how to use transistor to increase resistance

Alec_t

Joined Sep 17, 2013
12,818
and I used two pins for Air conditioner.
Re post #38, since you have no air conditioner, do pins 2 and 3 of the car's sensor connect to anything at all (apart from terminals 10 and 33)?
But there is a problem. After placing this circuit, the indicated temperature also increases with increasing resistance
That suggests the simulated resistance needs to connect to the +ve supply rail rather than ground, assuming the digital gauge works by simply sensing a voltage at the sensor.
 
Last edited:

MisterBill2

Joined Jan 23, 2018
11,919
INDEED! Post #38 clearly indicates that the temperature changes of the two sensors are in opposite directions. This tells me that IF the "A/C" sensor is not used for anything else, it might be possible to arrange a scheme so that the voltage dropped as temperature rises, by feeding the sensor with a regulated positive voltage and using a resistor in the lead from the other terminal to common. The it will just be a matter of setting gain and offset to allow the digital display to indicate correctly. Of course, that is still presuming that the digital display can function with one sensor terminal tied to common.
So far there is no indication that this is true.
Thus my suggestion that the sensor provided with the digital display be used with a separate exposure to the engine coolant loop. The benefit is that there would be no need for any access to the existing temperature system at all. In addition, adequately skilled mechanical support is often much more reasonably than electronic system expertise.
Has it even been mentioned, the brand , year, and model of the vehicle that this project is for??
 

Alec_t

Joined Sep 17, 2013
12,818
... and if the in-car gauge doesn't indicate over the full temperature range, is it possible that the in-car sensor is not the OEM one originally specified for this particular car model?
 

MrChips

Joined Oct 2, 2009
26,145
If you want to use the unused sensor for A/C then we can make this work.

1645541076076.png
Measure the resistance vs temperature and let's see what you have.
 

MisterBill2

Joined Jan 23, 2018
11,919
I think that's what the TS shows in post #22?
Both of those curves show the resistance DROPPING as the temperature rises, while the A/C sensor resistance INCREASES as the temperature rises. Thus the circuit for changing variable resistance into a change in voltage will need to be quite different, although simple to make.
What is still a totally unknown detail is if the connection to the sensor for the digital display has either terminal either a voltage source or to common. So really the next challenge will be discovering what iput the digital display unit requires.
This is why, for the third time, I am suggesting an arrangement to use the temperature probe supplied with the digital display that to sense the coolant temperature independently of any other sensor. That is entirely possible for most vehicles, although I dod imagine that the TS might not be able to achieve that themself.
 

Thread Starter

saeidbb

Joined Jan 21, 2022
25
Re post #38, since you have no air conditioner, do pins 2 and 3 of the car's sensor connect to anything at all (apart from terminals 10 and 33)?

That suggests the simulated resistance needs to connect to the +ve supply rail rather than ground, assuming the digital gauge works by simply sensing a voltage at the sensor.
At least by checking what I had in the car wiring diary, I found that these two wires were not used, and in fact, two thermistors were used inside the sensor.

There are 2 questions about the proposed circuit in post#17. First, specify the pins of the LM358 in circuit. I have connected the VCC pins to +12 and the GND pins to ground or the negative of the battery.
Also, if you specify the IRF530 bases (D,G,S) on the circuit, it will be very good and I can be sure of the correct arrangement of the circuit.

You said in this post that I should connect to + VE instead of ground. Can you tell me where to connect for this?
 

Thread Starter

saeidbb

Joined Jan 21, 2022
25
INDEED! Post #38 clearly indicates that the temperature changes of the two sensors are in opposite directions. This tells me that IF the "A/C" sensor is not used for anything else, it might be possible to arrange a scheme so that the voltage dropped as temperature rises, by feeding the sensor with a regulated positive voltage and using a resistor in the lead from the other terminal to common. The it will just be a matter of setting gain and offset to allow the digital display to indicate correctly. Of course, that is still presuming that the digital display can function with one sensor terminal tied to common.
So far there is no indication that this is true.
Thus my suggestion that the sensor provided with the digital display be used with a separate exposure to the engine coolant loop. The benefit is that there would be no need for any access to the existing temperature system at all. In addition, adequately skilled mechanical support is often much more reasonably than electronic system expertise.
Has it even been mentioned, the brand , year, and model of the vehicle that this project is for??
Unfortunately, I can not do this mechanically by placing the digital thermometer sensor somewhere in the engine.

this project is for proton gent 2 2007
 

Thread Starter

saeidbb

Joined Jan 21, 2022
25
... and if the in-car gauge doesn't indicate over the full temperature range, is it possible that the in-car sensor is not the OEM one originally specified for this particular car model?
No, this is what the factory has put on it, and this problem exists in all Proton Gen 2 cars produced in different years, and it is not corrected by replacing the temperature sensor. And I said before that the temperature read by the ECU is absolutely correct, and I checked this with the ODBD2 device. This means that the car's temperature sensor is perfectly healthy and working properly.
 

Alec_t

Joined Sep 17, 2013
12,818
I found that these two wires were not used
They may well not be used, but are they actually connected to anything? In other words, is the A/C thermistor actually floating electrically or not? If not, it limits how an adapter can be designed and configured.
There are 2 questions about the proposed circuit in post #17.
The post #17 circuit won't work with your digital gauge.
Instead of us guessing the digital gauge function it would help in designing an adapter if we had more info about the internal workings of your digital gauge. As you won't be using the new thermistor that comes with it, can you pierce the insulation on the two wires connecting it (let's call them A and B) to the gauge and make a few measurements with the digital gauge powered up? It would help to know, for each of two temperatures (e.g room temperature and a temperature > 50C) the voltages (a) between A and ground (i.e the -ve terminal of the gauge supply) and (b) between B and ground.
 

MisterBill2

Joined Jan 23, 2018
11,919
If there is some connection point that you have used to verify that the existing temperature sensor is accurate, then can you also connect to those two unused wires and observe the function of the second temperature sensor? That will both verify that it is present and also verify how the resistance changes with temperature. And this without doing anything ecept connecting to the two unused wires. Then we will know if the sensor should be connected to the positive voltage source, with a resistor from the other terminal going to common, or if one side should connect to common and the other side toa resistor connected to the positive source. So it will be simple to either have a voltage that rises as temperature rises, or a voltage that decreases as the temperature rises. That is something that we need to know.

BUT it would be so very much simpler to arrange for that temperature sensor with the digital display to sense the coolant temperature where the coolant leaves the engine block and flows toward the radiator. No complex electronics and No access to the electrical wiring, except to power the digital display device.
 

Thread Starter

saeidbb

Joined Jan 21, 2022
25
They may well not be used, but are they actually connected to anything? In other words, is the A/C thermistor actually floating electrically or not? If not, it limits how an adapter can be designed and configured.

The post #17 circuit won't work with your digital gauge.
Instead of us guessing the digital gauge function it would help in designing an adapter if we had more info about the internal workings of your digital gauge. As you won't be using the new thermistor that comes with it, can you pierce the insulation on the two wires connecting it (let's call them A and B) to the gauge and make a few measurements with the digital gauge powered up? It would help to know, for each of two temperatures (e.g room temperature and a temperature > 50C) the voltages (a) between A and ground (i.e the -ve terminal of the gauge supply) and (b) between B and ground.
Well, I did almost the same thing, and instead of a digital thermometer temperature sensor, I put a variable resistor, and I wrote down the value of that variable resistor and the temperature indicated by it, which is given in post 18. Since the car's temperature sensor is made up of two separate sensors, I turned on the car and wrote down the temperature and the amount of thermistor resistance used by the car's sensor, which is given in post 18.
I did not understand how opening the digital thermometer sensor cover helps us?
 

Thread Starter

saeidbb

Joined Jan 21, 2022
25
If there is some connection point that you have used to verify that the existing temperature sensor is accurate, then can you also connect to those two unused wires and observe the function of the second temperature sensor? That will both verify that it is present and also verify how the resistance changes with temperature. And this without doing anything ecept connecting to the two unused wires. Then we will know if the sensor should be connected to the positive voltage source, with a resistor from the other terminal going to common, or if one side should connect to common and the other side toa resistor connected to the positive source. So it will be simple to either have a voltage that rises as temperature rises, or a voltage that decreases as the temperature rises. That is something that we need to know.

BUT it would be so very much simpler to arrange for that temperature sensor with the digital display to sense the coolant temperature where the coolant leaves the engine block and flows toward the radiator. No complex electronics and No access to the electrical wiring, except to power the digital display device.
In fact, I did what you said and recorded the actual temperature of the car engine using the Diag OBD2 device, and at the same time I recorded the temperature sensor resistance of those two unused wires and brought it in post 17. And with a variable resistor instead of a digital thermometer sensor, I also wrote down the values shown by the digital thermometer, and in post 17, I drew a comparison chart of these values in post 22.

But about placing the thermometer sensor in a part of the engine. This was the first thing that came to my mind and I tried to do this, but due to the fact that this sensor must be in contact with the engine coolant and near the same temperature sensor of the car itself, it was not possible. And if it is embedded outside the engine, the temperature would not be displayed correctly, and if it is installed near the car radiator, the temperature will be displayed with a delay, and the reason is that if the car's thermostat is closed, the liquid will be cold. There is no current in the radiator.
 

MisterBill2

Joined Jan 23, 2018
11,919
The engine coolant circulates and it does not lose much heat at all as it passes through the engine. So the point that I am suggesting is at the connection where it leaves the engine block to enter the hose that connects to the radiator. That point is usually where the enginethermostat is located, which controls the flow of coolant to the radiator. That fitting is usually removable, so that drilling and threading thehole for the fitting to mount the sensor can be done easily with no risk of getting metal chips into the engine. So the delay in reading will be rather short, and the error in reading will not be present.
 

Alec_t

Joined Sep 17, 2013
12,818
I put a variable resistor
But where did you put it? Was one terminal connected to ground, or was it connected to the +12V rail? We still don't know how/where the thermistor is wired inside the digital gauge. The measurements I proposed would establish that.
Show us the test circuit you used.
 

MisterBill2

Joined Jan 23, 2018
11,919
OK, now, based on the first TS paragraph in post #52, we know that the temperature versus resistance curve shown is of the actual sensor that is available for use. What we do not know, I don't think, is the temperature versus resistance requirement of the digital display. from it's sensor. If the blue plot in post #22 is from the AC sensor that is available for use, and if the other plot is the temperature indicated versus the sensor resistance requirement for the temperature display, then with a combination of series resistor and possibly a shunt resistor the curves can be made to coincide, and the TS can obtain an accurate temperature display from the digital device. We have no need to examine the inner workings of that package because the adjustments will be with the two external resistances.
The result may not be exact at the lower temperatures but usually the concern about engine temperatures is at the higher levels.
 
Top