Detect sharp voltage increase against slow light change

Thread Starter

Roars

Joined Apr 23, 2022
8
Hi to all, I found this forum by chance whilst was looking for ideas for a project I have in mind . I have a set of daylight running lights working fine using one of these cheap voltage controller (eBay) IE on and off just detecting battery voltage. I have now fitted a solar panel which obviously even when the van if off the voltage on the battery ( during solar charging) can and will increase over the threshold so the lights turn on . I have been thinking how I could detect a sharp voltage increase ( IE engine on) but ignore a slow voltage increase. I know I could use a component like Arduino and with that I could do it just measuring voltage increase against time etc but I wanted to see if was possible to do it without such a components , something simpler ,but so far not come up with any ideas . Anyone would have any though ? Thank you .
 

DickCappels

Joined Aug 21, 2008
10,152
Short answer:
Connect all of the six Schmitt triggers in a HEF40106B or similar hex inverter with Schmitt Trigger inputs in series. Put the input from your sensor into the input of the first in string of Schmitt tiggers and take the output from the last.

Background:
I did that in about 1973. We planted a radio beacon in foothills surrounding our valley and because of heavy skip interference during the day wanted the transmitter to kept off during the day, thereby greatly extending the battery life.

The sensor was a photo resistor and a series dropping resister. I fed the signal from the sensor to the input of a Schmitt trigger to make the change in light cause a sharp transition. It worked great on the workbench, so off we went into the foothills to bury most of the transmitter and and run the antenna up a redwood tree.

The next day the transmitter did not turn off. Back up into the foothills, retrieved the transmitter and back on the bench connected five of the six Schmitt triggers in the IC package in series (as I recall it was a cd. It seemed to work, so back up into the foothills and buried the transmitter the same place, hung the wire antenna and went back to the valley. For months afterwards the transmitter came on in the evening and off at sunrise. It became a fixture at that frequency sending ".... .." in CW for months.
 

Thread Starter

Roars

Joined Apr 23, 2022
8
Short answer:
Connect all of the six Schmitt triggers in a HEF40106B or similar hex inverter with Schmitt Trigger inputs in series. Put the input from your sensor into the input of the first in string of Schmitt tiggers and take the output from the last.

Background:
I did that in about 1973. We planted a radio beacon in foothills surrounding our valley and because of heavy skip interference during the day wanted the transmitter to kept off during the day, thereby greatly extending the battery life.

The sensor was a photo resistor and a series dropping resister. I fed the signal from the sensor to the input of a Schmitt trigger to make the change in light cause a sharp transition. It worked great on the workbench, so off we went into the foothills to bury most of the transmitter and and run the antenna up a redwood tree.

The next day the transmitter did not turn off. Back up into the foothills, retrieved the transmitter and back on the bench connected five of the six Schmitt triggers in the IC package in series (as I recall it was a cd. It seemed to work, so back up into the foothills and buried the transmitter the same place, hung the wire antenna and went back to the valley. For months afterwards the transmitter came on in the evening and off at sunrise. It became a fixture at that frequency sending ".... .." in CW for months.
Hi thanks for your reply, but if understand correctly the antenna came on and off following the light what i need its something that detect a fast change of voltage (ie from 12.5 (engine off) to lets say 13.7 (engine on) ) but wont trigger when the voltage gradually increase. I think the circuit will have to have some sort of counter to measure how long it took for the voltage to go from min to max ...any way thank you for your input really appreciate it.
 

DickCappels

Joined Aug 21, 2008
10,152
The counter is one way of doing it. How about a simple differentiator?

1650702748190.png
Select the input capacitor for the change in voltage vs time and change the feedback capacitor to filter out noise from the electrical system. The ratio of the input to output resistors sets the gain for the transition. You can then pass the output of the differentiator through a Schmitt trigger if a very fast transition is needed. If you use a dual opamp the second amp in the package can be made into a Schmitt trigger.

1650702986108.png
 

Thread Starter

Roars

Joined Apr 23, 2022
8
The counter is one way of doing it. How about a simple differentiator?

View attachment 265645
Select the input capacitor for the change in voltage vs time and change the feedback capacitor to filter out noise from the electrical system. The ratio of the input to output resistors sets the gain for the transition. You can then pass the output of the differentiator through a Schmitt trigger if a very fast transition is needed. If you use a dual opamp the second amp in the package can be made into a Schmitt trigger.

View attachment 265647
errr thanks i think that is what i need ..although need to read it few times...and perhaps read abit more about differentiators..i see how it triggers if the voltage change is fast but how does stop the output to change when the increase in voltage is slow ? sorry might be obvious but cant see it ...and again thanks for your time and much appreciated help :)
 

DickCappels

Joined Aug 21, 2008
10,152
When the voltage change is slow there is little current flowing through the capacitor, so the output signal is small. When the voltage changes rapidly, a lot of current flows through the capacitor and the output signal is large.

This article explains the circuit in painful detail. Skip down to the section titled "Active Differentiator" for less pain : -)
https://en.wikipedia.org/wiki/Differentiator
 

Ian0

Joined Aug 7, 2020
9,674
How about a different approach, and look for the ripple on the battery voltage due to the alternator?
A cloud passing over the sun in a breeze can give a quite sudden current increase from the solar panel.
 

Thread Starter

Roars

Joined Apr 23, 2022
8
How about a different approach, and look for the ripple on the battery voltage due to the alternator?
A cloud passing over the sun in a breeze can give a quite sudden current increase from the solar panel.
I did think something about the alternator but was above...my knowledge. I think i will explore the approach suggested by DickCappels...but thanks any way. now lets go and by op amp and few bits :)
 

Thread Starter

Roars

Joined Apr 23, 2022
8
When the voltage change is slow there is little current flowing through the capacitor, so the output signal is small. When the voltage changes rapidly, a lot of current flows through the capacitor and the output signal is large.

This article explains the circuit in painful detail. Skip down to the section titled "Active Differentiator" for less pain : -)
https://en.wikipedia.org/wiki/Differentiator
Do yo think this will do ?
LM741 ?
 

Ian0

Joined Aug 7, 2020
9,674
Isn't it a good thing that it turns the lights on when the battery voltage reaches float voltage? It will prevent the battery from overcharging.
 

DickCappels

Joined Aug 21, 2008
10,152
I assume you don’t want to make a negative power supply. An LM741 cannot swing very close to either power supply rail and has a very limited common mode input range, so I would not recommend that one.

The only slightly newer LM358 dual opamp would probably work well here.

How would you use the output signal?
 

crutschow

Joined Mar 14, 2008
34,285
i tried the circuit in a simulator and it works great
But not with a 741 powered by one supply.

Below is an LTspice simulation using an LM324 / LM358 single-supply op amp in a non-inverting differentiator circuit, to get a positive output for a positive change in the battery voltage.

You may have to play with the value of C1 to get the desired output voltage for your particular application.
A larger capacitor gives a higher output voltage for a specific change in the input voltage with time.

1650735461865.png
 
Last edited:

Thread Starter

Roars

Joined Apr 23, 2022
8
But not with a 741 powered by one supply.

Below is an LTspice simulation using an LM324 / LM358 single-supply op amp in a non-inverting differentiator circuit, to get a positive output for a positive change in the battery voltage.

You may have to play with the value of C1 to get the desired output voltage for your particular application.
A larger capacitor gives a higher output voltage for a specific change in the input voltage.

View attachment 265674
whow just read msg now ......funny was trying to learn this ltspice.....thank you very much :)
 

crutschow

Joined Mar 14, 2008
34,285
Noticed that the plus op amp input could be damaged by a sudden negative change in input voltage, so added a resistor in series to minimize that problem (below):

1650837581364.png
 
Last edited:

MisterBill2

Joined Jan 23, 2018
18,176
How about a different approach, and look for the ripple on the battery voltage due to the alternator?
A cloud passing over the sun in a breeze can give a quite sudden current increase from the solar panel.
There are commercial products that do that, they have been available for years. So detecting alternator ripple is a valid method. But for an actual running light it only need to be lit when the vehicle is in gear. So just tap in to the neutral safety switch circuit.
BUT if you are afraid to touch the vehicle wiring at all, then detecting alternator ripple is the scheme to use.An LM324 quad op-amp with small value coupling capacitors, a simple diode rectifier, and one section configured as a comparator to drive a transistor to control the light switching relay. Or buy the whole package at your auto parts store.
 
Top