HELP me build a gas gauge????

[SPDCHK]

How do you know that the fuel level meter gives those exact Ω values? Did you measure it, or did you get the information from a datasheet or something similar?

Nevertheless, the sensor can still work, but the signal now needs to be amplified. As can be seen from the attached schematic, I now changed the sensor and voltage divider resistor network around, to cater for the grounded sensor signal and the feedback signal (for lack of a better word). I.e. the ONE wire that comes from the sensor (and I assumed here that the Ω measurement was made with one lead of the DMM (Digital Multi Meter) connected to the car chassis (i.e. ground) and the other lead was connected to the sensor).
The wire from the sensor is wired in series with R3 (1.452KΩ) I got this value calculating a volt drop of 1V over the sensor if the tank is empty (i.e. sensor = 132Ω).

With the sensor measurement at 12Ω, the circuit simulator gave an output of 5.03V, and with the sensor measurement at 132Ω, the output was 0.11V. The OPAMP output can be wired directly into the LM3914 signal input.

I hope this will help you. Don't feel too bad about not knowing too much. I was in the same boat about a month ago. Our forum colleague, Mr. Eric Gibbs helped me with the designs of OPAMP circuits.

Good luck.

 

[Leftyretro]

How do you know that the fuel level meter gives those exact Ω values? Did you measure it, or did you get the information from a datasheet or something similar?

Nevertheless, the sensor can still work, but the signal now needs to be amplified. As can be seen from the attached schematic, I now changed the sensor and voltage divider resistor network around, to cater for the grounded sensor signal and the feedback signal (for lack of a better word). I.e. the ONE wire that comes from the sensor (and I assumed here that the Ω measurement was made with one lead of the DMM (Digital Multi Meter) connected to the car chassis (i.e. ground) and the other lead was connected to the sensor).
The wire from the sensor is wired in series with R3 (1.452KΩ) I got this value calculating a volt drop of 1V over the sensor if the tank is empty (i.e. sensor = 132Ω).

With the sensor measurement at 12Ω, the circuit simulator gave an output of 5.03V, and with the sensor measurement at 132Ω, the output was 0.11V. The OPAMP output can be wired directly into the LM3914 signal input.

I hope this will help you. Don't feel too bad about not knowing too much. I was in the same boat about a month ago. Our forum colleague, Mr. Eric Gibbs helped me with the designs of OPAMP circuits.

Good luck.



View attachment 15835

That should do it, nice job. The op is scaling and reversing the voltage direction to make it compatible with the display chip. However one improvement might be to regulate the +12VDC down to say +10VDC and recalculate the resistors. Otherwise as the car's voltage varies somewhat with speed, idle, etc the reading is bound to move around some, but maybe not with just 5 'step' display.

Lefty

 

[browningbuck]

HOLLY ****!!! this is getting all kinds of complicated....as for how i know the resistance values , i did use my Fluke and moved the arm . all you guys have been really cool, and i wish i could express my appriciation.

now that, thats said....how the hell do i build this thing now??? i have the LM3914 hooked up as shown with the R1 as 1.2K and R2 as 3.8K and the cap. at 2.2uf. as for SPDCHK's diagrams(to amp it up) i dont understand the R3 value is that the same as 1.45K and what is that SPAN??? god i am sooo sorry that you have to walk me through this. and if i connect all this together it should work ???

 

[SPDCHK]

ZERO and SPAN is terminology used in instrumentation where an input signal (the signal that you want to measure) is converted into something more usable by electronic devices.

Your fuel gauge is a typical example. When the fuel tank is empty, the sensor gives out a certain value, and the same for when the tank is full. BUT, the value from the sensor is not usable for normal electronics. The ZERO setting is to adjust the minimum (or zero) value, and SPAN is for the full scale value. In this case 0 volt (ZERO) 5 volt (SPAN)

Leftyretro made mention of perhaps using 10V i.s.o. 12V. Personally I would still use a 7812 voltage regulator in a car. (Usually the charging and regulating voltage of a car battery runs around 13.4V). Mind you, I have never done this, so I might be wrong.

Basically, if you connect the two circuits as posted it "should" work. However, theory and practice very seldom agree at the first attempt. You will have some teething problems, circuit bugs and short circuits. Don’t worry, it’s normal.

 

[browningbuck]

okay that made a WHOLE lot more sense now you put it that way

okay but i still need to know about that R3 value in the amp curcuit is that 1k45 the same as 1.45K resistor??? anyway i will go build this today thanks a ton

 

[superbrew]

The 7812 needs at least 14.5V at the input. I would say to use at least the 7810, which requires 12-5V, or the LM2931AZ50R as 3v0 suggested earlier. Automotive environments tend to pretty harsh to electronics, as there is a lot of noise and voltage spikes.

 

[chemelec]

Although Not Precise because of Battery Variations, This Circuit should do what you want.

For better accuracy, You could possibly use a slightly lower or higher valuse for the 150 ohm resistor.

HOLLY ****!!! this is getting all kinds of complicated....as for how i know the resistance values , i did use my Fluke and moved the arm . all you guys have been really cool, and i wish i could express my appriciation.

now that, thats said....how the hell do i build this thing now??? i have the LM3914 hooked up as shown with the R1 as 1.2K and R2 as 3.8K and the cap. at 2.2uf. as for SPDCHK's diagrams(to amp it up) i dont understand the R3 value is that the same as 1.45K and what is that SPAN??? god i am sooo sorry that you have to walk me through this. and if i connect all this together it should work ???

 

[browningbuck]

oh you totally screwedm me up now... so would this still need that amp circuit that SPDCHK had designed....i will have to give this a try tonight

i have a 1.2K and a 3.8K resistor will that be okay?? i dont want to run to the store today but i might have to

 

[Darth Bagel]

Basically, if you connect the two circuits as posted it "should" work. However, theory and practice very seldom agree at the first attempt. You will have some teething problems, circuit bugs and short circuits. Don’t worry, it’s normal.

Your diodes are backwards (or should be connected to +5 instead of ground) in the first circuit. And check the value of R2, if you would. I'm guessing you copied most of it from the "typical application" circuit, in which R2 is 3.83k instead of 2.83k.

 

[superbrew]

The diodes are not necessarily connected backwards, although pull up resistors will be needed. The reason that they are connected like that is that the voltage drop across the sensor is inversely proportional to the quantity in the tank.

EDIT: This is only true if the opamp is not used.

 

[browningbuck]

okay the direction on the LEDs are correct...they light up how htey should

if i where to put this together would this work...please look at the alt to the fuel gauge part

**broken link removed**

 

[Darth Bagel]

I was referring to this schematic. My apologies if that wasn't the one being referenced.

 

[chemelec]

oh you totally screwedm me up now... so would this still need that amp circuit that SPDCHK had designed....i will have to give this a try tonight

i have a 1.2K and a 3.8K resistor will that be okay?? i dont want to run to the store today but i might have to

Yes, 3.8K is OK.
No you Don't need that Op-Amp.

I never Built this to test it, But it should work.
Give it a try.

Gary

 

[browningbuck]

Okay Chemelec i LOVE it now that ive looked at it, just one question. is it okay to shoot that 12V down to the potentiometer that is surrounded by GAS?????

 

[chemelec]

Okay Chemelec i LOVE it now that ive looked at it, just one question. is it okay to shoot that 12V down to the potentiometer that is surrounded by GAS?????

If it is a PROPER SENDER potentiometer, made to be in the gasoline, it should be OK.
But other than that, I can't say without knowing the real quality of that part.

 

[browningbuck]

okay sweet , yeah it comes straight from Honda and is made for it to be in gas (just didnt know if orginally a lower voltage was sent to it)



tomorrow morning i will have results!!!!!!!

 

[browningbuck]

DYUMN chemelec it worked...it has some small bugs to work out (like once i hit the very to of full it doesnt show a light..but just a slight bit off full it comes on) then i need to get some diodes in there so it acts like a fuel gauge(only one light is on at any give point) then the final thing i need to work out(inbetween positions from one light to the next there may be no light showing at all until i get to the next one(i think the diodes will fix that))

 

[superbrew]

How do you have pin 9 connected? If you want "dot" mode then leave it open, otherwise connect it to pin 3 and it will be in "bar" mode.

 

[browningbuck]

AHHHHHHH sweet that will solve that proble and i wont have to use diodes
that make me REAL happy

hey how do i open up the range a little (so that it would range from 11 ohms to 132ohms???) would that be off the 150ohm resistor that i would make bigger????

 

[chemelec]

Instead of only using the 5 LED's as you mentioned, Why not use ALL 10 LED's?

It will Work Better and definately show Full.

Edit: And As I said previously:
That 150 Ohm resistor was Only an Aproximate Value. It may need to be slightly Lower or Higher to get an Exact range for what you want.
(I only used the 150 Ohms, as it is a Standard, off the shelf value.)

Gary

DYUMN chemelec it worked...it has some small bugs to work out (like once i hit the very to of full it doesnt show a light..but just a slight bit off full it comes on) then i need to get some diodes in there so it acts like a fuel gauge(only one light is on at any give point) then the final thing i need to work out(inbetween positions from one light to the next there may be no light showing at all until i get to the next one(i think the diodes will fix that))