2

I've a 6-12V dc latching solenoid, which needs 60ms of pulse to set it to on/off.

I'll be using a microcontroller to generate the pulse. I've following questions.

  1. What if the generated pulse is longer (say 10x of the specified 60 ms) Will it damage solenoid? Will it still correctly turn on/off.?
  2. I was thinking of using pwm(pulse wave modulation) technique to generate pulse. So I was thinking of using 60% duty cycle for time period of 100ms. Then after sleeping for 100ms, I would disable pwm. So there is always potential of two pulses instead of one, (or more) going out. Am I thinking on correct lines?

Adding specifications:

enter image description here

Nick Alexeev
  • 37,739
  • 17
  • 97
  • 230
Ouroboros
  • 223
  • 1
  • 8
  • Do you have a datasheet for the solenoid? What is the reason you cannot generate a shorter pulse? – evildemonic Mar 07 '19 at 16:42
  • All I know is its rated 9ohm, 6-12v and needs 60ms. I can generate shorter pulse, but I want to understand what may go wrong. I would like to manually test this out using batteries by hand, and I just want to be sure if I keep it connected for a second or so, I don't burn the coil. – Ouroboros Mar 07 '19 at 16:49
  • 3
    If 60 ms is the minimum pulse width, there should also be a maximum specified if there is a danger of overheating with longer pulses. – Peter Bennett Mar 07 '19 at 16:49
  • I see, I didn't know there is max specified pulse duration. What happens if I keep applying 60ms pulse every say 100ms? Will it burn the coil? – Ouroboros Mar 07 '19 at 16:53
  • I was looking at one recently, not a latching version but one used to unlock a cabinet door. It had a maximum activation time, after which one risked burning up the coil. 800mS. A PWM signal is overkill to create one 60mS pulse. Your microcontroller shoud be able to handle that easily. Say about 6 lines of code with an Arduino in C++ for example. – CrossRoads Mar 07 '19 at 16:53
  • Post the datasheet for yours, or we can look one up at Digikey and go from there. – CrossRoads Mar 07 '19 at 16:54
  • Crossroads, I understand using sleep, I can generate pulse using gpio. Agreed pwm sounds like an overkill. – Ouroboros Mar 07 '19 at 16:55
  • @Ouroboros 12V²/9 ohms = 16 Watts of heat and 0.1s is 1.6Joules of energy which "May" overheat the fine wire. for 100ms within a short term repeatedly . The holding current can be reduced to few % if not latching. THe duty cycle must be low like <20% is my guess for rapid flips. These specs are usually minimum to activate with no load or max duration due to temp rise – Tony Stewart EE75 Mar 07 '19 at 16:58
  • Crossroads, added tech specs – Ouroboros Mar 07 '19 at 17:04
  • Thanks sunnyskyguy. I understand. So I believe trying to test it "manually" is a sure shot way to kill the valve. Since I'll be putting it under batteries for at least half a second, enough to heat the coils to limit. Right? – Ouroboros Mar 07 '19 at 17:05
  • @Ouroboros Just looking over the image you provided, I'm still trying to find the 60 ms you mentioned. I see 20 and 27 with two different voltages given. But not 60. Is 60 ms the maximum pulse width specification written out somewhere else? In any case, it's easy to set up a simple circuit to provide an appropriate pulse width without any timing required by the MCU. All it would need to do is toggle a pin to active and then inactive as fast as it can do it. – jonk Mar 07 '19 at 19:43
  • @jonk Thanks for your observation. I would like to know if code like so is okay? 1. Turn GPIO ON 2. Sleep 30ms 3. Turn GPIO OFF – Ouroboros Mar 08 '19 at 09:26

1 Answers1

0

What if the generated pulse is longer (say 10x of the specified 60 ms) Will it damage solenoid? Will it still correctly turn on/off.?

No it will actuate again if the timing is not followed, so make sure the timing and voltage are followed properly.

The electrical current applied to a latching valve is very brief, typically in pulses of 20 to 50 milliseconds. The device's electrical control system must control these pulses precisely. If they are too short, the valve will not respond and if they are too long the valve will open and then close again or vise versa. This is commonly referred to as re-latching. The voltage must also be stable not varying more than +10% or -15% of rated voltage.

Source: https://www.solenoidsolutionsinc.com/custom-solutions/latching-solenoid-valves--low-energy/

I was thinking of using pwm(pulse wave modulation) technique to generate pulse. So I was thinking of using 60% duty cycle for time period of 100ms. Then after sleeping for 100ms, I would disable pwm. So there is always potential of two pulses instead of one, (or more) going out. Am I thinking on correct lines?

A better way is to use a hardware timer on the microprocessor (which is available on almost any processor) to run the GPIO's. If that can't be done then use a loop and lock out interrupts while using the loop to avoid timing issues.

Voltage Spike
  • 75,799
  • 36
  • 80
  • 208