# Need a resistance divider (?) or gas gauge

Discussion in 'The Projects Forum' started by Jack_K, Jul 16, 2009.

1. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
I have a motorcycle with a gas gauge. The sending unit reads 10 ohms when the tank is full and 215 ohms when it's empty. OK so far.

I want to add a second gauge (2 total gauges). The second gauge needs 10 ohms when full but only 70 ohms when empty.

How can I have both gauges and have both of them read accuratlely? I think both gauges are the heating element type.

Jack

Last edited: Jul 16, 2009

Jul 7, 2009
1,585
141
You'll get better responses if you disclose what you're trying to accomplish. Making your own gas gauge isn't something many folks would likely undertake unless you know exactly what you're doing. One accidental spark and you've got a problem.

Based on what little information you've given, why not use an identical fuel sending unit, put 104 Ω in parallel with it and put 61 Ω in series with this. This should give you the desired range, but may change the shape of the response curve.

3. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
I guess I just wasn't clear. I don't want to build a gas gauge, I already have them. The motorcycle has a gas gauge on it. The sending unit for that gauge has 10 ohms when the tank is full and 215 ohms when the tank is empty. All that works fine.

I want to add a second gas gauge in parallel with the existing gas gauge. I have the gauge but it is built to use a sending unit that has 10 ohms when the tank is full, but only 70 ohms when the tank is empty.

So, if I simply parallel the two gauges, the additional one will read empty long before the tank gets empty (70 ohms vs 215 ohms). Also, I suspect that paralleling two gauges will make the original gauge read wrong. That really complicates the problem.

I had thought about changing the voltage supplied to the second gauge but I think it would have to be increased rather than decreased, and that isn't possible.

I can't install an additional sending unit in a motorcycle tank. No way to do that.

Any other ideas?

Jul 7, 2009
1,585
141
OK, Jack, now I see what you want -- I think you were clear enough; I just misunderstood.

You don't say what kind of gauge you're using, but let's assume it's an analog gauge that responds to current. I'm not sure what you mean when you say they're the "heating element type".

I'd also assume the existing gauge and sender are in series with a voltage across them as shown in the attached picture (but it might not be 12 volts). You'd get a larger current to deflect the meter to full scale with a small resistance and a smaller current for reading empty; thus, R is a low resistance for full and a high resistance for empty.

You're correct in assuming you can't just parallel the gauges -- you'll change the effective resistances and neither gauge will work correctly.

I would guess the easiest way to get what you want would be to utilize the voltage that is dropped across r, the existing gauge. You'd input this voltage into a high impedance amplifier and use the output to drive the second gauge. The purpose of the amplifier is to utilize the voltage across r without modifying it significantly.

In the pictured circuit, when the original gauge reads full, a large current is flowing through the meter, so the voltage across r will be large. At empty, the voltage across r is small. Thus, amplifying the voltage across r with a non-inverting amplifier will allow you to drive the second gauge without changing the reading of the first gauge. You'll have to adjust the DC offset of the lower input to the op amp (i.e., it might not be connected to ground) and the amplifier's gain to get the second gauge to read analogously to the first gauge and the amplifier will have to have enough current output to drive the gauge (or you can use an additional stage).

• ###### gas_gauge.jpg
File size:
74.1 KB
Views:
36
Last edited: Jul 18, 2009
5. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
I'm not completely certain, but I believe you have the gauge and sender swapped around. It would help a good deal if our OP would post that portion of his motorcycle's schematic.

12v is usually supplied to the gauge. The sender unit is normally grounded at or near the fuel tank. One wouldn't want much current flowing through the sender unit, as sparks in a fuel tank would be A Very Bad Thing.

6. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
12v is usually supplied to the gauge. The sender unit is normally grounded at or near the fuel tank.

You're correct. The 12 volts connects to the gauge and the gauge connects to the sending unit. The sending unit is inside the gas tank and gets grounded externally.

What I meant by heating element is that there's a bi-metalic strip in the gauge. As more current flows it heats and expands and moves the needle.

Jul 7, 2009
1,585
141
SgtWookie: undoubtedly you're right -- thanks for pointing that out. Actually, the current flow would be the same with either arrangement, but I'm betting the reason to keep the voltages to a minimum around the sender unit is in case something shorts to it -- as you say, you'd want to minimize the chance for a spark.

Jack_K: thanks for educating me. I've never taken an automotive meter apart and wasn't aware that's how they worked. But I can see how that would be much cheaper to make than a typical moving coil galvanometer.

Hopefully the basic method I gave above will work -- the circuit gurus here will help you pin down an actual circuit.

8. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
Jack, can you connect the new gas gauge up to a 14v supply with a resistor to ground, and measure the voltage across the resistor, and also measure the resistance? The measurement needs to be pretty accurate.

Also need to find out what the voltage measures on the wire from the existing gauge to the sender when the tank is full and when it's empty.

9. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
I measured with the second gauge since it's on my workbench. Here's what I got. I connected the 12 volt pin to a 12 volt gelcell battery. The sensor pin was connected to different resistors. The other end of the resistor was connected to the battery's negative terminal. I then measured the voltage across the resistor (with a DVM).

10 ohms - 1.06 v = 106 ma
20 ohms - 1.82v = 91 ma
30 ohms - 2.36 v = 78.6 ma
40 ohms - 2.80 v = 70 ma
50 ohms - 3.12 v = 62
60 ohms - 3.41 v = 56.8 ma
70 ohms - 3.63 v = 51.7 ma

You can see it's nonlinear, but not too far off.

I haven't measured the motorcycle's gauge. I'm going to try that tomorrow.

Jack

Last edited: Jul 21, 2009
10. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
I wasn't able to check the gas gauge on the bike today. Maybe tomorrow.

11. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
This is what I'm thinking is what I need but I don't know what the gain should be yet.

Jack

File size:
11.5 KB
Views:
25
12. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
The readings you posted in reply #9 today are quite different from those posted on the 18th.

The gauge needle is moved by a bimetallic strip that has a heating element. The heating element will change resistance somewhat as its' temperature changes, which will change the readings you get.

It takes time to warm up the bimetallic strip. Allow 2 minutes time for warm-up from when you first connect power until you take the reading.

Also, there may be a variance in the resistance of the resistors that you are using. You are reporting standard resistance values, but not the actual measured value of the resistance, which can vary significantly.

Additionally, the nominal voltage of your cycle's electrical system will likely measure somewhere in the vicinity of 13.8v when running above around 1,200 RPM with a charged battery. A gel-cell battery on the bench with no charger attached will probably measure 12.8v or less.

Misreporting or erroneously reporting these measurements will negatively impact the final result of the project, as these errors are cumulative.

13. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
The reason I changed the values is because I redid the measurements. I used a 10-turn 100 ohm pot and a DVM to measure and set the resistance and then measured the voltage across the pot. I then calculated the current.

I was assuming that, if I used the same battery when I duplicate the measurements on the bike's gas gauge, the ratio should be OK. Is that wrong?

14. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
OK, that's good to know. For now I'll just assume that the resistance values you posted today are within 1%.

They'll be close. Did you measure the battery voltage at the start and end of the test?

15. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
No. I never thought of that. I had just charged the battery over night and took it off the charger about an hour earlier.

16. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
Here's an update on my circuit. The op-amp will output the voltage across Gas Gauge 1. Right now I don't know exactly what voltage that will be but I do know the voltage required for Gas Gauge 2 is approximately 10.2 to 12.5 volts based on V1 being 13.8 volts.

I doubt that the two gauges need the same voltage so that would mean that I need a non-linear differential amplifier.

Anyone know how to do that?

Jack

File size:
13.1 KB
Views:
20
17. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
Jack,
This is more of what I had in mind for your circuit:

Right now, it's just a simple voltage follower circuit with a PNP transistor added to provide higher current.

However, you haven't provided any details about how your existing fuel gauge interacts with your existing sender.

Also, the data you've already provided may be skewed due to battery discharge over the time of the test.

Until you provide at least some data for the existing sender/fuel gauge, it'll be difficult to make progress.

File size:
11.1 KB
Views:
87
18. ### Jack_K Thread Starter Active Member

May 13, 2009
115
0
"Until you provide at least some data for the existing sender/fuel gauge, it'll be difficult to make progress."

Thanks Sarge. I saw this circuit on a previous post. I removed the gauge from the motorcycle and tried to measure it the same way I did the external gauge. Unfortunately, the "gauge" is one complete digital package and I was unable to accomplish the task.

I'm goung to try again using a method wherein the instrument cluster stays on the bike. Unfortunately I have to travel on business starting tomorrow and I'll be out for a week.

Also, I posted a new thread with my latest op-amp idea and question.

Jack