6 minutes ago, SteveJ said:

Wattage is proportional to current. It is also proportional to voltage.

I'm sorry Steve, I've never been able to really understand this. So, keeping all other variables the same, would the thermistor illuminate a lower watt bulb sooner than a higher watt bulb?

The thermistor has to have enough current flowing thru it to heat up. To low a wattage (high resistance bulb)and it will not heat up, how low wattage is to low is something the engineers can answer if they had the specs.

Here is a good example, if you read the voltage at the light socket with a reg VOM or DVM you will likely see 12v, as there is no load on the circuit. Put a load on it like say a test light and the voltage will drop, not zero but some, bigger load bigger drop, as the voltage drops at the socket, more current is flowing thru the thermistor, submerged I doubt it would ever heat enough to change it resistance (lower). its a very analog setup so you may or may not see the bulb of a test light come on.

1 hour ago, dmorales-bello said:

I'm sorry Steve, I've never been able to really understand this. So, keeping all other variables the same, would the thermistor illuminate a lower watt bulb sooner than a higher watt bulb?

I want to hear Steve's explanation about the whole balancing act and competing factors here... 

But in any event, it sounds like you DO have the correct bulb in there, so I think we're back to the thermistor.

I just completed the experiment with the 4.3 watt and the 6 watt bulbs and the malfunction was triggered after 25 minutes or so in both cases. The gas tank is still over half full.  So (as the Capn' pointed out) it looks like we're back to the thermistor.?

Just thinking "out loud" here: would some kind of resistance placed inline just prior to the lead reaching the bulb be a possible fix? Granted, it would have to be calibrated precisely to allow enough current to reach the bulb once the fuel level dropped and the thermistor was above it.

5 minutes ago, dmorales-bello said:

Just thinking "out loud" here: would some kind of resistance placed inline just prior to the lead reaching the bulb be a possible fix? Granted, it would have to be calibrated precisely to allow enough current to reach the bulb once the fuel level dropped and the thermistor was above it.

That thought popped in to my head too.  Maybe use the fuel tweaker trick on your fuel light.  A potientiometer (isn't it actually a rheostat, as it's used?) that you can tune until you get what you want.

http://www.atlanticz.ca/zclub/techtips/tempsensorpot/index.html

@Captain Obvious, do you mean like this?

So it seems like the thermistor isn't spec'd correctly from the experiments run by @dmorales-bello. As @Dave WM said, a lower wattage bulb would be higher resistance in the line. In theory you could use a potentiometer to adjust the resistance, but it would be cheaper just to try a lower wattage bulb to see if that's a viable workaround.

2 minutes ago, Zed Head said:

That thought popped in to my head too.  Maybe use the fuel tweaker trick on your fuel light.  A potientiometer (isn't it actually a rheostat, as it's used?) that you can tune until you get what you want.

http://www.atlanticz.ca/zclub/techtips/tempsensorpot/index.html

A pot runs a few dollars (plus shipping since Radio Shack is no more). Low wattage BA9 bulbs were pretty cheap at the link I posted. You would need no more than a 100 Ohm rating on the pot, and then you have to modify the wiring to incorporate it.

you would prob want to get a WW pot, those cheap carbon track pots would not like the current. I think I have some 100 or 500 ohms ones laying around if you cant get the fix with the different bulbs.

Fry's is like a super duper Radio Shack.  Georgia is close.  https://www.frys.com/ac/storeinfo/storelocator

I'll bet that zcardepot used a battery to determine which thermistor would work, and didn't consider alternator voltage.

Seems like at this point it needs to be determined if you're trying to find a fix that you can live with or trying to show that ZCarDepot used the wrong thermistor so that you can get a refund.  Or go even farther and determine the right thermistor to use for the application.

A cheap "pot" would be the easy experiment if you just want to know the answer.

 

The Fry's closest to me (about 3 miles from my office) was almost an empty warehouse the last time I visited. Are they back to stocking products?

its not a stock thermistor? oh well that could explain things.

 

11 minutes ago, Dave WM said:

its not a stock thermistor? oh well that could explain things.

 

It wouldn't surprise me if it's an "off-the-shelf" piece that was used in creating the reproduction sending unit. And by "off-the-shelf" I mean that it is designed for a more recent car/electrical system, maybe even for LED lights. That would explain why the light comes on so soon for the incandescent bulbs. Unless ZCarDepot has someone as thorough as the good Captain when specifying the behavior for the reproduction, a thermistor with the wrong curve could have been used.