Solenoid driver: unwanted oscillation

[crutschow]

Here's an alternative PWM approach, avoiding the need for a buck regulator

What buck regulator?

 

[alec_t]

Delete 'buck'.

what would be the advantage of using this over a 555 oscillator to drive a MOSFET gate?

None whatsoever. The aim was to avoid using a 7805 or similar regulator; so that everything runs from a single 12V source.

 

[crutschow]

The aim was to avoid using a 7805 or similar regulator; so that everything runs from a single 12V source.

Neither PWM circuit requires a 7805 regulator.
The regulator was only to provide the 5V holding voltage for the original approach, which will now be generated by the PWM.

 

[alec_t]

Understoood. I was working from the original.

 

[earckens]

Neither PWM circuit requires a 7805 regulator.
The regulator was only to provide the 5V holding voltage for the original approach, which will now be generated by the PWM.

Oh now I see too.. so, forget a 5V regulator, and use PWM after solenoid startup instead of change-over to 5V with PWM?

 

[crutschow]

Oh now I see too.. so, forget a 5V regulator, and use PWM after solenoid startup instead of change-over to 5V with PWM?

Yes.
It saves a part and is much more efficient.

 

[earckens]

Yes.
It saves a part and is much more efficient.

Hi crutschow, can you please make a design proposal where 12V is applied in full during say 50 or 500ms (somewhat as with IC3 in my initial design) and after that applying pwm using a second 555?
Thanks!

 

[crutschow]

.....can you please make a design proposal where 12V is applied in full during say 50 or 500ms (somewhat as with IC3 in my initial design) and after that applying pwm......

Okay, here's my KISS design.
When power is initially applied, capacitor C3 turns on MOSFET M2, which pulls timing capacitor C2 to ground.
This keeps the 555 output high and applies full voltage to the solenoid.
The value of R2C3 give a delay of about 400ms before M2 turns off and the PWM starts.
This can be varied by changing the value of R2 or C3.

The PWM frequency is about 1.3kHz with the indicated values of U2 and C2.
This can be varied by changing the value of C2.

Note that the solenoid current goes to its maximum at start and then settles to the hold current at about 500ms.
The hold current can be varyied by adjusting pot U2.

 

[earckens]

Okay, here's my KISS design.
When power is initially applied, capacitor C3 turns on MOSFET M2, which pulls timing capacitor C2 to ground.
This keeps the 555 output high and applies full voltage to the solenoid.
The value of R2C3 give a delay of about 400ms before the PWM starts.
This can be varied by changing the value of R2 or C3.

The PWM frequency is about 1.3kHz with the indicated values of U2 and C2.
This can be varied by changing the value of C2.

Note that the solenoid current goes to its maximum at start and then settles to the hold current at about 500ms.
The hold current can be varyied by adjusting pot U2.

View attachment 111928

Great!

Now when an optocoupler is used to trigger this circuit (ref my initial circuit), there is a signal available as long as the optocoupler is activated: can that be used on pin 4 of the 555? Ie: apply permanent supply voltage to pin 8, and keep pin 4 (connected to the optocoupler output) low until the optocoupler is triggered?

 

[crutschow]

........Now when an optocoupler is used to trigger this circuit (ref my initial circuit), there is a signal available as long as the optocoupler is activated: can that be used on pin 4 of the 555? Ie: apply permanent supply voltage to pin 8, and keep pin 4 (connected to the optocoupler output) low until the optocoupler is triggered?

Yes.
Here's the circuit modified to operate from an optocoupler input.
I added MOSFET M3 to hold the 555 Reset input and the delay capacitor low when the opto input is zero.
As you can see from the simulation, the solenoid current is off while the opto input (V(in) from V2 is low and goes through the energizer sequence when the opto input signal goes high.

(Note that the 555 DIP package pinout is shown looking at the bottom of the package.)

 

[earckens]

Fantastic, thanks a lot!!

 

[earckens]

Yes.
Here's the circuit modified to operate from an optocoupler input.
I added MOSFET M3 to hold the 555 Reset input and the delay capacitor low when the opto input is zero.
As you can see from the simulation, the solenoid current is off while the opto input (V(in) from V2 is low and goes through the energizer sequence when the opto input signal goes high.

(Note that the 555 DIP package pinout is shown looking at the bottom of the package.)

View attachment 111942

Theoretically this works well. But when loaded with the solenoid, the MOSFET does not manage to shut off and I get huge backpulses from the coil destroying the 555 within seconds.
D3 is a fast switching 1N4004, M1 is a FQP30N60 (60V 32A).

 

[crutschow]

the MOSFET does not manage to shut off

How do you know it doesn't shut off?
Do you have an oscilloscope?

 

[earckens]

Yes, the scope show attempts at pulsing on and off yet the load voltage remains at supply voltage.

 

[crutschow]

Yes, the scope show attempts at pulsing on and off yet the load voltage remains at supply voltage.

Where does it show "pulsing"?
Where do you measure the load voltage?
Can you post a snapshot of the voltage at the MOSFET gate and drain?

 

[earckens]

Where does it show "pulsing"?
Where do you measure the load voltage?
Can you post a snapshot of the voltage at the MOSFET gate and drain?

When the mosfet attempts to shut down (at each oscillator downpulse) a counter-emf spike (up and down) is generated and the coil remains at its supply voltage. The frequency of these pulses is the same as the oscillator frequency. I will post the scope screen in a few days, now I have broken down the breadboard setup: I will have to be real fast because within seconds the 555 goes up in smoke.
The measurements were done at the mosfet drain.
I may have to get a 5k or so resistor between G and S, if that would help?

 

[alec_t]

You might also want to add a gate-stopper resistor, and 100uF or so directly across the 555 supply pins.

 

[earckens]

You might also want to add a gate-stopper resistor, and 100uF or so directly across the 555 supply pins.

I will try that too; supply output impedance is virtually zero, but a cap might help..

 

[crutschow]

When the mosfet attempts to shut down (at each oscillator downpulse) a counter-emf spike (up and down) is generated and the coil remains at its supply voltage.

Of course it remains at the supply voltage when the MOSFET is off.
What voltage do you expect?

I don't understand how the 555 is getting zapped.
You might try adding a 10 ohm resistor in series with the power to the 555 and add a 100uF cap from the 555 power pin to ground.

Below are how the voltages should roughly appear at the MOSFET gate, V(out), and the MOSFET drain, V(d).

 

[earckens]

Of course it remains at the supply voltage when the MOSFET is off.
What voltage do you expect?

Sorry, I measured at pin 3 of 555

I don't understand how the 555 is getting zapped.
You might try adding a 10 ohm resistor in series with the power to the 555 and add a 100uF cap from the 555 power pin to ground.

Below are how the voltages should roughly appear at the MOSFET gate, V(out), and the MOSFET drain, V(d).

View attachment 111955

For sure the 555 got fried