Using a stepper motor driver with the prop to drive a proportional solenoid val

Siri

I am planning to drive a two proportional solenoid valves(data sheet attached) using Micro dual serial motor controller{data sheet attached).

Is it safe to do it.From what I have read so far looks it is an ok idea.If so will I be able controll the solenoid vlave adequately from close to sloww=ly
to fully open and back to close - using the serial object for the prop.

Thanks,

Siri

ElectricAye

I'm just curious: why is it necessary to use such a motor controller? Couldn't you drive these solenoids with a MOSFET (like IRF3708) controlled by PWM driven directly by the Propeller?

search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=IRF3708PBF-ND

Siri

ElectircAye,

I tried using MOSFETs and driven directly using transsistors but the proprtional valve did not respond well specially when the valve has to be opened very little.It works well when I wanted it to be fully open.

So I called the manufacturer of the VSO valve and was advised using a solenoid driver to use change of current rather than the voltage to get better control.

Thanks,

Siri

ElectricAye

Also just curious: how does this valve provide feedback? Or is feedback something you have to work out with external components, pressure sensors reading back to the microprocessor, etc.?

RinksCustoms

I see no current vs mechanical response curve data, but that does not mean the solenoid has a linear response (most likely not), so its not quite as simple as just PWM'ing a mosfet by the desired mechanical open percentage. This is due to the fact that you essentially have a variable inductor with the variable reluctance comming from the needle valve itself changing the magnetic permiabilaty of the core from moving in or out of that core.·A possible solution to your problem can be simple, (see attached photos *NOTE component values not real, pictures only depict a general idea for discusion purposes).

• You can either eliminate the PWM & Mosfet and just use the DAC·to control the·adjustable voltage regulator (which by ohms law would vary the current through the slolenoid) see picture 1
• Alternatively you could PWM a simple RC network from the prop, picture alt 1...
• Possibly buffered by a simple voltage follower using a 741 op-amp, picture alt 2,.. ·connected to the adj terminal on the 317 regulator which would give you much finer control over the voltage curve vs the DAC idea.
• Or a mix of the latter two, separate PWM's driving the voltage regulator and mosfet.. - the downside to these 4 ideas is that the regulator drops about 1.2V through it so achieving full voltage isnt possible with these designs. The next design will get you full supply voltage less about 0.5V(depending on the characteristic Vf of the mosfet & diode used).
• The other possible solution involves varying the voltage (and thus current) to the solenoid valve would be to use the coil inside the solenoid as shown in picture alt 4 as the coil in a buck voltage converter - a type of voltage regulator. You may need a bleeder resistor across the coil to make the circuit more responsive. If you are familiar with how a buck regulator works it uses pulses to C1 which generates a square wave. The charge on C1 is determined by C1's time constant and pulse duration, D1 then rectifies this voltage to DC and the coil stores the temporary charge rectified by D1. D1 also prevents oscilation by blocking back EMF when the coils magnetic field is collapsing thus using the coils resistance to disipate the collapsing magnetic field in the forward direction.

In any case you'll likely·have to deal with the fact that the solenoid is not a linear device and near open and close it will have a more/less exponential/logarithmic response curve so you will have to experiment some with an interpolation of a rough response curve to scale your programs input to the controlling code so that 10% open equals 10% open mechanically. IMHO, the alt·4 picture with the single PWM simply controlling the time constant of the Buck regulator utilizing the coil within the solenoid would be the simplest, most economical solution. Good luck, i hope i pointed you in the right direction.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Quicker answers in the #propeller chat channel on freenode.net. Don't know squat about IRC? Download Pigin! So easy a caveman could do it...
http://folding.stanford.edu/ - Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.

Post Edited (RinksCustoms) : 7/1/2010 5:12:55 PM GMT

ElectricAye

Thanks, Rinks,

I don't know if Siri appreciates your inputs, but I certainly do.

ErNa

Hello Siri, it is ok to do this with a full bridge driver, for this gives a symmetric driving characteristic. For a servo valve, opening and closing is the same, so if you open by applying a voltage and close just by having a current decay due to freewheeling, this will show different response and can not give symmetrical behaviour. Also, driving a valve with a dc proportional voltage may lead to slip/stick behaviour, so this is also an disadvantage. Indeed, PWM-ing is the best solution, as it is like "knocking" to the valve with a small hammer, slip/stick is reduced.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
cmapspublic3.ihmc.us:80/servlet/SBReadResourceServlet?rid=1181572927203_421963583_5511&partName=htmltext
Hello Rest Of The World
Hello Debris
Install a propeller and blow them away

RinksCustoms

@ErNa- did you look @ the pdf he atatched? Its not a servo but a solenoid - in and out, no spining.. so an h-bridge would be overkill to control a solenoid that needs current in only one direction.

@Electric- Thank you, more than half way through this AEES degree.. and the valve as stated in the pdf siri gave gives feedbvack via closed loop

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Quicker answers in the #propeller chat channel on freenode.net. Don't know squat about IRC? Download Pigin! So easy a caveman could do it...
http://folding.stanford.edu/ - Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.

Post Edited (RinksCustoms) : 7/2/2010 1:19:13 PM GMT

ErNa

No, I didn't first look to the pdf, but now . OK: it is a proportional valve, so it acts like a servo and should be driven like one. The difference to a normal valve is, that current creates a proportional force on the rod of the magnet and the rod pushes to a spring, giving a back force proportional to displacement. The medium itself has not much influence and doesn't disturb those balanced forces much. This behavior is originated from the design of the magnetic flux guide. In a switching valve, the flux changes as abruptly as possible, so whenever the magnets field reaches a certain level, he latches and if you now bring back the current, it will not be released. Very similar to a relay, where activating current can be 10 times the hold current.
The feedback in a proportional application is mostly done via a LVDT (linear variable differential transformer) in a closed loop, but the valve has no sensor, so feedback must use a different input, for example, measuring the media flow.
Anyway, the PWM-ing will always create some vibration and so prevent the actuator from sticking.
It is indeed not necessary to use a separate controller, the propeller could generate the PWM modulation by itself, lets say at 20kHz with an resolution of 12 bits. That should be more than can be expected by an external controller. And the Prop opens the way to vary PWM frequency, so adjust the vibration amplitude of the valve. This might be very useful especially for low throughput.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
cmapspublic3.ihmc.us:80/servlet/SBReadResourceServlet?rid=1181572927203_421963583_5511&partName=htmltext
Hello Rest Of The World
Hello Debris
Install a propeller and blow them away

ElectricAye

ErNa said...
...
The feedback in a proportional application is mostly done via a LVDT (linear variable differential transformer) in a closed loop, but the valve has no sensor, so feedback must use a different input, for example, measuring the media flow....

Okay, that's what I could not see. I saw no mention of an in-built sensor for this in his document, so I was wondering how he was measuring the feedback.

thanks,
Mark

Siri

Sorry for the delay in replying - was out town in DC and my lap top died - motherboard failure so I

had to wait until I got back home.

I appreciate all your help - as ElectricAye mentioned some of it is way beyond my comprehend capability -

as I am a physician nad elrctonic tinkerer.

To answer about feed back - TWO VSO valves are being used to mix oxygen and air in variable proportions-

and the feed back is from an O2 analyser via a DAC.- to get appropriate O2/air mixture.

I am now trying out the VSO valves to see if this could be done.I am in the process of looking into a

O2 analyser to suit my need.

Thanks for all your input.



·Siri

RinksCustoms

two precision pressure sensors could be used, one on each side of an orifice within the flow path.. there are formulas for calculating mass airflow with this method.. involves orifice size, temperature of the fluid, density of the fluid, and the two pressures on either side of the orifice. I dont know the exact formulas, but some digging on the internet about thermaldynamics and mass airflow should produce results. I looked into these formulas a few years ago when i was attempting to make a DFI for my truck using a mass airflow sensor from a corvette engine. Unfortunately the MAF sensors are varying from part number to part number and all give slightly different data. Tracking down airflow vs freq datasheets was imposible..

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Quicker answers in the #propeller chat channel on freenode.net. Don't know squat about IRC? Download Pigin! So easy a caveman could do it...
http://folding.stanford.edu/ - Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.

Siri

RinksCustoms,

Thanks for your help.

Siri

Siri

I received the Pololu Dual serial Motor controller- as suggested by some of you I am now able to control 2 VSO(variable solenoid valve) valves very well.
The valve responds well and just gets warm to touch.

Thanks to all of you who responded to my call for help.

Siri