hey all,

I am trying to figure out what kind of circuit I need to control a proportional solenoid valve. here is a link to the valve in question....

Soolenoid Valve

I was thinking I could use somthing like this...

MOSFET

driven from a PIC18xxx PWM to drive the valve. I would need 0% duty cycle (0v) coming from the PIC to equal 0v to the valve and 100% duty cycle (5v) coming from the PIC to equal 12v to the valve. basically, output would be proportional to duty cycle.

This will be used on a hydraulic pump trailer (to control flow) with a 12v system very similar to a car.

Any thoughts on design or help on choosing components in this situation would be much appreciated. I have never designed anything like this before.

Thanks.

 

You could use a 555 timer configured as a PMW modulator similar to this. The coil would be in place of the motor.

You need to add a transient suppresion diode directly across the solenoid coil (2A rectifier, cathode to positive side) and probably should reduce the value of R2 from 10k ohm to about 1k ohm for faster switching of the MOSFET.

 

You could use a 555 timer configured as a PMW modulator similar to this. The coil would be in place of the motor.

You need to add a transient suppresion diode directly across the solenoid coil (2A rectifier, cathode to positive side) and probably should reduce the value of R2 from 10k ohm to about 1k ohm for faster switching of the MOSFET.

Crutschow,

Thanks for the suggestions. I think using the PIC to control the pwm is the way I need to go. The pwm will change quite often because this is a user controlled valve and i'm not sure the 555 timer method is best suited for this.

 

Maybe something like this would work...

Am I missing something here or would this do the trick and be reliable?

 

Maybe something like this would work...

Am I missing something here or would this do the trick and be reliable?

View attachment 39018

The circuit should work but that MOSFET requires switching 28nC of gate charge to turn on and off. The switching time would thus be 28nC divided by the PIC output current rating. If that time is too slow (longer than a few microseconds) then you may need to use a driver, such as the MOSFET you referenced in your first post, between the PIC output and the MOSFET input.

 

I'm not sure I fully follow you here. could you explain this in a bit more detail?

If i needed to use one chip to drive the other, how would that look and how would it function.

Sorry for my lack of knowledge here, I'm still learning.

 

Maybe something like this??



But wouldn't that supply more voltage to the gate of the FET than it's rated for?

** Edit --

I see that it would not supply over voltage now. But I still lack the understanding of WHY use both components. If I am reading the datasheets correctly, it appears as though the MCP1407 Driver could do the job of both. Any input here would be helpful, I would just like to understand it

Thanks.

 

Maybe a bit more like this one to show how it would be wired. Does this drawing make more sense?

 

Mechanically it's very difficult to get a solenoid to maintain a stable fixed position between its end limits!

It will be very unstable (load dependent) and non-linear.

About the only way to do it well it to have a closed loop system where you have high speed feedback of the solenoid armature position.

 

Mechanically it's very difficult to get a solenoid to maintain a stable fixed position between its end limits!

It will be very unstable (load dependent) and non-linear.

About the only way to do it well it to have a closed loop system where you have high speed feedback of the solenoid armature position.

If you look at his referenced solenoid, it is stated that the flow is proportional to the current through it, so it may have internal electronics to perform that function.

 

I don't know why you're using an IRL530 when your Vcc will likely be limited to under 15v.

Look for a MOSFET with a much lower Vdss rating. You will have a much lower gate charge.

Also, in your last couple of attachments, you show a capacitor in parallel with the MOSFET gate. That is not good.

 

Mechanically it's very difficult to get a solenoid to maintain a stable fixed position between its end limits!

It will be very unstable (load dependent) and non-linear.

About the only way to do it well it to have a closed loop system where you have high speed feedback of the solenoid armature position.

Thanks for the feed back guys! Any chance you can give me a example of what you are saying here with the closed loop system? I'm not clear on what this is.

I will have a sensor measuring flow through the valve and the PIC can make adjustments to PWM duty cycle accordingly.

On another note, thanks for the tips. I will remove the cap on the gate line and see about getting a MOSFET that has lower Vdss

 

...................
If I am reading the datasheets correctly, it appears as though the MCP1407 Driver could do the job of both. Any input here would be helpful, I would just like to understand it

That driver has a maximum on-resistance of 2.5Ω which will give a voltage drop of 4V at the 1.6A maximum solenoid current and this will dissipate 6.4W. This is way more than the DIP package can dissipate.

Thus you need a lower on-resistance MOSFET to drive the solenoid.

 

That driver has a maximum on-resistance of 2.5Ω which will give a voltage drop of 4V at the 1.6A maximum solenoid current and this will dissipate 6.4W. This is way more than the DIP package can dissipate.

Thus you need a lower on-resistance MOSFET to drive the solenoid.

Your saying this is why I need both the driver and the power FET? or are you saying this driver will not work at all?

BTW. I did go ahead and order an assortment of different MOSFETs and Drivers to play around with (part of the learning curve) and they should be here next week. At least then I can start breadboarding a few things and ask more pointed questions.

Again, sorry for my lack of knowledge here, I am just trying to learn as I go so thanks for all your input.

 

Yes, for fastest switching you would use the driver to control the MOSFET.

 

If you look at his referenced solenoid, it is stated that the flow is proportional to the current through it, so it may have internal electronics to perform that function.

Thanks Crutschow, I stand corrected. I didn't download the PDF and check that specific solenoid, and so assumed it was a standard on/off type.

If it is controlled by current then an open loop PWM voltage from a FET is not the best way to go about it. It needs current feedback.

What about using a motor driver IC that has a current feedback resistor? Most are good for 2A or so and have an analogue voltage input (that gives a constant current output into a coil or motor).

Some of the ICs for stepper motor control also accept an analogue voltage input and give a constant current output.

 

.............

If it is controlled by current then an open loop PWM voltage from a FET is not the best way to go about it. It needs current feedback.

What about using a motor driver IC that has a current feedback resistor? Most are good for 2A or so and have an analogue voltage input (that gives a constant current output into a coil or motor).

Some of the ICs for stepper motor control also accept an analogue voltage input and give a constant current output.

If the solenoid inductance is sufficient to maintain a low ripple current through its coil with high frequency PWM, then the solenoid circuitry make work properly.

Do those analog voltage drivers have a variable current output proprotional to the voltage input which is what the solenoid needs?

 

Oh boy, now I'm really lost

 

If the solenoid inductance is sufficient to maintain a low ripple current through its coil with high frequency PWM, then the solenoid circuitry make work properly.

Do those analog voltage drivers have a variable current output proprotional to the voltage input which is what the solenoid needs?

Yep, many of the stepper driver ICs do exactly that.

Some ICs have the "8th stepping" ie the DAC built in, but some ICs have a voltage input, that is used to control the PWM inside the IC to provide a constant current output (which is proportional to the voltage input).

Here's the TI page;
http://www.ti.com/ww/en/motor_drive_and_control_solutions/motor_control_integrated_motor_drivers.htm

Of course for this simple app it can be done like the old discrete stepper driver designs, a comparator IC (as an oscillator) is used to control a switching FET, with a current feedback resistor.

 

This thread has been a great help so far guys, Thanks!

Now that I am getting closer on this circuit, what would be the best method to monitor the current being sent to the solenoid in question?

Im thinking just a simple .1ohm resistor style circuit to measure current but I am concerned about too much heat at full current (1.7Amps)

maybe one of these with a heat sink and one of these to clamp the voltage?

What are your thoughts?