Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Electronic fuel gauge

Status
Not open for further replies.

fingaz

Member
Hi,

I have just brought a motorbike, which has been 'modified' by removing the fairing. Only problem is that the guy who removed the fairing has also got rid of some of the electrics.

Basically, the bike has no fuel gauge. It has the sensor, and wiring. The original fuel level sensor works by varying the resistance depending on the level of fuel. According to the specs, it should read between 4 to 10 ohms when full, and between 90 to 100 ohms when empty.

I basically want to make some kind of 'gauge', though all it needs to do really is show when the tank is full, half full, and empty. I can't just stick the original gauge back on because I'm restricted to space. If I can make something, i can make it fit the space I have.

Can anyone make suggestions as to how I could do this? PIC's are not really an option unless someone already has the 'software/code' (I have a pic programmer, but am not experienced at all with writing code)

All suggestions are welcome, any help would be gratefully received.

Thanks
 
Last edited:
Do you mean 4 to 10 ohms when empty?

You'd need to do some electronic design but a LED bar-graph IC **broken link removed** connected to an LED light bar indicator should do what you want.

You would want to power it from a regulated voltage (about 9V) using an LM317 regulator.

Energizing the fuel gauge sensor from 9V through a 500Ω resistor will give about a 1.5V to .15V signal as the sensor goes from full to empty.

Setting the sensitivity of the LM3914 to 1.5V full-scale will give you the full LED bar output when full, and one or two bars when empty.
 
Hiya,

Thanks for the info and suggestion. The specs state that the gauge should read 4 ohms to 10 ohms when FULL, and 90 ohms to 100 ohms when EMPTY. I will have a look at the link you suggested.

I will see what I can do with it, and come back on here if I have any problems.

Neil
 
Last edited:
Thanks for the info and suggestion. The specs state that the gauge should read 4 ohms to 10 ohms when FULL, and 90 ohms to 100 ohms when EMPTY. I will have a look at the link you suggested.
Well, that will invert the signal from what I suggested, which complicates things somewhat (unless the fuel sensor has two leads, but I suspect one end of the sensor is connected to the tank ground.)

One solution is to run the signal through an op amp connected as an inverter with 1.5V offset at the + input. This would give near 1.5V output when the tank is full.
 
The specs state that the gauge should read 4 ohms to 10 ohms when FULL, and 90 ohms to 100 ohms when EMPTY.

10 mV in series with a 1 mA full scale meter movement of very low resistance in series with this sending unit will give 1 mA with 10 ohms and 0.1 mA with 100 ohms.
So, 1.00 vdc in series with a 100 mA meter (i.e., an ammeter) movement will do the same thing.
You need to draw the meter scale to show full and empty, instead of using level shifters and trimpots.
 
Last edited:
My Kawasaki 750 GPZ has a bad LCD and the fuel level readout is so dim its is impossible to read while moving.

I just drive until it sputters like its running out of gas and turn the little knob under the tank from main to reserve and head for the fuel pump!
Never failed me yet!
 
Hi all,

Thanks for the suggestions and ideas.

Temtech -> Unfortunately the XJ900 does not have a reserve position on the fuel tap. I did originally consider this as an option.

Anyway, I found a schematic which is here : **broken link removed**.

I have built this, but only using 1 LM339, and 4 led's. The only thig I have noticed is that the led's sometimes flicker when the petrol shloshes around the tank.

I assume that if I put a small electrolytic across each of the led's that this will reduce some of the flickering. Can anyone confirm this? or make an other suggestions on how to solve this?

Thanks
Neil
 
The impedance across the sender is a few Ω. You need a low-pass filter with a cut-off frequency of ~0.05Hz, or to put it another way, you want the fuel gauge to show the "average" fuel level over about a 10sec window.

To get a 10 second time-constant, with a sender source resistance of say 100Ω, it would take a capacitor of 0.1 Farads, which is 100,000uF, a huge capacitor.

You could put a series 10K resistor between the sender and the 8 comparator inputs without running into an offset due to the input bias current of the comparators. With 10K, the shunt capacitor would be 1000uf, still quite big.

Another alternative is to add an opamp active filter between the sender and the eight comparator inputs, where you can scale the series resistance up to ~1meg, and that makes the capacitor ~10uF. See the attached circuit.
 

Attachments

  • LPF.jpg
    LPF.jpg
    183.5 KB · Views: 1,318
Last edited:
Hi Mike,

Thanks for the quick response, and for the very easily understandable explanation. I will give this a try over the weekend (when I'm next off work).

Will let you know how I get on with it.
Oh, I assume pretty much and op-amp would be ok. (I have some lm741's somewhere, and some lm358's too)

Thanks again,
Neil
 
Last edited:
I posted a better schematic above.

A 741 won't cut it. The input bias current will cause too much voltage drop across the ~1meg input resistance. Besides, you need an amp which has a common-mode input range down to ground. Its output must also pull close to ground, which a 741 won't do. The LM358 is better. I added another resistor to balance offset in light of input bias current. Maybe someone else can comment.
 

Attachments

  • LPF.jpg
    LPF.jpg
    181.3 KB · Views: 898
That's a brilliant idea. Maybe he can gin up a 'chopper' circuit, and use samples of the reading to charge his filter. I've never done a project like that, but I think it's theoritically feasable.
 
Yes, a 'chopper' or switched capacitor circuit can be use to give an apparent high resistance for low frequency filters. You use a multivibrator, such as a 555, to generate low duty-cycle pulses that periodically turns on a CMOS switch (such as a CD4016 or CD4066) in series with a resistor and a capacitor to ground, forming a low-pass filter. The duty-cycle determines the relative increase in resistance, for example a 1% duty-cycle would increase the apparent resistance by a factor of 100.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top