Go Down

### Topic: Looking for a way to measure the fuel level for an RC car tank ! help ! (Read 3551 times)previous topic - next topic

#### vinceherman

#30
##### Nov 22, 2017, 03:04 pm
and will have a rotary switch to control the throttle pedal input
You can do that.
Or you can read the value coming out of the receiver in the car.

Quote
what i did not understand is number 7 I how to get fuel info from throttle and rpm ?
This is the core of my proposal.
The engine uses fuel when it is running.
At any particular RPM, the engine is using more fuel at higher throttle settings.
At any particular throttle setting, the engine is using more fuel at higher RPM.

I doubt that either of these relationships are linear, but let's assume they are for a starting point.

At the start of your program, set a fuelConsumed variable to zero.
Set up a method to count revolutions of the motor.  I guess external hardware or an interrupt.

In a loop, read the throttle position.  If you are reading the throttle channel from your receiver, I would guess that the throttle reading would return the number of microseconds of the throttle servo pulse, maybe a value ranging from 500 to 1500.  Map that to a range that starts at zero.

Now, multiple the throttle reading by the number of revolutions since the last pass through.
Add the result to the fuelConsumed variable.
Reset the number of revolutions to zero.
Transmit the fuelConsumed number to the pit.
Repeat the loop.

The number transmitted can be translated to an estimate of the fuel consumed.
It will be some goofy number.  But a number that grows at a similar rate to the rate of fuel consumption.
You just need to multiply it by some factor to get a usable fuel consumed number.

What value for the factor?  You need to determine that.  And adjust it through trial and error.

Quote
what do you think about this method of making Liquid capacitive sensor?
Try it out.  Calibrate it to read Full when the bottle is full.  Empty when the bottle is empty.  And a percentage in between.
Then fill the bottle with 1/10 of the full capacity.  Now shake the bottle and see what reading you get.
Is that usable?
I am not saying that it is not doable.  But I think it is hard to do it well.

#### Gerry48

#31
##### Nov 22, 2017, 09:12 pm
The normal way to measure fuel is with a float and pot.   Measure voltage with Arduino.  You'll have to cut hole in tank to insert float and glue electrical connector in place.

https://www.ebay.com/itm/Fuel-Level-Sensor-Gas-Gauge-Direct-Fit-for-Chevy-GMC-Isuzu-Oldsmobile-Pontiac/132034473951?hash=item1ebdde2fdf:g:htAAAOSwG7lZwRaA:sc:ShippingMethodStandard!93117!US!-1&vxp=mtr

#### firashelou

#32
##### Nov 23, 2017, 07:03 am
You can do that.
Or you can read the value coming out of the receiver in the car.
This is the core of my proposal.
The engine uses fuel when it is running.
At any particular RPM, the engine is using more fuel at higher throttle settings.
At any particular throttle setting, the engine is using more fuel at higher RPM.

I doubt that either of these relationships are linear, but let's assume they are for a starting point.

At the start of your program, set a fuelConsumed variable to zero.
Set up a method to count revolutions of the motor.  I guess external hardware or an interrupt.

In a loop, read the throttle position.  If you are reading the throttle channel from your receiver, I would guess that the throttle reading would return the number of microseconds of the throttle servo pulse, maybe a value ranging from 500 to 1500.  Map that to a range that starts at zero.

Now, multiple the throttle reading by the number of revolutions since the last pass through.
Add the result to the fuelConsumed variable.
Reset the number of revolutions to zero.
Transmit the fuelConsumed number to the pit.
Repeat the loop.

The number transmitted can be translated to an estimate of the fuel consumed.
It will be some goofy number.  But a number that grows at a similar rate to the rate of fuel consumption.
You just need to multiply it by some factor to get a usable fuel consumed number.

What value for the factor?  You need to determine that.  And adjust it through trial and error.

Try it out.  Calibrate it to read Full when the bottle is full.  Empty when the bottle is empty.  And a percentage in between.
Then fill the bottle with 1/10 of the full capacity.  Now shake the bottle and see what reading you get.
Is that usable?
I am not saying that it is not doable.  But I think it is hard to do it well.
well not easy at all, what should the factor be like how to determine it ?

and what do you think of the sensor that Gerry48 suggested in reply #31 ?
Aiming for the unreachable !

#### firashelou

#33
##### Nov 23, 2017, 07:04 am
The normal way to measure fuel is with a float and pot.   Measure voltage with Arduino.  You'll have to cut hole in tank to insert float and glue electrical connector in place.

https://www.ebay.com/itm/Fuel-Level-Sensor-Gas-Gauge-Direct-Fit-for-Chevy-GMC-Isuzu-Oldsmobile-Pontiac/132034473951?hash=item1ebdde2fdf:g:htAAAOSwG7lZwRaA:sc:ShippingMethodStandard!93117!US!-1&vxp=mtr
I was looking for a sensor like this but how to get the value to be read by arduino ?
Aiming for the unreachable !

#### Boardburner2

#34
##### Nov 23, 2017, 08:58 am
Is there enough space in your model to accomodate one of these, they tend to be physically large.

#### firashelou

#35
##### Nov 23, 2017, 09:30 amLast Edit: Nov 23, 2017, 09:31 am by firashelou
Is there enough space in your model to accomodate one of these, they tend to be physically large.
oh right i think not because the tank is 750ml
but what about the sensor used in it ? maybe i can find something similar but small
Aiming for the unreachable !

#### Boardburner2

#36
##### Nov 23, 2017, 09:34 amLast Edit: Nov 23, 2017, 09:44 am by Boardburner2
I doubt that either of these relationships are linear, but let's assume they are for a starting point.

But I think it is hard to do it well.

My thoughts exactly.

There could b considerable work in getting an accurate map.

oh right i think not because the tank is 750ml
but what about the sensor used in it ? maybe i can find something similar but small
Even with a baffled tank i doubt you will get reliable measurements in a mooving vehicle.
(Smaller system less damping.)
Best i would hope for is 1/2 1/4 full.

Others have suggested a flowmeter which could be much simpler to implement

#### Boardburner2

#37
##### Nov 23, 2017, 09:49 amLast Edit: Nov 23, 2017, 09:57 am by Boardburner2
Perhaps an older idea could be used, use a separate much smaller reserve tank.

Do not know how to physically implement such a fuel system though.
EDIT
Perhaps use a conical base tank with a simple float level switch at the top of the cone.
These used to be available in 1 Litre sizes for testing fuel consumption on full size cars.

#### Gerry48

#38
##### Nov 23, 2017, 05:09 pm
Glue a micro switch to the bottom of the tank.  Tie a string to the end of the switch and connect the other end of the string to a fishing bobber.  The bobber holds the switch closed until fuel level is lower than the string length.  The switch opens.

#### Boardburner2

#39
##### Nov 23, 2017, 06:05 pm
Good idea.
The fishing bobber could then stop the fuel flow from the main tank to stop air entrainment and draw from a reserve.

#### firashelou

#40
##### Nov 24, 2017, 07:11 am
Glue a micro switch to the bottom of the tank.  Tie a string to the end of the switch and connect the other end of the string to a fishing bobber.  The bobber holds the switch closed until fuel level is lower than the string length.  The switch opens.

looks interesting but i can't drill an extra hole for the bobber and string, and it would give only when the tank is low
Aiming for the unreachable !

#### Gerry48

#41
##### Nov 24, 2017, 05:14 pm
No need to make holes in the tank other than for the two small wires.  You assemble the micro switch with bobber through the fuel tank fill hole.  You don't even need a string on the bobber.  Just clip the bobber to the switch lever and glue it in place.

You could also attach a small potentiometer inside the tank and use a float (small bobber) to get analog measurements.  Again, all assembly can be make from the tank fill hole.

#### firashelou

#42
##### Nov 27, 2017, 06:51 am
No need to make holes in the tank other than for the two small wires.  You assemble the micro switch with bobber through the fuel tank fill hole.  You don't even need a string on the bobber.  Just clip the bobber to the switch lever and glue it in place.

You could also attach a small potentiometer inside the tank and use a float (small bobber) to get analog measurements.  Again, all assembly can be make from the tank fill hole.

ok but the tank fill hole will be closed with a cap so how should that be done ?
and about the potentiometer  how would it take it's place back to default position ?
Aiming for the unreachable !

#### polymorph

#43
##### Nov 30, 2017, 09:09 pm
Pick the pulse off of pin 3 before the RC smoothing network, the pulse width after subtracting the empty tank width should be a linear function of how full the tank is.

The tough part is vibration foaming up the fuel.
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - https://tinyurl.com/Technote8
Multitasking: forum.arduino.cc/index.php?topic=223286.0
gammon.com.au/blink - gammon.com.au/serial - gammon.com.au/interrupts

#### Gerry48

#44
##### Nov 30, 2017, 09:51 pm
Don't understand you question.   You open the tank fuel fill hole and glue in the potentiomenter.  Attached to the pot shaft is a 90deg elbow.  Attached to the other end of the elbow is a short rod.  Attached to the other end of the rod is a float.  All these items are scaled to the size of your tank.  You attach two wires to the pot and use a voltage divider to measure voltage.  That's how virtually all car fuel tanks are measured.  It's simple and accurate.

ok but the tank fill hole will be closed with a cap so how should that be done ?
and about the potentiometer  how would it take it's place back to default position ?

Go Up