have studied it and know it to be true, as well as the engineer in Detroit who designed it.

After you study it, see if it seems logical to you, then let the engineer in Detroit know you have put the "Sandwich Stamp of Approval" on it so he will rest easy tonight.


"The fuel gauges in our old trucks use two coils (electromagnets) to position the needle.

One coil (called the limiting coil) pulls the needle to the left or empty position. The other coil (called the operating coil) is positioned at 90 degrees to the limiting coil and pulls the needle towards the full side.

Current from the ignition switch, enters the limiting coil when the switch is in the 'on' position. The other end of the limiting coil is connected to both the operating coil and the tank sending unit. The other end of the operating coil is connected to ground.

If the gas tank is empty, then the resistance of the sending unit is near zero ohms and all of the current from the limiting coil goes to ground through the sending unit. With all of the current flowing through the limiting (empty side) coil then needle is pulled to that side.

If the gas tank is full, then the sending unit has about 30 ohms of resistance and most of the current from the limiting coil will pass through the control coil to ground. This current will cause the needle to be pulled away from the limiting coil (empty side) and towards the operating coil (full side). The operating coil has more windings than the limiting coil so it takes less current to generate enough magnetic force to overcome the limiting coil and pull the needle to the full side."

Stuart

Tks Stu, printed it out and gona go to the throne and do a case study of it. Let ya know if it passes scruteny, or rather if it makes any sence to me.
Denny G

The fuel itself doesn't ignite at lower temps but the vapors, it's that whole flash point thing that gets you. A cigarette is more than hot enough to ignite gasoline vapors, but in liquid form, not a chance.

You can put a match out in gasoline, but you have to allow it to stabilize and vent off any heavy vapors. Not a reccomended practice, but entirely possible.

At my first job we kept small one lb coffee cans about half full of gas right there on the workbench, & used it to refill carburetors/check accel pump operation after being rebuilt. One day a fella watching us build his carb asked Dan about smoking around all that gas, Dan says "gas don't burn" & promptly tossed his lit cigarette in there(went right out). You shoulda seen the look on that guy's face, it was priceless, he also never came back behind the counter again.

Gasoline/air doesn't have to be an exact ratio to ignite. The lower flammable limit is about 1% (by volume) in air. The upper flammable limit is just over 7% (by volume) in air. This means any mixture above or below these limits won't ignite. It also means any mixture between these limits can ignite if an adequate source of ignition is pesent.

true.

i've seen PGE (gas company) mig weld a valve on a 7'' natural gas pipe line w/ gas inside it
NO explosion
cause the ratio is wrong and in this case the pipeline was high pressure ie 175psi
a cross country pipe line

Thank ya Harold, I know your schooled in this area. I still taint had a chance to digest your info Stu, but I will get to it buddy.
Denny G

Some of you that have gotten to know me, (I didn’t say like me) since I joined the forum a little less than a year ago, know that I have this insatiable curiosity, this “need to know everything” that crosses my path. I’m part Scotsman and part Irishman, one part means that I’m a real cheap skate and the other part means that I’m a real bullheaded. Usually the two halves manage to work together to find a way of fixing things with out parting with the green, of course sometimes they work against each other and the bullhead ends spending more of the Scotsman’s money than he should.
Anyway that’s why I just have to fix up these parts as much as I can on my own rather than simply replace them with a rebuilt or reproduction. A positive side affect is that I learn a lot. I’m already 65 and I’m only scratching the surface, so much to learn!

Stu, I got a chance to read your excerpt and I see what you guys are referencing to. I've never having had the opportunity to actually do an autopsy on an AD fuel gauge so I really don't know what kind of a movement it has in it. I was assuming that they were simply an inexpensive D’Arsonval or Permanent Magnet Moving Coil movement using an internal shunt.
I’ve also down loaded a few other articles one of which is by Randy Rundle describing the movement used in fuel gauges. From his discription this meter movement cannot quite be considered a true electrodynamic meter movement but I think it’s in the same family, I’m still looking for that answer.
By the way, what was that Detroit engineer’s name???
Denny Graham
Sandwich, IL

Denny, As far as I can determine, there were two types of sending units used on AD trucks. The earlier style, ( I believe '49 and '50 ) had a gear arrangement on the float arm with the electricals mounted in a housing externally on the unit. The gears transferred the up and down movement of the float arm into a rotation of a rod that ran to the actual guts in the housing. Kinda hard to explain, I've got an old one around here somewhere. If I find it, I'll send a pic.
In '51, they went to a submerged type, such as what you have, only it's all metal construction.
I believe they changed because of a propensity of the gears to corrode and freeze up on the earlier style, plus they were probably cheaper to make.

Bob

Boy I'd really like to see that Bob. I've been hoping we would get some pictures that we could add to our reference libraries. I conceder all of this as continuing education so keep it coming guys.
Yep, this sender couldn’t get much more basic than it is. It would be pretty inexpensive to produce.
Denny Graham
Sandwich, IL

agrihawg/ builder The match in the container does work. BUT and i mean BUT .
THE CONTAINER MUST BE FULL TO THE BRIM/ TOP.
NO VAPER LAYER
I have done this and it works. BUT USE A VERY SMALL CONTAINER 1 OZ [ NO SPILLAGE] FULL TO TOP.
AND AWAY FROM ALL OTHER GAS. AND DO NOT FUMBLE THE THE MATCH

Here you go Denny,

1948 and 1955.2 fuel sender units

Looks to be real similar to the sending unit in my 38. Sorry didn't take any pics while I had it apart.

<click one>
<click two>
<click three>

I had an 98 Chevy that would "run not run" for about a month. I finally pulled the tank to replace the fuel pump. One of the leads was completely burned in two. The end of the wire was black and burnt and there was melted black burnt plastic on the pump. I Never drive a vehicle below a 1/4 tank which means this burning took place while submerged. If it had been empty would I have been speaking to JESUS?

An important point I want to make ( been said here but not clearly). Liquid gasoline does not burn, what you see burning is the vapor part as the gasoline vaporizes. The vapor mixes with the oxygen in order to support combustion. There is no oxygen in the liquid itself. Once the vapor is lit it may appear that the liquid is burning but in reality what is happening is the heat from the reaction is causing a much faster vaporization of the fuel and therefore a bigger fire. The reason the container must be full for the match to fail to light it is because it gives the vapor nowhere to collect. But if its warm enough outside when you do this, enough vapor will be forming off the top to light it anyway. So PLEASE don't burn your head off PEOPLE!!!!!!!!
The idea that there is no oxygen present in a fuel tank is false. We all should know by now that a tank will draw air in through the vented cap as we burn fuel driving. Anybody who has every had a bad vent on a fuel cap (older vehicles) knows that familar "sucking sound" as you take the cap off. And those vapors will burn in a low oxygen enviroment ( they just won't explode).
So I say YES denny I agree with you, if for some reason you had a spark inside a tank with a low level of gas there would be some statistical chance. I also know how sending units work they read resistance difference between a known ground and the one on the unit at a millivolt level. ( try and make one spark I bet you can't )

OK I'll shut up now.

P.S. Anybody know a good site to buy a frame tank kit from. For some reason I am very leary off my cab tank now. LOL

The other reason a match or cigarette doesn't ignite a full container of gasoline is that the liquid has enough heat capacity to extinguish the small flame by cooling it without raising the gasoline above the ignition pont.

Most of the information you will find on flammable liquids will use flammable limits and explosive limits interchangably.

It's not that there's no oxygen in the tank space that prevent's an ignition it's that there's not enough. For gasoline with a UFL/UEL limit of 7% (by volume in air) to reach the UFL/UEL in the tank would require that the vapor space in the tank be 93% air. This condition is for ignition whether you call it burning or explosion it's still the same. That's the minimum amount of air needed and that's a a lot of air.

The vapor space in a partially filled gasoline tank with a typical Reid Vapor Pressure of 9.5 PSI does not become flammable until the temperature has dropped to less than 10 deg F and does not become a serious hazard until it's between 0 deg F and -40 deg F.

On a truck I used to have the PO had replaced the stock fuel gauge (which had burned out) with a gauge from something else, and used a 10ohm 10 watt resistor shunted across the battery and sender terminals to 'equalize' the gauge to read correctly when used on the stock sender. The gauge was erratic, and read FULL until it was under 1/2 tank... but it worked and I never messed with it.

Now regarding passing high currents to the sending unit, in retrospect I wonder how safe this arrangement really was?

From my days in the service, you have to remember the fire triangle, heat - oxygen - combustible material, each making up a leg of the triangle. Without all three there is no fire. Remove one or more and there is no fire.
So all said from this thread while there is a combustible material (gas) there is not enough oxygen or heat therefore no fire.

I have two gauges one new from carter and the one from the truck both test no resistance from connection to connection, and the gauge pegs full all of the time, would you think that the new sending unit is faulty? this is a 55 gmc ist series

pgossett,

When you say "no resistance", you are telling us that you have continuity..there is a good path between both terminals that offers no resistance.

I actually don't know what a meter would indicate connected between power and sending unit posts, on the bench with no other wiring!

In the dash, in the truck, with power applied to the power terminal, and the sender disconnected, the gauge SHOULD read FULL all the time, this is good!

If you do the same drill, but connect the sender wire and it still reads full, either the tank IS full, the wire back to the tank is broken, or the sending unit/tank isn't grounded properly.

I really don't know what values to expect just testing the gauge with a meter out of the truck...sorry.

Stuart