Maker Pro
Maker Pro

Op Amp with a/c output

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Sorry to bang on, but I'm still having a problem understanding the logic behind this. Correct me if/where I'm wrong :
1) If the ISS sensor output is only 1.5VAC at idle, it (or its connection) is faulty.
2) If you build a circuit which has output pulses whose negative-going excursions go at least abou 7V below the battery +ve voltage you can clear the error light.
3) The visual test will pass.
3) The inspector will then proceed to do the emissions test.
4) The test is done at more than one input shaft speed.
5) Your faulty ISS will be in use for the emissions test.
6a) The built circuit is also in use during that test. If the test is passed you have cheated the test.
OR
6b) The built circuit is not used for the emissions test. The error light will come back on. The test is failed.
6c) Cheating the test is illegal where you are.

Note: Encouraging unlawful activity is forbidden by the Terms of Service of this site.
As I understand it, he only wants to amplify the signal from the sensor because it is inadequate. Since it is the frequency of the signal, not the amplitude that the ECM is using, this should make no difference in the operation of the engine, and the emission testing would have the same results as it would with a sensor that had adequate output. So, perhaps technically cheating, but in reality, the engine is doing what it should and not producing any extra emissions, so not cheating by the spirit of the law.

Bob
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
Since it is the frequency of the signal, not the amplitude that the ECM is using, this should make no difference in the operation of the engine, and the emission testing would have the same results as it would with a sensor that had adequate output.
In that case, I think this circuit could work :
SensorAdapter-b.PNG

When Q1 switches on, C2 acts as a bootstrap to keep the opamp supply up.
 

HomerDodd

Sep 6, 2017
53
Joined
Sep 6, 2017
Messages
53
Harold: can you give me a schematic of how to build/install this circuit? Thanks.

Thank you very much. A schematic or diagram would be such a great help to me, so I can buy the materials, and build it. Much appreciated !
 

HomerDodd

Sep 6, 2017
53
Joined
Sep 6, 2017
Messages
53
As I understand it, he only wants to amplify the signal from the sensor because it is inadequate. Since it is the frequency of the signal, not the amplitude that the ECM is using, this should make no difference in the operation of the engine, and the emission testing would have the same results as it would with a sensor that had adequate output. So, perhaps technically cheating, but in reality, the engine is doing what it should and not producing any extra emissions, so not cheating by the spirit of the law.

Bob

Exactly right Bob. The transmission otherwise runs normally. The sensor just needs amplification. The car otherwise has clean emissions. No tampering with the system involved here.
 

HomerDodd

Sep 6, 2017
53
Joined
Sep 6, 2017
Messages
53
In that case, I think this circuit could work :
View attachment 36487

When Q1 switches on, C2 acts as a bootstrap to keep the opamp supply up.

WOW !! Thanks for the details. A few questions: which pin do I input the supply voltage from the car? Is Q1 a transistor? Is UA1 and op amp LM339? Are those values below the capacitors exactly what I need to ask for at the store, along with all the diode numbers? Thanks again, this is an immense help.
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
The circuit I posted was designed in connection with a different thread (concerned with emulating a high-side Hall sensor) and for an ECU having a pull-down input resistance of around 100 Ohms. If your PCM doesn't have such an input arrangement then the circuit might not do what you want. Also, depending on how your car does its initial checks when you turn on the ignition, it may not recognise the circuit as a fault-free inductive sensor so would throw an error signal. There is also a risk that the low resistance path created when transistor Q1 switches on could damage your PCM. Putting a 1k resistor in series with the wire going to the 'V' terminal should avoid that risk.
 
Last edited:

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
Although adding the 1k series resistor as mentioned above would avert the risk, it would also prevent the circuit doing what you want if the PCM has a low resistance input and is intended for a high-side switching Hall sensor.
Unfortunately, the input characteristics of your PCM are a mystery :(.
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
Here's an alternative, much simpler circuit which I think stands a better chance of working with your setup. It does, however, require a ground (chassis) connection. Inputs 1 and 2 are where the sensor connects.
R4 emulates the sensor coil resistance. D3 prevents Q1 from shorting terminal V to ground and also prevents reverse current if S and V got accidentally transposed.
The 1k trimpot is for setting the optimum transistor bias position to maximise the circuit sensitivity.
Q1 could be almost any low power NPN transistor.
Sensor Amp.PNG
If anyone wants to play with this circuit I've attached the LTspice sim file.
 

Attachments

  • SensorAdaptor9.asc
    3.1 KB · Views: 132

HomerDodd

Sep 6, 2017
53
Joined
Sep 6, 2017
Messages
53
Here's an alternative, much simpler circuit which I think stands a better chance of working with your setup. It does, however, require a ground (chassis) connection. Inputs 1 and 2 are where the sensor connects.
R4 emulates the sensor coil resistance. D3 prevents Q1 from shorting terminal V to ground and also prevents reverse current if S and V got accidentally transposed.
The 1k trimpot is for setting the optimum transistor bias position to maximise the circuit sensitivity.
Q1 could be almost any low power NPN transistor.
View attachment 36515
If anyone wants to play with this circuit I've attached the LTspice sim file.


Wow again!! Thanks so much for the help and explanation. Sorry I dragged this out so long. I really needed the help from the people in this forum.
 

HomerDodd

Sep 6, 2017
53
Joined
Sep 6, 2017
Messages
53
Here's an alternative, much simpler circuit which I think stands a better chance of working with your setup. It does, however, require a ground (chassis) connection. Inputs 1 and 2 are where the sensor connects.
R4 emulates the sensor coil resistance. D3 prevents Q1 from shorting terminal V to ground and also prevents reverse current if S and V got accidentally transposed.
The 1k trimpot is for setting the optimum transistor bias position to maximise the circuit sensitivity.
Q1 could be almost any low power NPN transistor.
View attachment 36515
If anyone wants to play with this circuit I've attached the LTspice sim file.


Alec,
I am still trying different circuits. I'm making some progress. What I wanted to ask was how does this circuit amplify the low AC to a higher AC ? I always thought you could never get more out of what you put in. Is it done through capacitors? Thanks.
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
What I wanted to ask was how does this circuit amplify the low AC to a higher AC ? I always thought you could never get more out of what you put in.
The input voltage changes the base current of the transistor, which causes a bigger change in the collector current (that's the way transistors work) and hence a big change in the collector voltage. The output voltage change is bigger than the input voltage change and the power out (voltage times current) isn't less than the input power because it mostly comes from the power supply rather than from the sensor.
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
I always thought you could never get more out of what you put in.
You don't. The higher energy (voltage) at the output of the amplifier is supplied by the power source used to power the amplifier. All in all the sum of power input (from the power supply + the ac input) is more than the output power available at the ac output. The difference is lost as power dissipation (essentially heat) within the circuit elements.

Compare this to a water installation: you need comparatively little energy to open or close a tap, but the water can drive e.g. a lawn sprinkler because the power is supplied by a pump in the water supply.
 

HomerDodd

Sep 6, 2017
53
Joined
Sep 6, 2017
Messages
53
The circuit I posted was designed in connection with a different thread (concerned with emulating a high-side Hall sensor) and for an ECU having a pull-down input resistance of around 100 Ohms. If your PCM doesn't have such an input arrangement then the circuit might not do what you want. Also, depending on how your car does its initial checks when you turn on the ignition, it may not recognise the circuit as a fault-free inductive sensor so would throw an error signal. There is also a risk that the low resistance path created when transistor Q1 switches on could damage your PCM. Putting a 1k resistor in series with the wire going to the 'V' terminal should avoid that risk.


Thanks Alec. Sorry to ask so many questions, but I am convinced I/we can solve this problem. Another wrinkle to throw into the mix: I have since learned that the AC output must, of course, be at least 5.5 VAC, with initial frequency of about 500-600 Hz, both rising as the speed increases. Some of the Cadillac guys said the computer is looking for a "sine wave" from the AC. Will this circuit provide this, or can I "tweek" it slightly to achieve this? Hey, thanks again..............
 

HomerDodd

Sep 6, 2017
53
Joined
Sep 6, 2017
Messages
53
Here's an alternative, much simpler circuit which I think stands a better chance of working with your setup. It does, however, require a ground (chassis) connection. Inputs 1 and 2 are where the sensor connects.
R4 emulates the sensor coil resistance. D3 prevents Q1 from shorting terminal V to ground and also prevents reverse current if S and V got accidentally transposed.
The 1k trimpot is for setting the optimum transistor bias position to maximise the circuit sensitivity.
Q1 could be almost any low power NPN transistor.
View attachment 36515
If anyone wants to play with this circuit I've attached the LTspice sim file.

This one works fairly well, too Alec. However, it only works at harder acceleration and speed. The Check Engine lamp went off ! Maybe because the AC voltage from the sensor rises more at higher rpm. However, at lower rpm and speed, the light comes back on. Any way to get the gain amplitude higher at lower speed? Maybe a resistor, capacitor or trim pot change or "tweak".......Thanks man. We are making some progress.
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
Try replacing D3 with a 100uF >=25V-rated capacitor, cap negative terminal connected to Q1 collector.
 

HomerDodd

Sep 6, 2017
53
Joined
Sep 6, 2017
Messages
53
Try replacing D3 with a 100uF >=25V-rated capacitor, cap negative terminal connected to Q1 collector.

I will. Thanks so much Alec. I could never have figured this out on my own. Still a novice, but getting better. When that light when off during steady driving, I was so thrilled !!
 

HomerDodd

Sep 6, 2017
53
Joined
Sep 6, 2017
Messages
53
I will. Thanks so much Alec. I could never have figured this out on my own. Still a novice, but getting better. When that light when off during steady driving, I was so thrilled !!

Just wondering if a sine wave is still being produced? I'm beginning to think that the cars computer (ECM) is looking for a sine wave with 500-600 frequency and rising with speed.
 
Top