JOB OFFER: Automotive Application

shortbus= said:

I know you want to do it yourself, but, what is your time and car worth? Is it worth burning up the wireing harness and possibly the car? Is it worth more than the cost of one of these?
http://docskustom.com/shop/viewcategory.php?groupid=38

Click to expand...

thanks for the link, i saw those before, not sure why i didn't see the second one. i wonder if i can just hook that up to the gauge and call it a day, or use two of them at the sensor....

also, why would i burn up my harness? i appreciate your concern, but i don't see AT ALL how someone with my experience and knowledge would do that. i put a corvette motor in a nissan, rewired the entire assembly from scratch, including modification of lots of electrical components. if it was to burn up it would have already done so. the link you sent me has pretty much what i want to do, why would that cause my harness to burn? by that logic wouldn't these boxes burn my harness too?

EDIT: http://static.speedwaymotors.com/images/charts/FuelGaugeInterfaceModule.pdf

found a manual. i think all resellers are peddling the same box.

KeepItSimpleStupid said:

There are some measurements that won't hurt.
ECU
Open circuit voltage (I'm assuming 5V)
With a 50-250 ohm potentiometer (50 ish ohms in series with 200 ohm pot) see what the range of voltage is.
Adjust so voltage is 1/2 and measure the resistance of the POT out of circuit.

You kinda also have to do a similar thing and simulate the "original gas gauge sensor" with the gas gauge.
Pay some attention to linearity

Click to expand...

I will measure this weekend, thanks!!

gauge is simple ground signal provided from cluster and measured thru sender. 12v is provided via switched wire.

You might consider getting a sender that is ECU compatible and adding that potentiometer to your original sending unit. If you can't do it, maybe a machine shop can. Then you have both outputs from one sender.

ClydeCrashKop said:

You might consider getting a sender that is ECU compatible and adding that potentiometer to your original sending unit. If you can't do it, maybe a machine shop can. Then you have both outputs from one sender.

Click to expand...

i tried that, the custom ones i see need to have the module outside of the tank. the factory sender is completely immersed. it's just a rod with a float around it and 2 terminals.

these makers either wouldn't want to do it or they were wayy to expensive.

Anyway, my feeling is that hopefully, the sender is connected to ground, which makes life easier. You won't actually simulate a resistance, but what value of voltage is needed to ground at X resistance, tank level So, you get a set of tuples (resistance, voltage, tank level), then use a FET and set that particular drop across the FET.

Reading the gas gauge may be as simple as a voltage divider to say 5V. So, you can use a microcontroller with an A/d to read the gas guage and two D/A's and look up tables for as many points as you need (32, 64 or 128).

The measurements suggested, would help nail down what's required.

For fun, take a look at www.picaxe.com for a really simple controller. The I2C bus can be used to access a D/A or A/D.

So, theoretically, not hard.

gauge is simple ground signal provided from cluster and measured thru sender. 12v is provided via switched wire.

Click to expand...

If you are using the original gauge, driven by a 0-50 Ohm sender, to actually indicate fuel-level then presumably any fuel-level indicating signal sent out from the ECU will not be fed to that gauge. In which case a simple 150 Ohm fixed resistor to ground will fool the ECU into thinking there is a half-full fuel tank. I don't see any need for more complex emulation of a variable resistance.