Needing a Resistance Divider Cct

Discussion in 'The Projects Forum' started by Johno8, Mar 14, 2009.

  1. Johno8

    Thread Starter New Member

    Mar 14, 2009
    8
    0
    Hi:
    I have a circuit for a motorcycle fuel gauge that needs adjusting. The fuel sender in the gas tank varies from 250 down to 35 ohms, depending on the fuel level. The gauge only reads resistance from 5 ohms to 100 ohms. I need to connect something that will change the resistance coming from the tank to match the specifications for the gauge input. Example...when empty of fuel, the tank resistance is 250 ohms...it should be 100 ohms. When it is full, the tank resistance is 35 ohms, should be 10 ohms. Can you help me with this? Thanks so much.
    Phil Johnson
     
  2. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    With the fact the sender is immersed in a flammable fluid, and had electricity running through it, perhaps it would be easier to change the circuit?
     
  3. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Thatoneguy's correct; there is a big safety issue involved. Sparks in a fuel tank must be avoided at all costs.

    With that said, the fuel gauge is putting out a very low constant current to the rheostat (fuel level sender) in the gas tank, and measuring the resultant voltage drop across the rheostat.

    For starters, it would help a great deal to know what that current is.

    Remove the wire from your rheostat/fuel level sender. Take a 100 Ohm resistor, connect one end to the wire that's supposed to go to the fuel level sensor, and the other end to chassis ground (or the negative terminal on the battery). Measure the voltage across the resistor using a digital multimeter. (this is much safer than attempting to measure current)
     
  4. Johno8

    Thread Starter New Member

    Mar 14, 2009
    8
    0
    This is why I wanted to change the circuit outside the tank. There wouldn't be any modification inside the tank...I am hoping to add the circuit in the wiring harness on the motorcycle instead. Do you still need me to measure the current? :confused:
     
  5. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    That's the idea. ;)
    Yep!
     
  6. Johno8

    Thread Starter New Member

    Mar 14, 2009
    8
    0
    Ok .15vdc With a 100 ohm resistor to ground. The voltage jumps from 0 to .15 vdc because the digital gauge flashes once per second when empty, so I suspect that is the reason for the flucuation. What now?

    Phil
     
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    OK.
    Ohm's Law says that Current = Voltage/Resistance, or I=E/R, so:
    0.15V/100 Ohms = 1.5mA

    But since it was flashing when you were taking the reading, that might affect the accuracy of your meter readings.

    Can you try it again with a lower value of resistance, say 75 Ohms, 68 Ohms, etc.?

    If you happen to have two 100 Ohm resistors, you could use two in parallel, which should come out to 50 Ohms.
     
  8. Johno8

    Thread Starter New Member

    Mar 14, 2009
    8
    0
    I'm using a Fluke digital meter, but I'll do the 50 ohm reading for you tomorrow...thanks.

    Phil
     
  9. Johno8

    Thread Starter New Member

    Mar 14, 2009
    8
    0
    Ok, using a 50ohm resistor, the fuel guage shows 3 bars, and the voltage drop across the resistor is +.14 vdc.
     
  10. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    1/100th volt is the difference between empty and full tank?
     
  11. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Yeah, that's a pretty slim margin.

    Set your meter to read in the 200mV range. Need to get more than 2 digits of accuracy.
    OK, let's try taking measurements with a few different resistors:
    10 Ohms
    75 Ohms
    82 Ohms
     
  12. Johno8

    Thread Starter New Member

    Mar 14, 2009
    8
    0
    I thought the same thing, so I triple checked and got the same readings. With 50 ohms, I get 3/7 bars while 100 ohms gives me an empty tank. If I actually connect the sender and manually move the float, the indicator varies from empty to 3/4 tank maximum as that sensor only goes down to 34 ohms. Looks like I might be better off looking for another sender with the proper resistance as this idea may be too much of a pain for everyone and I don't want to come across this way. Thoughts?
     
  13. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    Do you have a schematic of the electronic fuel gauge, or a make & model number?

    There may be a calibration adjustment on it somewhere, or a matching sender. The more info you can provide, the better the odds of finding the easiest solution are.
     
  14. Johno8

    Thread Starter New Member

    Mar 14, 2009
    8
    0
    It's an ACEWELL digital gauge and in conversation with them, the resistance levels are an industry standard for most cases exept Harleys, which is what I have. They suggested finding another sender unit, but motorcycle units are not plentiful. I even thought of modifying the resistance track on the sender, but wouldn't know where to start, which is why a resistance network seemed a possibility. You folks here are awesome in your expertise for sure.
     
  15. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Well, if you connected 167 Ohms to ground in parallel with the tank sensor, you'd wind up with a range of about 28 to 100 Ohms. You'd only get a 7/10 when the tank was full, but the empty part would be about right. This is the easiest and least expensive "quick fix"; certainly not ideal, but you would at least get an idea of how much fuel was in the tank.

    You could use a 120 Ohm and a 47 Ohm resistor in series to get 167 Ohms.
    Other possibilities:
    160 + 6.8 = 166.8 (-0.12 %)
    200 || 1000 = 166.667 (-0.2 %)
    180 || 2400 = 167.442 (0.265 %)
    220 || 680 = 166.222 (-0.466 %)
    150 + 18 = 168 (0.599 %)
    100 + 68 = 168 (0.599 %)
    130 + 36 = 166 (-0.599 %)
    110 + 56 = 166 (-0.599 %)
    91 + 75 = 166 (-0.599 %)
    240 || 560 = 168 (0.599 %)
    270 || 430 = 165.857 (-0.684 %)
    330 || 330 = 165 (-1.198 %)
    300 || 390 = 169.565 (1.536 %)
    160 + 0 = 160 (-4.192 %)

    Your fuel gauge is more than a fuel gauge, it's a multifunction unit, right? Those things have a microcontroller in them. I'll bet if you looked at the signal going to the fuel sensor, you'd see a ramp signal rather than a straight DC level.
     
  16. Johno8

    Thread Starter New Member

    Mar 14, 2009
    8
    0
    thanks so much for the info...will give it a try. A ramp voltage would also be the reason why the voltage jumps so much as well.
     
  17. gitbox

    New Member

    Apr 27, 2009
    2
    0
    I am doing the same thing as John08. I have a fuel sender that is not compatible with the fuel gauge. I'm sure there is some circuit that can be created to do the proper conversion.

    Here is some data that should help the design engineers on this forum:

    Fuel sender Meter current
    ========= ===========
    Empty 235 ohms 35mA
    Full 35 ohms 75mA

    Available supply: 13.8v

    Schematic of the stock fuel gauge circuit:
    +13.8-----meter-----fuel gauge-----ground.

    From my meter readings, it appears to be a linear relationship.

    I have a good supply of electronics parts and enough experience to build whatever circuit you come up with. (barring multi-layer PC boards and SMT components - I only do "hobbyist" circuits).

    Thanks so much for your help.
     
  18. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Sgt. you've got a little Adrian Monk in you. An astute observation which explains the seemingly minute DC voltage change over the range of the sensor!
     
  19. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
Loading...