Oh mighty Hot-Rod collective, I recently ran out of gas, in my 53 Cadillac. No surprise, as my gauge was sloppy. As the gas would slosh in the tank, the gauge would float up and down. So I replaced the sending unit, with a new one. I also re-terminated the wire the sending unit as it was frayed and unhealthy looking. It seems to do the same thing, just not as bad. Is there somewhere else I should look? My car buddy suggests a resistor that may be bad or the gauge itself. Does anyone have any insight?
Take the sending unit out, ground it, and power up the system. Move the sender full travel, and watch the gauge. If it follows it, all should be well, as it ( the sender), should travel from zero to whatever ohm rating it is. Cycle it fast, and if it follows fast, you are probably seeing just what you said, the "gas sloshing", as there's probably no baffle near the sender. While the sender is out, look inside the opening with a mirror and a flashlight, to check for baffling. Most movements are "dampened", best I recall, so you don't see too rapid of movement on the gauge, and I would hazard a guess, that if a bad resistor was involved somewhere, you would have no reading, ( tank empty), or a full reading.
In 53' you should not have an external resistor associated with the gauge. Most often, the sender is at fault due to the environment it is exposed to. The gauge can have connection issues at several points, but generally, if it is working then it should be fine. When you say "sloppy", what are you observing? Does the gauge pointer bounce around? Is the calibration off? Also, in 53' you should be 6 volt. Have you converted to 12 volts?
'53 was the first year of 12 volt. The gauge pointer is bouncing, and as I said it is a new sending unit. The gauge seems fine, I wonder if I can put some sort of resister in-line to the gauge to slow its response time. The tank is not baffled.
Is there any way to dampen the gauge movement? I've just been gauging it on how low the needle gets, during acceleration, but I would like to make it a little less "ghetto".
Doing some reading, it looks like this should be possible, and not all that difficult. I haven't tried it yet, but my tank has no baffles, so the gas sloshes around, and the gauge is all over the place. I intend to try this later this week. The basic gas gauge is a calibrated volt meter in the dash hooked to a variable resistor in the tank. Like: To dampen the movement of the gauge, add a capacitor and a resistor, like: Depending on the resistor and capacitor values chosen, it will now take some time for the gauge to initially start up and read correctly, but it will also take some time for the gauge to react to the changing fuel sender as gas sloshes around. The values shown should, if I did the math correctly, take about 10 seconds after power on to reach a steady state where the gauge reads correctly. There is a small voltage drop across the cap, but I don't think that will be a problem for this application.
You re-terminated the wire as in you are still using the old wire? You may try a new piece of wire it could be shorting out anywhere between the tank and the gauge.
Good luck finding a radio shack. All the ones in my area are gone. I end up buying my resistors and capacitors online. No other place to get them. Maybe hobby lobby.
Being that the ohm ratings (resistance) change with the movement of the sender, adding more resistance would make the gauge inaccurate, no?
A slight drop, yes, but not very much of one once the cap charges up. But, I drew the schematic wrong. It should have been like this: This is a resistor capacitor (RC) network. There are a couple of things you can calculate about it. The first is T, the time constant. T = R x C So with R = 2.2K and C = 100uF, T = 2.2 seconds. An RC network reaches a steady state at 5 x T, for this circuit that is 11 seconds. At that point the cap is fully charged and acts like an open circuit. There is a slight voltage drop, though. To calculate that, you start with V (12 volts, here). To get Vc, the effective voltage after the RC network, is: Vc = V (1 - e ^ -t / RC) e is a constant, approximately 2.71828 t is the number of seconds R is the resistor value C is the capacitor value So, for R = 2.2K and C = 100uF, with t = 5 seconds, Vc is 10.76 volts. The gas gauge will read a bit lower than it would without the RC network. But at t = 10 seconds (almost to the 5T value above), Vc = 11.87 volts. For a circuit that runs 12v, with a bunch of wire, and some connectors, the difference between the theoretical perfect 12v and a voltage drop to 11.9v is probably insignificant. As the gas float drops, the voltage level in the circuit changes (could be up, or down, depending on the gauge and sender design). The cap will resist this change by absorbing voltage, or by adding voltage to the circuit. The resistor slows down how fast that happens. By varying the resistor and cap values, you can change T, and vary how quickly Vc stabilizes at almost V. I haven't tried this in my own car yet, but if you google search for anti slosh damper circuits, this is what you'll find. There's one site I found selling what appears to be exactly this, with a "plug in" harness (ie: crimp on bullet connectors) for European sports cars. He's getting $50 for them.
The Jalopy Journal
