Faulty fuel senders

[CarDrake]

I have installed a brand new fuel sender on my 40 Super and it is wildly inaccurate; registers about 1/8 of a tank with 6 gals. (in a 17 gal. tank.)

 

My Buick Guru told me that he installed 2 senders and both gave him bad readings. He said that he even checked it before he installed it, noting a full gas gauge with the float raised to it's highest position, zero upon dropping it down to the bottom.

 

Are these senders that are available just crap? It sounds like faulty (i.e., leaking) floats might be the culprit.

[neil morse]

I know that "all" available senders are not crap because I got one from Bob's Automobilia that's working fine in my '41 Super. 

 

Your post is a little unclear -- did you install this sender or your "Buick Guru?"  Are you saying that he installed two different senders in your car that gave bad readings even after checking the operation before installation?  It would help to know what kind of "bad readings" they were giving.  With regard to the one currently in the car, apart from giving an inaccurate reading with 6 gallons in the tank, what other "wildly inaccurate" readings is it giving?  A more detailed description of the symptoms would help diagnose the problem.

 

With some senders, you need to bend the arm a bit to calibrate it to fit the size of the tank (i.e., the distance of the "drop" from full to empty).  That seems to be a more likely explanation for the symptom you describe than a leaky float.  If the float is leaking, the gauge will just always read empty because the float doesn't "float" and remains at the bottom of the tank.  It also seems unlikely that a float would be leaking on a brand new sending unit.

[Ben Bruce aka First Born]

Ground problem.  I run a ground wire from one of the sender mounting screws to the frame.   Dollar to a donut that will make you a happy camper.

 

 Ben

[Gary W]

1 hour ago, CarDrake said:

My Buick Guru told me that he installed 2 senders and both gave him bad readings. He said that he even checked it before he installed it, noting a full gas gauge with the float raised to it's highest position, zero upon dropping it down to the bottom.

 

If you checked it before installation, and it gave proper readings when not in the tank.....

 Raising the arm gave a "F" on the dash gauge,  and should measure 30 OHMS with an ohm meter

 Dropping the arm shows an "E" on the dash,  and should measure 0 OHMS with an ohm meter.....then it seems the actual sending unit is working properly.  

 

In a 17 gallon tank, assuming you want a reserve, your dash gauge should measure approximately a quarter tank for every 4 gallons of gasoline.

 

I had just the opposite problem...  My dash showed 1/4 but the tank was actually BONE DRY.  I was able to fix it by grinding on the "stop" of the sender AND bending the arm up.

 

In your case, you would have to bend the arm downward, (so the float hits the gasoline "earlier") which will effectively "raise" the arm,  giving a more accurate reading at the dash. 

(** If your dash unit shows 1/8 with 6 gallons of fuel in the tank, does it also show 3/4 when your tank is totally filled? **)

 

This assumes, of course, that:

1. Your new sending unit is operating in the proper 0 - 30 - 0  OHM range.  (empty - full - empty) (float arm dropped - raised - dropped)

2. The sender is properly grounded

3. Your dash gauge is functioning properly

 

I recently went through all of this, and it is a very detailed discussion.  It starts with me running out of gas, and continues on.  If you have a little time to go through the discussion, it will be very valuable.  

 

Touch that arrow in the top right corner to begin the discussion and the steps to take to check your sender and hopefully get it reading correctly:

 

[neil morse]

What Gary said.

 

Ben, I think a bad ground will just cause the gauge to read "full" all the time.

[Bloo]

This system uses 0 ohms for Empty, and so there is no tolerance or fudge factor anywhere. There are a bunch of things to check while the tank is out of the car.

 

1) The float should not hit the bottom or the top of the tank (but almost). If it hits it is probably going to eventually sink.

 

2) It needs to swing from 0 ohms empty to 30 ohms full. If there is not 30 ohms difference, the needle cannot swing the whole way. A little over 30 is ok. If it is way over 30, the gauge will be wildly inaccurate, staying pegged full for the first few gallons. Check an empty tank by checking right side up (0 ohms) and upside down (30 ohms)

 

3) The float needs to be far enough from the bottom that there is an appropriate amount of reserve. Put some gas in the tank, enough to bring it above 0 ohms, then suck the gas back out with a hand powered transfer pump until you get to 0 ohms just barely. The original senders have a brake, and if you have one of those you need to keep shaking the tank while you do this. Once you are at 0 ohms, transfer the hose to a different container and keep pumping until no more gas comes out. Check how much gas is in your second container. This is your reserve. If you suck air before you get to 0 ohms, you will run out of gas before you get to "E". If the reserve is there but is not an appropriate amount (too little or too big), now you know. Either way you can fix it while the tank is still out. Don't use any kind of electric pump. Also avoid the Harbor Freight red-colored plunger pump, it's crap. Any other hand powered transfer pump should do.

 

4) Always add a ground to the tank, Run it from the frame or body to a sender mounting screw.

 

5) The gauge needs a good ground to work as well. The instrument cluster can have a bad ground due to fresh dash paint. If you have the gauge out of the cluster, and you look at it closely, you can see where it grounds to the cluster.

 

6) With the key on, if you disconnect the tank wire and ground it, and the gauge goes to "E", and you let the wire hang loose and the gauge pegs full, there is a 99 percent chance that the gauge and wiring are ok, and only the sender and/or sender ground need attention.

 

7) I recommend adding a sock filter to the fuel pickup if the sender does not have one.

 

Good luck!

 

[CarDrake]

Thanks to all, including Gary W, Bloo and all the others who chimed in; I never expected such a lengthy and informative discussion!

 

I had to drop the tank and remove the sender to do work (unlike the service manual states.) I then checked the resistance and found I had 32.2 ohms when "full", and .9 ohms when "empty." Since this appeared to be within tolerance, I was happy with the readings: I think that (as Bloo stated), there is very little I could do to get the upper number any closer (as you need to grind off the stop, thus actually increasing the resistance.)The lower number was certainly within range.

 

With the tank laying level on a table, I Used a straight edge to eyeball the distance between the bottom of the tank and the flange for the sender. It measured about 7 3/4" Then, I took a measurement from the bottom flange of the sender itself and the lowest point of the float, and found it to be about 5 1/2." I then bent the float arm down to about 7 1/2" and reinserted the sender into the tank,

then checked the resistance readings again (tank level, then upside down, as mentioned.)

 

After reinstalling the tank, I poured in between 5 and 5 1/2 gallons, which put the fuel gauge right on the 1/4 full mark; this should reflect the 17 Gal capacity; first mark at about 5-5 1/2 (4 Gal plus 1-1 1/2 Gals. reserve.)

BTW, I'm keeping my fingers crossed about the high end reading!

 

Hopefully, I have solved the problem. However, IMO, the vendors of these fuel senders should caution the prospective buyers that they will almost probably have to titrate (i,e., bend) the float arm to suit their particular application.

Thanks again to all,

Charlie Drake, Potomac, Md.,1940 Buick Super

 

 

 

 

[Bloo]

It is a hazard of this particular design. There is no way to have a calibration adjustment anywhere because E is 0 ohms. I still marvel that GM used it in production for so many years. It must have been a real PITA for them, as it probably caused a lot of rework to be necessary on the line.

 

Glad you got it worked out.

 

Edited September 13, 2021 by Bloo (see edit history)

[Larry Schramm]

14 hours ago, Bloo said:

It is a hazard of this particular design. There is no way to have a calibration adjustment anywhere because E is 0 ohms. I still marvel that GM used it in production for so many years. It must have been a real PITA for them, as it probably caused a lot of rework to be necessary on the line.

 

Glad you got it worked out.

 

 

The first job that I had at GM was building fuel tank sending units on the line at AC Spark Plug.   Believe it or not, adjusting the float level was not that big of a deal.  We put the unit in a fixture and with a special set of pliars would bend the arm between the pivot point and stops on the resistor housing.  Then adjust the arm to get the correct travel for the tank height.  Maybe took 30 seconds.

 

Surprising, very few tank units needed much adjustment.