Packard Forum Index » Pre-War (1899-1942)

The gas gauge in my 37 120 coupe convertible jumps irratically from full to empty when traveling. I thought I found a bad connection where the wiring harness attaches to the wire to the tank transmitter. I repaired that but it does not seem to correct the problem.

When I disconnect the wire to the tank transmitter the gauge reads full when the ignition switch is in the "on" position and it does not fluctuate while driving the car. This seems to be exactly opposite from what I have read in previous forum discussions on the fuel system gauge. I believe the discussions have said that the gauge will read "full" when the tank lead is grounded out, not open ended? It seems odd that with no electrical flow, the gauge would go from empty (ignition switch off) to full (ignition switch on) when the wire is open ended.

Although I do think the gauge is reading inaccurately, it does appear to move from full to empty as the fuel is consumed.

Has anyone else experienced this problem. I would like to have a better understanding of the system before I end up dropping the tank.

Thanks for your help,

Rod

According to the eleventh edition of Motors Manual, 35-37 120's use AC gauges. Those are magnetic type and will read full if wire is disconnected. Some of the later cars used King-Seeley which are thermal and those will read empty if disconnected.

While the senders are different in construction, the symptoms sound exactly the same as later ones exhibit when in spots, the resistance winding in the sender has a poor connection with the wiper. On those units the winding is on a strip of phenolic or fiber material which can twist or warp. There could be some oxidation buildup also. An intermittent or poor ground between the sender and tank or body also contribute.

I don't know the resistance range on that particular sender but since it's not the bimetal type, should be able to be checked with an ohmmeter. Would expect a continuous smooth change over the travel. Ones that have had issues seem to jump several numbers high or low before coming back to the proper spot.

EDIT: Since they are supposed to be the same, here is the description out of the 35 120 technical data manual.

Attach file:

---

  (49.37 KB)


  (105.40 KB)


  (111.90 KB)

Thanks Howard! Your description and attachments really help. I now understand how the system works which will help in troubleshooting my problems.

From what I have read, my system must have a poor ground or the float unit is defective. I will insure that I have a good ground before removing the unit.

thanks again,

Rod

Hope it's an easy and inexpensive fix. If you do take the unit out to check and measure the resistance, perhaps you could post the values for empty end and full end.

I checked my 38 sender out and I came to the conclusion that it is a 0-90 ohm unit.

I bought a 1940 120 back when I lived in Dallas. While showing off the car to Jim Hollingsworth, we out for a drive late one evening in south Dallas. All went well until the car went silent as the last drop of gas was consumed, while the gauge continued to read 1/2 full. Jim suggested that I check the ground to the tank. I ran a wire to the frame and attached it with a alligator clip, then later a toothed washer and screw, and did the same on the lip of the gas tank. Viola - a working gas gauge! It continued to work fine for the next six years or so when I sold it. Don't know if this is the problem, but it's a quick and easy check. Good luck.

Thanks David for the advise. That is exactly what I had planned for my next move before I drop the tank. The gauge seems to work fine except while the car is in motion so I am presuming it is a ground problem or the sending unit.

Rod

Yup, it was a ground problem! I drilled a hole in the seam of the gas tank and ran a lead over to the frame and my problems went away.

Thanks for all the help,

Rod

Glad my cheap trick worked. Amazing what a little grounding can accomplish.