Help with fuel guage PLEASE!
I had the same issue on my 1967 Dodge Dart, stock gauge, aftermarket sender.
Do you have the ground strap on the fuel sender to fuel line ( I assume it is still used in 75)
My aftermarket sender's pickup was almost an inch farther above the gas tank floor than the OEM unit, which would indicate the tank empty sooner than OEM, I saw a Chrysler TSB regarding fuel senders on a different car needing their pickup tubes bent to get it closer to the floor of the tank. might apply here as well.
I know this is not alot of help here but I feel your pain.
I also know the below info does not solve your issue but it does shed light on the linear vs non linear aspect of things.
I found an OEM NOS sender (04051004 latest replacement number from Mopar) Now my gauge reads much better. I Filled the tank full, then I attached a multimeter to the sender output and outlet tube on sender, then I siphoned out one gallon at a time shook the car and let it settle. Then I measured and recorded the resistance. I did this with the aftermarket sender installed then when the tank was empty (I could only get 17.5 gallons out. I have a dented area in the tank BUT I don't think it accounts for 1/2 gallon). I changed the sender to the OEM Mopar unit and added one gallon at a time and measured the resistance the same as before. ( yes the car was on a flat garage floor and it is not jacked up and sits very flat) The measurements are from the senders ONLY no chassis wiring involved. The first is the chart of resistances and the second is the graph of those values.
I have not worked out the resistance values for the dash gauge yet but if you take the 74 = empty, 23 = 1/2, and 10 = full the mopar unit comes MUCH closer to reading proper resistance values per gallon than the aftermarket.
I am building a resister bridge to chart the fuel gauge including the chassis wiring. I'm going to change the resistance by one ohm at a time, wait a length of time and take a photo of the gauge from a fixed location than using CAD determine the angle of the needle. (I do this type of data acquisition for a living)
Once I gather that I'll post my findings
I had the same issue on my 1967 Dodge Dart, stock gauge, aftermarket sender.
Do you have the ground strap on the fuel sender to fuel line ( I assume it is still used in 75)
My aftermarket sender's pickup was almost an inch farther above the gas tank floor than the OEM unit, which would indicate the tank empty sooner than OEM, I saw a Chrysler TSB regarding fuel senders on a different car needing their pickup tubes bent to get it closer to the floor of the tank. might apply here as well.
I know this is not alot of help here but I feel your pain.
I also know the below info does not solve your issue but it does shed light on the linear vs non linear aspect of things.
I found an OEM NOS sender (04051004 latest replacement number from Mopar) Now my gauge reads much better. I Filled the tank full, then I attached a multimeter to the sender output and outlet tube on sender, then I siphoned out one gallon at a time shook the car and let it settle. Then I measured and recorded the resistance. I did this with the aftermarket sender installed then when the tank was empty (I could only get 17.5 gallons out. I have a dented area in the tank BUT I don't think it accounts for 1/2 gallon). I changed the sender to the OEM Mopar unit and added one gallon at a time and measured the resistance the same as before. ( yes the car was on a flat garage floor and it is not jacked up and sits very flat) The measurements are from the senders ONLY no chassis wiring involved. The first is the chart of resistances and the second is the graph of those values.
I have not worked out the resistance values for the dash gauge yet but if you take the 74 = empty, 23 = 1/2, and 10 = full the mopar unit comes MUCH closer to reading proper resistance values per gallon than the aftermarket.
I am building a resister bridge to chart the fuel gauge including the chassis wiring. I'm going to change the resistance by one ohm at a time, wait a length of time and take a photo of the gauge from a fixed location than using CAD determine the angle of the needle. (I do this type of data acquisition for a living)
Once I gather that I'll post my findings