Engine Temperature using an AD590 on the Oil Pressure Wire to the engine

[Danwvw]

I was thinking of a way to see my engine temperature without having to thread a wire to the engine.
Planning on using an AD590 IC Transducer which will run on my cars 6 volts fine and only has 2 wires it's pretty high impedance operating at just 1 ua/Kelvin
255 ua = 0' F (Degrees Fahrenheit)
388 ua = 240' F.
Need help figuring out the circuits! It could power one of the little cheap voltage displays to read Zero Volts at Zero Degrees F and 2.40 Volts at 240 Degrees F.
It would run on the Oil Pressure wire and that wire would still be active operating Oil Pressure Light in the car.
This is what I have so far:

 

[crutschow]

You circuit is novel and should conceptually do what you want.
However the devil is in the details and, in trying to implement it with real parts, I had a problem with generating the required stable gain and offsets with the normal variation in battery voltage and the high common mode voltage of the signal.
One of the problems is that the bridge circuit output varies with the applied voltage when it is not balanced, as is the case here.

So I generated a, perhaps odd looking circuit (LTspice simulation below), that stabilizes the voltages with a programmable shunt voltage reference (U2) and tried to minimize errors by running the sensor current through an op amp feedback resistor to directly measure the current.

Due to feedback, Op amp U1a biases the temperature sensor (simulated by current source I_Temp) at 4V from the 4V Ref2 voltage.

The op amp feedback also forces the temp sensor current to go through feedback resistor R1 to generate the output voltage proportional to the current (temperature).

The current through R2 from the 6V Ref voltage, offsets the 255uA bias current at 0°F to give a zero output at that point.

To avoid interaction of the two signals, I separated the function of the the oil light from the rest of the circuit by op amp U1b acting as a comparator which drives a transistor to turn on the oil light when the oil switches closes.

The simulation below starts with the oil switch (red trace) closing at the 0.1s point and opening at the 0.5s point.

The temperature current then starts increasing (yellow trace) from 255µA and going to 388µA, with the output voltage (green trace) starting from near 0V and going to ~2.4V at the maximum temperature.

Due to component tolerances, some tweaking of R2 (offset) and R4 (gain) may be needed to get the proper output offset and range.

 

[rjenkinsgb]

One of the problems is that the bridge circuit output varies with the applied voltage when it is not balanced, as is the case here.

Could that be used to an advantage for simplicity, by taking the display reference voltage from a resistive divider fed from the top of the resistor bridge?

The whole thing should then be ratiometric?

 

[crutschow]

Could that be used to an advantage for simplicity, by taking the display reference voltage from a resistive divider fed from the top of the resistor bridge?

The whole thing should then be ratiometric?

If the display has an external reference voltage input.
But all that would do is eliminate the inexpensive voltage reference, U2

 

[rjenkinsgb]

It should work with Dan's far simpler circuit?

 

[crutschow]

It should work with Dan's far simpler circuit?

It's far simpler because it's a conceptual circuit which doesn't show all the additional circuitry needed to generate the required offset and gain functions with reasonable accuracy and stability.
Note that he asked " Need help figuring out the circuits! ".

 

[Danwvw]

Thanks, Just the sort of things I was wondering. Years and Years ago I used AD590 Chipped probes to measure things like Water Temp, Air Temp at Work and the Oil temp on my car by modifying offset and gain on a standard common mode rejection interface circuit board that was design to Output -10 to +10 volts that represented temperature in the range of -100F to 100'F, Can't remember any of the specifics Though. I still have a 590 probe somewhere, I should give it a try, Not sure it will be feasible though as it's a pretty high impedance transducer, I calculated it to be in the 22K ohm range at Zero Degrees F and chances are the oil lite wire will have too much leakage to ground to use it without a driver at the engine. Kind of wanted to keep it real simple though:
Thanks crutschow for the Schematic, I was thinking of using a (Low Dropout 5 volt regulator) if regulation was needed and seeing if the oil pressure light would still operate on the feed side. I do really like the idea of separating the functions though.
Thanks rjenkinsgb for the input.
It would be acceptable to add a power wire to the circuit at the engine too as key switched power is available there on the ignition coil.

 

[Diver300]

A bridge circuit with a constant current element is a really bad idea.

The idea of a bridge is that it is balanced when the ratios of resistance are the same, no matter what the supply voltage is. That will never work with a constant current element, because supply voltage variation will unbalance it.

The big advantage of a constant current element is that the supply voltage doesn't matter. The obvious way to have that work is to have a load resistor, and the voltage across that is proportional to the current, and therefore proportional to the absolute temperature.

You need to offset to 0 °C, or 0 °F, but for that you need a fixed voltage, not a fixed fraction of the supply voltage. You should have a voltage reference diode and a potential divider instead of the 1 kOhm resistor. Something like a LM385BZ-1.2G would do.

I like the idea of sharing the wire with the oil pressure switch. That is neat.

 

[crutschow]

The obvious way to have that work is to have a load resistor, and the voltage across that is proportional to the current, and therefore proportional to the absolute temperature.

In my post #2 circuit, that is the feedback resistor R1.

 

[Danwvw]

View attachment 129055

Diver300 Where is there a constant current supply? Not one in my bridge circuit? Looking at the above circuit isn't it pretty much a bridge just drawn a little differently? The only difference I see is how it balances the bridge. Ref 2 is also part of the shunt regulator voltage control.

 

[crutschow]

Where is there a constant current supply?

He's referring to the temp sensor, which generates (sinks) a constant-current signal.

Looking at the above circuit isn't it pretty much a bridge just drawn a little differently?

It's pretty much of a stretch to call my circuit a bridge in the conventional electronic sense of the term.
One side is a reference voltage, and the other side is an op amp configured as a transimpedance amp (current-to-voltage converter) with the constant-current signal source.

Typically all elements in a bridge consist of resistive elements.

 

[Danwvw]

Yeah I see what you're saying but I tend to think outside of the Electronic Engineering Terminology box mostly because I don't know it as well as you do. However I still see a bridge in it's simplified resistor equivalent circuit. The AD590 just becomes a 22.5K ohm resistor at Zero F in the equivalent bridge circuit.

 

[crutschow]

The AD590 just becomes a 22.5K ohm resistor at Zero F in the equivalent bridge circuit.

No, the AD590 is a very high impedance current source whose output current changes very little with voltage.
That's not the same as a resistor.

And the 6484 (-) minus input is a virtual ground with an essentially zero ohms impedance.

So if you still want to call that a bridge, that's your prerogative, but I doubt that many others will agree with that.

 

[Danwvw]

Yeah, But just do I=E/R and that is it's equivalent resistance! Whatever, Glad you're keeping this straightened out.
Ok yes I finally see what you guys are saying, I did more math: Looks like the supply voltage does needs to be regulated for the AD590 to be linear.

 

[crutschow]

Yeah, But just do I=E/R and that is it's equivalent resistance! Whatever, Glad you're keeping this straightened out.
Looks like the supply voltage does needs to be regulated for the AD590 to be linear.

Yeah, no.
The equivalent resistance of a constant-current device is not the DC voltage divided by the DC current, it's the ΔV / ΔI (change in current versus change in voltage).
That's a measure of the "stiffness" or how close it is to the ideal (infinity resistance) constant-current device.

For example the AD590 data sheet shows a ΔI / ΔV (Power Supply Rejection) between 4V and 5V of 0.5µA/V. That means its equivalent resistance is 1/ 0.5µA = 2 megohm.
Thus the AD590's voltage does not have to be precisely regulated for it to be linear (change in current versus temperature) although it does slightly improve accuracy.

The regulation I added was mainly to keep the offset current through R2 stable.
Regulating the AD590 voltage was required to keep the voltage across R2 constant (and thus its current), but it wasn't necessarily needed for the AD590 accuracy.

 

[Danwvw]

OK so if i used a bridge to balance the 0 Degrees Fahrenheit at 5 volts supply would I need to be in the 20K ohm area or the 2M ohm area for the resistor to match it.
As 2M ohm is likely going to be really noisy! Sounds like what I need to use an AD590 is to have the Op Amp located in the engine compartment.

 

[crutschow]

OK so if i used a bridge to balance the 0 Degrees Fahrenheit at 5 volts supply would I need to be in the 20K ohm area or the 2M ohm area for the resistor to match it.
As 2M ohm is likely going to be really noisy! Sounds like what I need to use an AD590 is to have the Op Amp located in the engine compartment.

You don't match the resistor value to the AD590 equivalent, you would use a resistor to balance the bridge at it's nominal room temperature current (so about 20k ohms).

But as I stated before, and Diver300 noted, a bridge won't work well with a constant current device for what you want to do.
If you have a NIH syndrome and don't want to use my circuit that's fine; continue down the path to the bridge and have fun.
But I'm not really interested in helping you try to make a bridge circuit work where it's not the best or simplest way to do what you want.

 

[rjenkinsgb]

I did not notice in the original "bridge" that it was an IC rather than a thermistor.
As crutshaw says, a bridge is not at all suitable for that type of sensor.

For a constant current transducer IC, all you need to extract the value is a single resistor in series, and measure the voltage across that with a differential amplifier.

Supply voltage changes should be pretty irrelevant; just a decoupling cap at the power source should be sufficient.

If you use an appropriate value for the measuring resistor so the voltage across that never exceed eg. 0.5V, you can just bypass it with a rectifier so when the oil pressure switch closes, the lamp gets near enough full current with no harm to anything.

 

[crutschow]

Supply voltage changes should be pretty irrelevant; just a decoupling cap at the power source should be sufficient.

But you need a stable voltage to generate the output signal offset the TS desires.

 

[JimB]

At the very beginning of this thread Danwvw said:

I was thinking of a way to see my engine temperature without having to thread a wire to the engine.

Since then, all sorts of convoluted schemes have been proposed to combine a temperature measurement with the oil pressure light, using one wire.

Now, may be I am being a bit thick here and have missed some vital point, but the bridge, OP-AMP and display unit are all in the engine compartment with the AD590 and the pressure switch.
So what is the advantage here?
Is it so difficult to run a wire or two from the engine compartment and make a proper job rather than some ??? solution.

JimB

Edit:
OK I have just looked back at the circuit and see what is going on and yes it may just work.
But my opinion is "just run a second wire".

JimB