Tapping into Fuel Gauge wiring

#1

Hi guys, I signed up for this site just for this question, as you lot seem to be the most knowledgeable part of the E30 world on this topic.

I’m installing a PC into my E30 (1989 325i Touring), with the aim of displaying and recording live engine data. I’ve tapped into the existing instrument cluster wiring to get signals such as tacho, speed, and coolant temp, and I’ve added my own oil temp and pressure sensors.

But I’m struggling with the fuel level gauge. I’ve identified the wire and with ignition off I get the appropriate resistances through the wire. But as soon as I turn IGN on I get voltage, and with the engine running I get a fixed voltage of 3.9V, no matter what amount of fuel is in the car.

I don’t want to disconnect the fuel gauge in the cluster, but I want to be able to read the sensors - is there any solution for this?

#2

Contact Jim Levie via “Members”. He probably has the most experience in this.

#3

Start here

http://spece30.com/forum/41-electrical-gauges-and-sensors/43975-fuel-level-gauge#61725

#4

I’m not very knowledgable on the subject, but I know that the OBC has a fuel level signal as well. When you press the “1000” and “1” buttons together, it will tell you how many liters of fuel remain. I have no idea how to go about getting that signal into your PC, but it’s a possible alternative.

#5

Thanks guys. I’ve successfully accessed the external temp sensor, the coolant, and my own oil temp and pressure sensors. But the fuel, speed and tacho signals have got me confused - I’m not sure what signals I should be looking for (voltage, square wave frequency etc). And this voltage on the fuel level wire is very confusing, since it’s just not going down - fixed at 3.9V.

#6

The vehicle and engine speed signals are square waves. The fuel level sensors are variable resistors to ground, with 0 ohms being a full tank and 60 (55L tank) or 120 (63L tank) ohms being an empty tank. The voltage on the wire to the sensors will vary with fuel level.

#7

Quick response, much appreciated!

That’s what I thought, but I can’t seem to replicate those results with my equipment. But it must be a user* or equipment error in that case.

*me, the idiot.

#8

I’m not sure what you are doing wrong. But the way to play with it is to remove the sensor(s) from the tank, which will allow you to invert the sensor(s) to simulate a full and empty tank.

#9

I feel guilty reviving this after five months, but I want to thank those that helped. I’ve now got readings from my fuel line using a voltage divider circuit, and got accurate tank levels. I’ve also managed to get Tacho readings using a frequency-to-voltage converter plus an extra 1k resistor so that there’s no interference with the existing gauge.

My new problem is with the speedo. The same F2V converter can’t get any reading from either the blue/red wire that fed the OBC (long gone) or from the green plug at the bottom of the dash cluster. All the wiring checks out, I just can’t get a reading and I don’t have an oscilloscope to check for a square wave.

Any ideas?

#10

Does the cluster speedometer work? If it doesn’t, there is a problem with the signal from the sensor in the differential. If it does, that signal may be too low a level for your F/V converter to handle. I’ve not ever looked at the signal, so I don’t know what it’s level is.

#11

[quote=“Grrrmachine” post=71403]I feel guilty reviving this after five months, but I want to thank those that helped. I’ve now got readings from my fuel line using a voltage divider circuit, and got accurate tank levels. I’ve also managed to get Tacho readings using a frequency-to-voltage converter plus an extra 1k resistor so that there’s no interference with the existing gauge.

My new problem is with the speedo. The same F2V converter can’t get any reading from either the blue/red wire that fed the OBC (long gone) or from the green plug at the bottom of the dash cluster. All the wiring checks out, I just can’t get a reading and I don’t have an oscilloscope to check for a square wave.

Any ideas?[/quote]

Can you share your setup on the fuel circuit? Wires tapped and how you got the signal back? Are you using a 0-5V reading on it?

#12

Yes, the cluster works. I’ve been told the signal is a 5V square wave, in the region of 1.5mA, which may well be too low for an LM2907 to handle. Trouble is, how do I boost that signal to make it readable?

I’m using 3.3V on this setup rather than 5V, but the principle is the same. Tap into the blue/green wire that comes from the yellow plug behind the dash (pin 6). Create a voltage divider using either a 60-Ohm or 120-Ohm resistor, depending on the size of your tank (55L = 60 Ohms). To do this, connect both your datalogger and the new resistor to the wire, and ground the resistor.

#13

Yes, the cluster works. I’ve been told the signal is a 5V square wave, in the region of 1.5mA, which may well be too low for an LM2907 to handle. Trouble is, how do I boost that signal to make it readable?[/quote]
If you are getting the signal from the green connector, it should a 5v square wave. I think that the raw signal from the sensor drives the speedometer directly and a circuit in the speedometer produces a 5v square wave from that for the OBC and CC.

If that is the case, the problem could lie in the speedometer.

#14

SOLVED! The problem appeared to by my Frequency-to-voltage design.

The F2V that came with my data-logger was based on a tiny LM2907N-8 chip - very minimalist design. That worked for the tacho, but not the speedo.

So I downloaded the datasheet for that chip and built another, using the same basic circuit board but with the bigger LM2917 chip. Again, worked on tacho, didn’t for speedo.

Then I used found deisgn, which is from a poster over on the e30tech webste:

http://foz11.tripod.com/DIY_MOD_s/Digital_Speedo/digital_speedo.html

I assembled it with no modification, and the circuit accepted the 3.3V input (instead of the 8V recommended) with no problems. At 60kmh it pumped out 0.25V, rising steadily to about 0.375V at 80kmh. Since it uses the same LM2917 chip as before, the issue can only be with the capacitor/resistor layout.

Glad I’ve sorted it, and hopefully this will prove useful to others wanting to keep their existing gauges.