Looking for ideas for sharp turn on from analog voltage.

Thread Starter

mbohuntr

Joined Apr 6, 2009
444
Good Morning!
I have built a 12V battery powered solar circuit to water our garden while we are away on vacation. It consists of a 555 set up as a one shot, triggered by a 2n7000 mosfet. It will hold open a solenoid valve long enough to fill a 5 gallon pail from a storage tank. The pail then gravity drains into drip tubes.

I can trigger it with switch contacts easy enough, and it works well. Here's the rub... I don't want a digital arduino type set-up. And I cannot use an op-amp because of the need to pull down to the ground rail from a single supply. The mosfet would never turn off. I cannot use a bjt transistor because of the analog solar voltage. No sharp turn on.
I cannot use a relay because of the variable solar panel voltage output being too high for a 5-10 volt coil, and a 12volt coil would retrigger when clouds go by.
The only two solutions I can come up with is a 7805 voltage regulator powering the relay coil, or use a 556 dual timer to use the second timer as a comparator and it would sink current when low, shutting off the mosfet.
Does anybody have any better ideas?
Thanks
Mike
 

Attachments

Thread Starter

mbohuntr

Joined Apr 6, 2009
444
Good Morning!
I have built a 12V battery powered solar circuit to water our garden while we are away on vacation. It consists of a 555 set up as a one shot, triggered by a 2n7000 mosfet. It will hold open a solenoid valve long enough to fill a 5 gallon pail from a storage tank. The pail then gravity drains into drip tubes.

I can trigger it with switch contacts easy enough, and it works well. Here's the rub... I don't want a digital arduino type set-up. And I cannot use an op-amp because of the need to pull down to the ground rail from a single supply. The mosfet would never turn off. I cannot use a bjt transistor because of the analog solar voltage. No sharp turn on.
I cannot use a relay because of the variable solar panel voltage output being too high for a 5-10 volt coil, and a 12volt coil would retrigger when clouds go by.
The only two solutions I can come up with is a 7805 voltage regulator powering the relay coil, or use a 556 dual timer to use the second timer as a comparator and it would sink current when low, shutting off the mosfet.
Does anybody have any better ideas?
Thanks
Mike
I just had an idea. What if I removed the mosfet, went with a negative pulse from a 741 op-amp comparator? Would that drop pin 2 quick and low enough to trigger it?
 

ericgibbs

Joined Jan 29, 2010
16,803
hi mbo,
This image shows how to generate a low going pulse on pin #2, Trigger Input.

BTW: You can get OPA's that will switch close to Vout= 0V, even when powered from a single supply

E
EG 1470.gif
 

eetech00

Joined Jun 8, 2013
3,418
Good Morning!
I have built a 12V battery powered solar circuit to water our garden while we are away on vacation. It consists of a 555 set up as a one shot, triggered by a 2n7000 mosfet. It will hold open a solenoid valve long enough to fill a 5 gallon pail from a storage tank. The pail then gravity drains into drip tubes.

I can trigger it with switch contacts easy enough, and it works well. Here's the rub... I don't want a digital arduino type set-up. And I cannot use an op-amp because of the need to pull down to the ground rail from a single supply. The mosfet would never turn off. I cannot use a bjt transistor because of the analog solar voltage. No sharp turn on.
I cannot use a relay because of the variable solar panel voltage output being too high for a 5-10 volt coil, and a 12volt coil would retrigger when clouds go by.
The only two solutions I can come up with is a 7805 voltage regulator powering the relay coil, or use a 556 dual timer to use the second timer as a comparator and it would sink current when low, shutting off the mosfet.
Does anybody have any better ideas?
Thanks
Mike
So..you've told us all the things you don't want.
Tell us what YOU DO want. :)
 

crutschow

Joined Mar 14, 2008
31,128
Your requirements are not completely clear, but it appears that you want the solar panel to trigger the circuit once, but not when a cloud passes.
For that, it would seem you need a timer, triggered by the solar panel voltage, to generate only one pulse per day.
Is that correct?

Or does the cloud cover still generate some voltage that can be detected?
In that case, a comparator (e.g. LM339/393) to detect a voltage above the minimum solar panel voltage to trigger the 555 should work.
The comparator works to ground level, and will give a sharp transition that can directly trigger the 555, no MOSFET needed.
 
Last edited:

DickCappels

Joined Aug 21, 2008
9,513
Taking a 90° turn, as for using the relay because the voltage can get too high: Your relay can probably safely be driven my more voltage than you think, after all it is only copper wire and iron. If you are still concerned, you can use a two transistor current limiter in series with your resistor. In the circuit below the current will me limited to 0.7V/Rsense.
1647193617675.png
 

Thread Starter

mbohuntr

Joined Apr 6, 2009
444
Your requirements are not completely clear, but it appears that you want the solar panel to trigger the circuit once, but not when a cloud passes.
For that, it would seem you need a timer, triggered by the solar panel voltage, to generate only one pulse per day.
Is that correct?

Or does the cloud cover still generate some voltage that can be detected?
In that case, a comparator (e.g. LM339/393) to detect a voltage above the minimum solar panel voltage to trigger the 555 should work.
The comparator works to ground level, and will give a sharp transition that can directly trigger the 555, no MOSFET needed.
Texting is hard to communicate myself clearly. I apologize for the ambiguity. You are correct, once per day.
I was looking to use the rising voltage on the solar charger panel to trigger the one shot. Either by positive or negative voltage pulse. That only occurs once a day if the threshold is low enough. Too high a threshold and the solar panel swings from 11V to 21V with passing clouds and would trigger the water too often. A comparator would be ideal, except it won't sink the trigger capacitor using only a single supply .
Yes, You're correct! now that I think about it, a lm339 SINKS current! Thank you!
 

Thread Starter

mbohuntr

Joined Apr 6, 2009
444
Taking a 90° turn, as for using the relay because the voltage can get too high: Your relay can probably safely be driven my more voltage than you think, after all it is only copper wire and iron. If you are still concerned, you can use a two transistor current limiter in series with your resistor. In the circuit below the current will me limited to 0.7V/Rsense.
View attachment 262657
Whoa... I think I get it... as the voltage rises across R1, T1 turns on slowly, that saturates T2, but the current is limited by R1? What about the sense resistor? How do you calculate the maximum current? By the peak anticipated voltage?
 

k1ng 1337

Joined Sep 11, 2020
701
Any reason in particular for not wanting a microcontroller? An afternoon coding an Arduino or Pi Pico would work nicely with plenty of available IO for other work.
 

Thread Starter

mbohuntr

Joined Apr 6, 2009
444
Any reason in particular for not wanting a microcontroller? An afternoon coding an Arduino or Pi Pico would work nicely with plenty of available IO for other work.
I have done some small things with an Arduino, and it was OK, I had an idea I wanted to make some stand alone boards, but I got bored with it.
 

crutschow

Joined Mar 14, 2008
31,128
Here's my take on using a comparator to trigger the 555 (LTspice simulation below).
The input resistor values are selected so the comparator output goes low (red trace) to trigger the 555 (yellow trace) as the solar panel (green trace) goes above 5V.
R7 provides about 1.7V of hysteresis at the trigger point to minimize the chance of more than one trigger, thus the input voltage must go below about 3.3V before the 555 can again be triggered.

I used the timing 555 RC values in your post but I assume you need more than 2 seconds to fill the pail.

1647228155873.png
 
Last edited:

Thread Starter

mbohuntr

Joined Apr 6, 2009
444
Here's my take on using a comparator to trigger the 555 (LTspice simulation below).
The input resistor values trigger the 555 as the solar panel goes above 5V.
R7 provides about 1.7V of hysteresis at the trigger point to minimize the chance of more than one trigger.
I used the timing 555 RC values in your post but I assume you need more than 2 seconds to fill the pail.

View attachment 262671
Yes! that's exactly right! Thank you very much for all the work!
 

DickCappels

Joined Aug 21, 2008
9,513
Whoa... I think I get it... as the voltage rises across R1, T1 turns on slowly, that saturates T2, but the current is limited by R1? What about the sense resistor? How do you calculate the maximum current? By the peak anticipated voltage?
In the circuit the current will me limited to 0.7V/Rsense.

If you want to limit the current to 50 ma (for example) the resistor would be 0.7V/.05 = 14Ω
 
Top