Thermistor in a gas tank

Discussion in 'Misc Electronics' started by cxturbö, Jul 26, 2003.

  1. cxturbö

    cxturbö Guest

    I'm having problems with the "Fuel Reserve" light on my motorcycle.
    When the fuel level drops to 1.3 gallons the Fuel Reserve light on the
    dash board is supposed to light, few weeks ago it quit working.

    I removed the fuel gauge sending unit from the tank which has the
    gauge rheostat and a separate metal capsule for the Reserve sensor. I
    split open the capsule and a very small thermistor...maybe a 1/16"
    fell out. It broke off the two wires soldered at both ends.
    After looking over the schematic in my shop manual, the circuit seems
    rather disturbing. 12 VDC is supplied to the 4 watt bulb in the dash
    which then leads to the negative coefficient thermistor in the gas
    tank. Submerged in fuel the circuit is open. Out of fuel it heats
    up, closes and completes the circuit to ground.
    I always thought that there were very low voltage value's sent into a
    tank in order to prevent explosions? When that thermistor broke off
    the base wire could have easily touched the metal casing and it's at
    least able to handle 4 watts @ ~14 volts?
    Is the fuel to air ratio so high that an explosion is not possible?
    Even so, being a motorcycle with a 5 gallon tank I would think there
    would be a risk when removing the cap to refuel?

    Next problem is repairing it. I managed to solder the wires back on
    but that didn't last but a few weeks. This time the microscopic metal
    tab came off the black thermistor body. The bike is a very limited
    production 1983 Honda CX650 Turbo...needless to say the sending unit
    assembly is no longer available.
    After searching the net I managed to get some info from a Goldwing
    owner. He recommended a:

    NTC Thermistor TYPE RL2004-582-97-T10
    Rated at 1,000 ohms at 25 degrees Celsius.
    Cold resistance is between 900 and 1200 ohms.

    This evening I made a mockup circuit using that thermistor. With a 12
    volt power source and a 4 watt bulb I left it energized for a few
    hours. What has me worried is that the thermistor gets too hot to
    touch. Submerging it in a very small vile of gas causes the light to
    turn off. It's hot enough that I hear a very faint hiss as it goes in
    the gas.
    It works exactly as it should but I'm worried about the heat.

    Anyone have experience with these thermistors? I would assume the
    same style setup is used in most cars with "Low Fuel" warning lights?

    Thanks for reading,

    George
     
    cxturbö, Jul 26, 2003
    #1
    1. Advertising

  2. cxturbö

    CWatters Guest

    I think it would be safest to look for a sensor form another bike that uses
    a similar circuit but...

    If' I'm looking at the right data sheets ...

    http://www.thermometrics.com/assets/images/rl20.pdf

    the NTC material is D9.7A and the data for that suggests the resistance is
    on the high side..

    http://www.hhv.co.uk/PDF's/thermistors/keystoned97.pdf

    I made it 107 Ohms at around 85C.

    If that's correct I'm surprised the bulb lit up for your test.

    I think the cold temp is about right (could be a bit lower) but the graph
    needs to be steeper with a lower hot temp. I think the resistance of the NTC
    needs to be less than the resistance of the bulb at whatver temp you feel is
    safe (say <10Ohms at 50C?)



    "cxturbö" <> wrote in message
    news:...
    >
    > I'm having problems with the "Fuel Reserve" light on my motorcycle.
    > When the fuel level drops to 1.3 gallons the Fuel Reserve light on the
    > dash board is supposed to light, few weeks ago it quit working.
    >
    > I removed the fuel gauge sending unit from the tank which has the
    > gauge rheostat and a separate metal capsule for the Reserve sensor. I
    > split open the capsule and a very small thermistor...maybe a 1/16"
    > fell out. It broke off the two wires soldered at both ends.
    > After looking over the schematic in my shop manual, the circuit seems
    > rather disturbing. 12 VDC is supplied to the 4 watt bulb in the dash
    > which then leads to the negative coefficient thermistor in the gas
    > tank. Submerged in fuel the circuit is open. Out of fuel it heats
    > up, closes and completes the circuit to ground.
    > I always thought that there were very low voltage value's sent into a
    > tank in order to prevent explosions? When that thermistor broke off
    > the base wire could have easily touched the metal casing and it's at
    > least able to handle 4 watts @ ~14 volts?
    > Is the fuel to air ratio so high that an explosion is not possible?
    > Even so, being a motorcycle with a 5 gallon tank I would think there
    > would be a risk when removing the cap to refuel?
    >
    > Next problem is repairing it. I managed to solder the wires back on
    > but that didn't last but a few weeks. This time the microscopic metal
    > tab came off the black thermistor body. The bike is a very limited
    > production 1983 Honda CX650 Turbo...needless to say the sending unit
    > assembly is no longer available.
    > After searching the net I managed to get some info from a Goldwing
    > owner. He recommended a:
    >
    > NTC Thermistor TYPE RL2004-582-97-T10
    > Rated at 1,000 ohms at 25 degrees Celsius.
    > Cold resistance is between 900 and 1200 ohms.
    >
    > This evening I made a mockup circuit using that thermistor. With a 12
    > volt power source and a 4 watt bulb I left it energized for a few
    > hours. What has me worried is that the thermistor gets too hot to
    > touch. Submerging it in a very small vile of gas causes the light to
    > turn off. It's hot enough that I hear a very faint hiss as it goes in
    > the gas.
    > It works exactly as it should but I'm worried about the heat.
    >
    > Anyone have experience with these thermistors? I would assume the
    > same style setup is used in most cars with "Low Fuel" warning lights?
    >
    > Thanks for reading,
    >
    > George
    >
     
    CWatters, Jul 26, 2003
    #2
    1. Advertising

  3. cxturbö

    Joey Guest

    "CWatters" <> wrote in message news:<a7uUa.33865$-ops.be>...
    > I think it would be safest to look for a sensor form another bike that uses
    > a similar circuit but...
    >
    > If' I'm looking at the right data sheets ...
    >
    > http://www.thermometrics.com/assets/images/rl20.pdf
    >
    > the NTC material is D9.7A and the data for that suggests the resistance is
    > on the high side..
    >
    > http://www.hhv.co.uk/PDF's/thermistors/keystoned97.pdf
    >
    > I made it 107 Ohms at around 85C.
    >
    > If that's correct I'm surprised the bulb lit up for your test.
    >
    > I think the cold temp is about right (could be a bit lower) but the graph
    > needs to be steeper with a lower hot temp. I think the resistance of the NTC
    > needs to be less than the resistance of the bulb at whatver temp you feel is
    > safe (say <10Ohms at 50C?)
    >
    >
    >
    > "cxturbö" <> wrote in message
    > news:...
    > >
    > > I'm having problems with the "Fuel Reserve" light on my motorcycle.
    > > When the fuel level drops to 1.3 gallons the Fuel Reserve light on the
    > > dash board is supposed to light, few weeks ago it quit working.
    > >
    > > I removed the fuel gauge sending unit from the tank which has the
    > > gauge rheostat and a separate metal capsule for the Reserve sensor. I
    > > split open the capsule and a very small thermistor...maybe a 1/16"
    > > fell out. It broke off the two wires soldered at both ends.
    > > After looking over the schematic in my shop manual, the circuit seems
    > > rather disturbing. 12 VDC is supplied to the 4 watt bulb in the dash
    > > which then leads to the negative coefficient thermistor in the gas
    > > tank. Submerged in fuel the circuit is open. Out of fuel it heats
    > > up, closes and completes the circuit to ground.
    > > I always thought that there were very low voltage value's sent into a
    > > tank in order to prevent explosions? When that thermistor broke off
    > > the base wire could have easily touched the metal casing and it's at
    > > least able to handle 4 watts @ ~14 volts?
    > > Is the fuel to air ratio so high that an explosion is not possible?
    > > Even so, being a motorcycle with a 5 gallon tank I would think there
    > > would be a risk when removing the cap to refuel?
    > >
    > > Next problem is repairing it. I managed to solder the wires back on
    > > but that didn't last but a few weeks. This time the microscopic metal
    > > tab came off the black thermistor body. The bike is a very limited
    > > production 1983 Honda CX650 Turbo...needless to say the sending unit
    > > assembly is no longer available.
    > > After searching the net I managed to get some info from a Goldwing
    > > owner. He recommended a:
    > >
    > > NTC Thermistor TYPE RL2004-582-97-T10
    > > Rated at 1,000 ohms at 25 degrees Celsius.
    > > Cold resistance is between 900 and 1200 ohms.
    > >
    > > This evening I made a mockup circuit using that thermistor. With a 12
    > > volt power source and a 4 watt bulb I left it energized for a few
    > > hours. What has me worried is that the thermistor gets too hot to
    > > touch. Submerging it in a very small vile of gas causes the light to
    > > turn off. It's hot enough that I hear a very faint hiss as it goes in
    > > the gas.
    > > It works exactly as it should but I'm worried about the heat.
    > >
    > > Anyone have experience with these thermistors? I would assume the
    > > same style setup is used in most cars with "Low Fuel" warning lights?
    > >
    > > Thanks for reading,
    > >
    > > George
    > >

    I recently replaced the sending unit and fuel pump in my aerostar van
    and was shocked to find out that all of the wiring to the pump and the
    sending unit, which was basically a big potentiometer, were all
    exposed and submerged into the gasoline. I can't see why these things
    do not create a danger of explosion.
     
    Joey, Jul 26, 2003
    #3
  4. cxturbö wrote:

    > I'm having problems with the "Fuel Reserve" light on my motorcycle.
    > [...]
    > This evening I made a mockup circuit using that thermistor. With a 12
    > volt power source and a 4 watt bulb I left it energized for a few
    > hours. What has me worried is that the thermistor gets too hot to
    > touch. Submerging it in a very small vile of gas causes the light to
    > turn off. It's hot enough that I hear a very faint hiss as it goes in
    > the gas.


    Maybe you had the bulb replaced with a higher wattage type accidentally in
    the past?
    That might explain the failure and the not-so satisfying experiment.

    Just a thought,
    Michael


    --
    One smooth landing is mostly luck; two in a row is all luck; three in a row
    is prevarication.

    Promoting Penguin Power. Web home: http://www.qsl.net/dc1rn
     
    Michael Hofmann, Jul 26, 2003
    #4
  5. cxturbö

    cxturbö Guest

    On Sat, 26 Jul 2003 22:54:46 +0200, Michael Hofmann
    <> wrote:

    >cxturbö wrote:
    >
    >> I'm having problems with the "Fuel Reserve" light on my motorcycle.
    >> [...]
    >> This evening I made a mockup circuit using that thermistor. With a 12
    >> volt power source and a 4 watt bulb I left it energized for a few
    >> hours. What has me worried is that the thermistor gets too hot to
    >> touch. Submerging it in a very small vile of gas causes the light to
    >> turn off. It's hot enough that I hear a very faint hiss as it goes in
    >> the gas.

    >
    >Maybe you had the bulb replaced with a higher wattage type accidentally in
    >the past?
    >That might explain the failure and the not-so satisfying experiment.


    I think the original thermistor just failed after 20 years. Bulbs in
    the dash are still factory.
    Actually the experiment worked exactly as it's supposed to. From
    ignition "on" the Fuel Reserve light typically takes 30 seconds to
    start glowing. The new thermistor does the exact same thing. Touch
    it with a piece of paper soaked in gas and the bulb goes off.

    My biggest concern is the temperature of the thermistor getting hot
    enough to ignite the fuel.
     
    cxturbö, Jul 27, 2003
    #5
  6. cxturbö wrote:

    > My biggest concern is the temperature of the thermistor getting hot
    > enough to ignite the fuel.


    The ignition temperature of gasoline is approx. 240°C, so I wouldn't be too
    worried about that.

    Michael
     
    Michael Hofmann, Jul 27, 2003
    #6
  7. cxturbö

    cxturbö Guest

    On Sat, 26 Jul 2003 12:04:22 GMT, "CWatters"
    <> wrote:

    >I think it would be safest to look for a sensor form another bike that uses
    >a similar circuit but...


    Correct. I stopped by the library today and went through quite a few
    books. Unfortunately all they have are Clymers and Haynes which don't
    go into detail. No shop manuals.

    >
    >If' I'm looking at the right data sheets ...
    >
    >http://www.thermometrics.com/assets/images/rl20.pdf
    >
    >the NTC material is D9.7A and the data for that suggests the resistance is
    >on the high side..
    >
    >http://www.hhv.co.uk/PDF's/thermistors/keystoned97.pdf
    >
    >I made it 107 Ohms at around 85C.
    >
    >If that's correct I'm surprised the bulb lit up for your test.


    Actually it works great. Even the warm up time from ambient until the
    bulb starts to glow is about 30 seconds...exactly how the original
    used to work. Just touch it with a piece of paper dabbed in gasoline
    and the bulb goes out.


    >I think the cold temp is about right (could be a bit lower) but the graph
    >needs to be steeper with a lower hot temp. I think the resistance of the NTC
    >needs to be less than the resistance of the bulb at whatver temp you feel is
    >safe (say <10Ohms at 50C?)


    The specs on the thermistor I'm using mention it's good for up to
    150ºC.
    I measured the temperature of the thermistor today. Highest reading I
    obtained was ~85º C. From what I understand it would need to reach
    232º C in a perfect environment to ignite gas vapors.

    It should work perfect. I also have a 500 Turbo which is nearly an
    identical bike with the same sending unit. I'm going to pop the
    thermistor out of that one and do a comparison just to be safe.

    I've never really thought about what goes on in the fuel tanks of cars
    (or motorcycles) until now. Fuel pumps that are not hermetically
    sealed, rheostats that are designed for far less than 12 volts and go
    into meltdown from a short or failed resistor, heated thermistors,
    ect...I guess the engineers get things right since we don't hear of
    cars going off like bombs everyday :)

    George
     
    cxturbö, Jul 27, 2003
    #7
  8. cxturbö wrote:

    > ect...I guess the engineers get things right since we don't hear of
    > cars going off like bombs everyday :)


    You're not watching Hollywood movies? They do :)

    SCNR,
    Michael
     
    Michael Hofmann, Jul 27, 2003
    #8
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Replies:
    3
    Views:
    605
    Bob Masta
    Jan 11, 2005
  2. Tanty

    Gas tank meter

    Tanty, Jul 24, 2003, in forum: Electronic Design
    Replies:
    12
    Views:
    448
  3. Replies:
    0
    Views:
    283
  4. Replies:
    0
    Views:
    296
  5. esm.
    Replies:
    8
    Views:
    1,586
    (*steve*)
    Mar 8, 2012
Loading...

Share This Page