Need help with reversing output voltage (NOT polarity)

Status
Not open for further replies.

Nossie

Newbie level 2
Joined
Mar 19, 2014
Messages
2
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Visit site
Activity points
24
Hi, I have got a problem on a current project and hoping some of you might help.

I am installing an LPG system in my vehicle (92 Subaru Liberty).
The LPG system has a controller that connects to various sensors and regulates mixture of LPG to air based on several parameters.
One of these sensors is the TPS (throttle position sensor). The LPG system is expecting 0.5VDC on closed throttle and 5VDC on open.
My Subaru however is doing the opposite - 5VDC at throttle position closed and 0.5VDC at throttle fully opened.
Another thing might worth mentioning is that the LPG controller has only one wire coming out to the
TPS, which would suggest that it is using its own earthing and not sharing the TPSs.
The TPS has 4 wires - Earth, 5VDC, signal on throttle operation (0.3VDC to 5VDC as soon as the throttle is operated) and signal - which is what the LPG controller is connecting to.

Is there any way I can deliver the expected voltage to the LPG controller based on the voltage readings of my TPS? I am after the simplest way of doing so.
I am not very strong in electronics so please do not get too technical

Thanks in advance!
 

I think a simple transistor inverter interface board will solve your problem. Select the transistor according to your current requirement. Good luck. If it helps, click.
 

Looking at what you have said about the voltage ranges you need, you can see it requires a level shift AND a small change in gain, not an inversion.

The range of the Subaru control is -5V (throttle closed) to +0.5V (open) = 5.5V.

The range of the throttle sensor is +0.5V (closed) to +5V (open) = 4.5V

The Subaru range, if shifted positive by 5.5V, would put the throttle closed position at +0.5V suitable for the throttle sensor. That same shift would put the open position, previously +0.5V, up to 6V, which is somewhat over the top for the throttle sensor

Using a very simple op-amp circuit, called a non-inverting summing amplifier, you can apply bias, to give you the shift, up one input, and offer the available Subaru range up the other input. The gain can be set by the op-amp feedback resistor to make the output range become 4.5V. A diode with resistor can add a limiting feature to prevent the output going higher than might be safe.

This is in control of LPG fuel - so take some precautions. Use a small 5V regulator so supply the whole circuit, with it's input protected by a zener diode with resistor to limit the maximum that can ever come out of it, and choose the op-amp to be a type that operated right near the positive power input (called "rail-to-rail"). Use a polyfuse, a low-cost self-resetting component chosen to limit the current to the circuit. Add a varistor overvoltage surge protector (cost pennies!). Do not put up with failure in the supply. Use over-specified filter capacitors and some kind of choke to
keep transients out. Filter the output from ignition RF (3 small components + a resistor). It goes in a metal box, and give attention to the temperature. It does not have to live bolted to a hot engine part or suffering vibration.

All the above would fit on a couple of postage stamps. There are bound to be other ideas out there, and if this thread is left to itself for a few more hours, you might start getting suggestions. It is understood you may not have much electronics knowledge, but this is your chance to gain some, because a operational amplifier of this sort can be put together by a hobby enthusiast, or school class exercise. If you get no other help, I will try and make a quick sketch.

P.S. It occurs to me that we need to get clear which end is at the iNPUT.
Is it the throttle sensor that has to deliver to the Subaru? .. or is it Subaru voltage that has to be changed to deliver to the sensor?

This immediately raises the question, where does the negative voltage come from? Surely the Subaru does not have its positive battery pole connected to the chassis! ??
 

Attachments

  • Handbook of OPerational Amplifier Applications.pdf
    2 MB · Views: 94
Last edited:

I guess the OP doesn't mean -5 V. It's a hyphen rather than a minus sign. A -1 amplifier with about 2.75 V refernce should do.
 

I guess the OP doesn't mean -5 V. It's a hyphen rather than a minus sign. A -1 amplifier with about 2.75 V refernce should do.
What can I say? I already posted before it occurred to me that there was a quite unlikely condition in there.

It shows the need for absolute clarity, and no ambiguity when choosing words .. or typing.
Not a minus sign changes the whole scene to require an inversion.
Either way, a very simple amplifier will do. Thanks for spotting what I should have earlier.
 

I think a simple transistor inverter interface board will solve your problem. Select the transistor according to your current requirement. Good luck. If it helps, click.

a transistor in CE configuration would do it

http://electronics.stackexchange.com/questions/30238/how-to-invert-a-digital-signal
www.downloads.reactivemicro.com/Public/.../Transistor Circuits.doc‎


http://www.electronics-tutorials.ws/amplifier/amp_2.html
http://www.talkingelectronics.com/Download eBooks/Principles of electronics/CH-10.pdf
**broken link removed**
 
Last edited:

I think a rail-to-rail opamp should be used because a transistor has a low input impedance especially if it has negative feedback from its output to its input to reduce its voltage gain to be "inverting one".
 
Reactions: FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
I think a rail-to-rail op-amp should be used because a transistor has a low input impedance especially if it has negative feedback from its output to its input to reduce its voltage gain to be "inverting one".
Absolutely agreed! This job is not like inverting an audio signal. It is an instrumentation task that requires stable, accurate, predictable inversion gain in a device that is in control of a moving vehicle! It can do the equivalent of stomping on the accelerator if it goes into runaway.

Nor should this be a race to the bottom to see if it "can be done with one transistor". Use a op-amp with a robust, protected output, and enough current to drive the impedance of the gadget it gets connected to; or, if needs be, use a transistor on the output of the op-amp, but inside the feedback loop. An op-amp allows the operation to depend only on fixed resistor values. The design attention that should go into limiting the supply, forcing fail-safe output, protecting from transients, vibration etc. seem at least equal to the core job of contriving the right operation range.
 

Thanks all for your advice. I will look into op-amp setup.

I will post back once its all set up.

Thanks again!
 

Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…