Voltage Sensing Circuit for Low Voltage Values

Thread Starter

Deve

Joined Dec 28, 2015
95
I know. Been around awhile. It is ALWAYS something silly. Let me know where I went wrong! And Thanks!
 

MikeML

Joined Oct 2, 2009
5,444
Look again at post #22.

You might need to do something with the inputs of the unused half of the LM358; floating is bad.
 

Thread Starter

Deve

Joined Dec 28, 2015
95
I do not know why this changed, but the sender's spread is 100mv to 3.5v. So that may be part of it. I told you before, if I turn it around, it doesnt fix it. I even used a new TL431 to do it in case I burned it out.
 

MikeML

Joined Oct 2, 2009
5,444
I do not know why this changed, but the sender's spread is 100mv to 3.5v. So that may be part of it. I told you before, if I turn it around, it doesnt fix it. I even used a new TL431 to do it in case I burned it out.
Turning it around wont fix the miswire. Ref and Cathode are tied together to create the 2.495V node. Anode is tied to ground.
 

Thread Starter

Deve

Joined Dec 28, 2015
95
Ok, so now there is 2.48 volts where its supposed to be. The potentiometer acts the same as it did before. Far left I get a light if the sender is very near the end of its cycle. Moving the sender off empty causes the light to go out. No adjustment. I must have something else wrong too.
 

MikeML

Joined Oct 2, 2009
5,444
Ok, so now there is 2.48 volts where its supposed to be. The potentiometer acts the same as it did before. Far left I get a light if the sender is very near the end of its cycle. Moving the sender off empty causes the light to go out. No adjustment. I must have something else wrong too.
Set the damn pot near mid scale and leave it alone. Then change the sender voltage. Once set, the pot never moves again.
 

Thread Starter

Deve

Joined Dec 28, 2015
95
RIght. That's what you would expect to happen. Far left the light comes on IF the sender is all the way down to 400mv or less. I want to be able to move the sender to anywhere on the scale while adjusting the pot and get the light to come on anywhere I want. The Pot right now has almost no affect.
 

Thread Starter

Deve

Joined Dec 28, 2015
95
One of the pics has a picture of the whole setup. The sender is not adjustable other than where it sits in the tank between empty and full. I want the light to come on when it approaches empty. That is a determination that would be made by the end user. So the pot should adjust where that light comes on. Right?
 

MikeML

Joined Oct 2, 2009
5,444
One of the pics has a picture of the whole setup. The sender is not adjustable other than where it sits in the tank between empty and full. I want the light to come on when it approaches empty. That is a determination that would be made by the end user. So the pot should adjust where that light comes on. Right?
Yes, but the setting procedure would be this:

Drive the car until the gauge reads the critical fuel level (the point where you want the LED to come on). Tweak the pot so that the light just comes on, goes off when the pot is moved ever-so-slightly clockwise and then counter-clockwise. Leave the pot alone from that point. Fill the tank; the light should be off. Drive; as the fuel is used the led should go on as the fuel level decreases below the original critical fuel level. You are making it harder than it needs to be...
 

Thread Starter

Deve

Joined Dec 28, 2015
95
I am saying there is no adjustment when you adjust the pot, NOTHING happens for 3/4 of the turn. The only time the light comes on is the first 100 to 400 mv. That is not adjustable either. That is just where the light comes on. What is wrong in my circuit that could cause this? Again, the voltages I am getting from the sender is 100mv -3.5 volts. It just happens that the resistance at ONE point on the Pot is sufficient to turn on the light. No adjustment whatsoever.
 

MikeML

Joined Oct 2, 2009
5,444
You have bad pot? Measure the voltage at the non-inverting input of the opamp while moving the pot. It should smoothly increase from zero to ~1.2V as you rotate the pot.
 

Thread Starter

Deve

Joined Dec 28, 2015
95
This is a very touchy vintage sender with very little window. What you say is happening fine. But what is not happening with your values is the sender is not lighting up the light except for when the tank is almost completely empty. If I change the Pot to a 50K pot and change the 15k resistor to 10K, I can get it adjustable from about half full (2 volts) to empty (100mv) which works for me. But I do not like to arbitrarily change values like that because who knows what other effects its having. In any case, you have gone above and beyond the call in helping others Mike. You have had patience and compassion outwardly which is all that matters. Thanks. If you run your model with those two values changed, I would like to hear what you get so there are no surprises, but thanks many times over for doing this for me. Truth is, I would like to go to a 15 turn pot and try to get more resolution. I want this LED to mean something to the driver.
 

MikeML

Joined Oct 2, 2009
5,444
This is a very touchy vintage sender with very little window. What you say is happening fine. But what is not happening with your values is the sender is not lighting up the light except for when the tank is almost completely empty. ...
Then you were wrong when you said that the highest trip point you ever wanted is 1.2V. The circuit met that spec originally.

With a 50KΩ pot and R4=10KΩ, the circuit never trips if the pot wiper is at the bottom, trips at 0.411V (red) if the pot wiper is at 20%, trips at 0.832V (org) if the pot wiper is at 40%, trips at 1.25V (yel) if the pot wiper is at 60%, trips at 1.66V (grn) if the pot wiper is at 80%, and trips at 2.08V (blu) if the pot wiper is at 100%.

170b.gif
 

Thread Starter

Deve

Joined Dec 28, 2015
95
That's perfect. Yeah, the 1.2 volts was what I thought was half full. Then going back and checking later, about 2 volts is half full. So if those changes are okay I am good to go. THANKS AGAIN!
 

Thread Starter

Deve

Joined Dec 28, 2015
95
Out of curiosity, what is the Capacitor for? and also, I am thinking you chose the 3.3 meg resistor to reduce hysterisis? Its another part that doesn't change the behavior of the circuit if its removed, but I like to know what the designer was thinking. FYI, exchanging the LM358 for the LM393 IS a bad idea. It introduces a slow transition from LED on to off.
 

MikeML

Joined Oct 2, 2009
5,444
Out of curiosity, what is the Capacitor for? and also, I am thinking you chose the 3.3 meg resistor to reduce hysterisis? Its another part that doesn't change the behavior of the circuit if its removed, but I like to know what the designer was thinking. FYI, exchanging the LM358 for the LM393 IS a bad idea. It introduces a slow transition from LED on to off.
A. Capacitor is a bypass to keep the LM358 from oscillating; also to filter spikes from the car's electrical system.
B. 3.3 meg creates a small amount ( a few mV) of hysteresis. If you get a lot of jitter in the LED due to fuel sloshing, make it lower in value, maybe 1meg.
C. The LM393 lacks a pull-up transistor; it can only pull down. You could fix it by shunting the existing LED/resistor combination with a 4.7K pull-up.
 
Top