The Stovebolt Site Forums The Shop Area The Electrical Bay Different approach to fuel level sender...

I received the fuel level sender I ordered from China yesterday and it is interesting. It seems to be made of non magnetic stainless steel, welded and made nicely and comes with a gasket and screws. You can see some images below. As soon as possible I'll try it in the tank. A couple of observations: Just like the one I made of this same style the resistance change isn't really a smooth sweep. I tried to make it smooth however I could not incorporate enough reed switches nor enough resistors in the central tube and I suspect the man in China had the same limitation. It starts at 30 ohms (full) and then works its way to empty in stages (to 0 ohms at empty). It should register empty with still about 1.5 to 2 inches of fuel left in the tank so you won't just hit empty and be stranded. The reason this is so is that just like the senders used in today's fuel tanks magnets and magnetic reed switches are used to register levels. So, the upshot is I expect this sender to work well and to be very durable since the reed switches will be kept dry at all times. I expect the vertical travel of the float to be both predictable and free from any sort of obstructions/problems. And I think once I get used to the way the needle travels (in steps), it should be okay. The sender was $30 USD plus about $8 USD to ship plus sales tax.

edit...the one I got was 260mm in length

Did you try to determine how many "steps" (how many reed switches) are in the tube?

Very interested in seeing how it works for ya once installed.

Hi Bill,
I believe there are 11 steps which are (in ohms) 31 (full), 28.4, 25.1, 21.7, 18.3, 15 (half tank), 11.6, 8.3, 5, 2.7 and .4~.5 ohm (empty). I will have to dig out my notes, but when I constructed one of these I think I only ended up being able to fit 10 steps into the tube. Obviously if the same number of steps can be built into a shorter tube (which for the gentleman who had the 7 inch tall tank will be the case) the needle movement will be smoother. I don't know if that will be possible, though. It will require some very compact components. The wiring, the reed switches and the resistors take up a certain amount of space and at the end of the day you do run out of room inside the tube...or at least I did. However I do see a possible benefit to the (roughly) 3 ohm steps the unit takes: the needle may be less susceptible to movement from motion, turns, etc. and some may find that a positive. As you know common values of resistors don't always allow you to hit exact step-down values, but this is pretty close. The more I've fiddled with it, the more pleased I am. Somebody did some nice work here it seems and it could last a very long time without problems.

I have a 7" long one of those that I got from Tanks, Inc.

The version I ended up with only had 8 steps, which made for really chunky needle movement. I probably would have been ok with 9 steps (one for each 1/8 tank plus empty), but Ibwas not happy with the sweep and replaced it.

I agree that the fully sealed/magnetic design looks solid and should work great for someone who is of with the number of steps it has.

I'm afraid that is going to be the case with this design no matter who makes it. I should mention because of the design there is 9 inches of travel with this one...or roughly 230mm. With 30 ohms to play with and 11 steps you get 2.7 ohms per step (roughly) and that's pretty close to what they used. Not having the float go all the way to the top isn't a problem in my opinion. There are reasons you should not over-fill the tank of the AD.

As I said, more steps are better and getting used to the difference may be the best answer. It reminds me of my grandmother when she said "well, you just haven't learned to like liver yet, but I'm sure you will." The magnetic reed switches are made in different lengths and in my experience the most reliable ones are 14mm in length (ignoring the connecting wires). The smaller ones fail very often and when they do you get a "dead spot" so the sender won't work at all until the fuel level rises or falls to allow the next switch to be activated.

I know how to make one of these that will show stepless variable resistance (using Kanthal wire and a sliding contact which would be drawn up and down with a magnet) but the challenge is in finding a wire that is flexible enough to allow a U bend in the wire without causing the sliding contact to hang up in the process of movement. In my experience you just don't have enough room inside the tube to do it and the wire gauge of the sender must match that of the fuel gauge or the gauge "sees" the wrong resistance. For anyone thinking about this, it is just like the dimmer part of the headlight switch. That's simply a coil of resistance wire and a contact wiper which acts as a current resistor.

And around 4 years ago I used flat Kanthal wire to make some senders (both flat-type wipers and the C shape---a design I do not prefer) which provided a perfectly smooth and absolutely linear variable resistance output from 0 to 30 ohms however the process is time consuming and expensive so there's that to consider. I could not make one of those units for $30. I tried, though.

That's pretty neat looking. I like the straight up and down float instead of the lever arm.

Are the 5 holes clocked correctly? Isn't there one hole on the AD tank that's slightly off from a perfect pentagram?

Yes, they appear to be clocked correctly but I haven't had time to remove the existing one and try this one yet. I'll let you all know asap.

Edit...yes, they're the same and the gaskets are interchangeable.