Shop OBEX P1 Docs P2 Docs Learn Events
closed loop control using basic stamp — Parallax Forums

closed loop control using basic stamp

ArchiverArchiver Posts: 46,084
edited 2003-05-29 18:50 in General Discussion
Hi everyone,

I'm new to the group and just learning about the basic stamp. Before
going further with this technology, I'd like your opinion on wheather
or not the basic stamp is capable of doing what I intend.

I need an inexpensive way to automatically control the pressure
within a vacuum chamber. Process gas is introduced into the chamber
and the pressure is controlled by adjusting a "throttle" valve
between the chamber and the vacuum pump (currently done by hand with
reference to the pressure guage). The guage has a 0-5V linear output
which I monitor with a DVM. I would like to use a pc and Visual Basic
to provide the brains but don't know where to go from there. Stepper
motor to actuate the valve?, motor controller required?,other
hardware needed?

Does anyone know of a similar application that has been developed
that is published on the web or somewhere else? Any help would be
greatly appreciated.

Thanks

Comments

  • ArchiverArchiver Posts: 46,084
    edited 2003-05-26 21:24
    Hi cskjdk,

    Your description is a touch vague, but I think you can
    accomplish what you're looking for. The main
    consideration is the feedback-loop speed. But, since
    you mentioned adjusting the valve by hand, I guess
    loop speed is not a big issue. The BASIC Stamp would
    work for what you want by measuring the 0-5 volt
    output and acting accordingly. But, acting accordingly
    begins the tough (and potentially expensive) part.
    First, I think you have to determine the electrical
    requirements for the valve control. Could this be done
    with an R/C servo? If not, I've seen motorized valves
    in Product Design and Developement magazine (search
    the web). But this is industrial type stuff and not
    cheap. Would it be possible to attach a DC
    gear-reduction motor to the valve shaft? These motors
    are on eBay for around $10.00. Once the determination
    is made about the valve control, then comes the signal
    to control the valve. This would likely involve a
    motor controller and appropriate power supply and
    control signal. From there, a series of if-then
    statements, plus adequate dead-band time (to avoid
    "hunting") is what's required.

    You mentioned PC and VB control. You could use a PC,
    loaded with a simple VB application in conjunction
    with a USB I/O controller to do the job. Look into the
    LabJack U12 controller. It is plug-n-play USB
    connected, had 8 x analog inputs, and 2 x analog
    outputs along with 20 x digital I/O channels and
    timers. You could easily establish a closed-loop
    control system with this. I use one for a robot and it
    is slick. At $119, it's a bargain and is compact and
    robust. With this type of digital controller, you
    could monitor the 0-5 volts directly, and use a 0-5
    volt output to drive a Curtis DC motor controller
    (eBay @ $47) to drive the DC motor. Very doable.

    Dr. Diode
    --- cskjdk <jkostura@s...> wrote:
    > Hi everyone,
    >
    > I'm new to the group and just learning about the
    > basic stamp. Before
    > going further with this technology, I'd like your
    > opinion on wheather
    > or not the basic stamp is capable of doing what I
    > intend.
    >
    > I need an inexpensive way to automatically control
    > the pressure
    > within a vacuum chamber. Process gas is introduced
    > into the chamber
    > and the pressure is controlled by adjusting a
    > "throttle" valve
    > between the chamber and the vacuum pump (currently
    > done by hand with
    > reference to the pressure guage). The guage has a
    > 0-5V linear output
    > which I monitor with a DVM. I would like to use a pc
    > and Visual Basic
    > to provide the brains but don't know where to go
    > from there. Stepper
    > motor to actuate the valve?, motor controller
    > required?,other
    > hardware needed?
    >
    > Does anyone know of a similar application that has
    > been developed
    > that is published on the web or somewhere else? Any
    > help would be
    > greatly appreciated.
    >
    > Thanks
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed.
    > Text in the Subject and Body of the message will be
    > ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to
    > http://docs.yahoo.com/info/terms/
    >
    >


    __________________________________
    Do you Yahoo!?
    The New Yahoo! Search - Faster. Easier. Bingo.
    http://search.yahoo.com
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-27 02:37
    I don't really think you need a PC and VB. Based on your description,
    the stamp should have all the "brains" you require.

    I will do some looking thru some info at work... might have something on
    a servo operated valve. I know I have something on a motor actuated ball
    valve. What are the flow requirements, medium, temperature, etc. etc?


    On Mon, 26 May 2003 06:59:47 -0000 "cskjdk" <jkostura@s...>
    writes:

    > I need an inexpensive way to automatically control the pressure
    > within a vacuum chamber. Process gas is introduced into the chamber
    > and the pressure is controlled by adjusting a "throttle" valve
    > between the chamber and the vacuum pump (currently done by hand with
    > reference to the pressure guage). The guage has a 0-5V linear output
    > which I monitor with a DVM. I would like to use a pc and Visual
    > Basic to provide the brains but don't know where to go from there.
    Stepper
    > motor to actuate the valve?, motor controller required?,other
    > hardware needed?

    ________________________________________________________________
    The best thing to hit the internet in years - Juno SpeedBand!
    Surf the web up to FIVE TIMES FASTER!
    Only $14.95/ month - visit www.juno.com to sign up today!
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-27 05:46
    I agree - my original question was more than a touch vague. I'll try
    again: Initially, the vacuum chamber is evacuated to around 10E-3
    Torr (760 Torr = 14.7psi or one atmosphere pressure). Process gasses
    are then let into the chamber at a very low flowrate with the vacuum
    pump operating. If the gas is introduced faster than it can be pumped
    away, the chamber pressure rises, but the system is designed to allow
    the necessary gas inflow and still maintain a low pressure (vacuum)
    in the chamber. The chamber pressure varies by plus or minus 5% of
    setpoint due to the process taking place within the chamber (a
    metal/gas plasma). In order to control the chamber pressure with a
    given (constant) gas inflow rate, a "throttle valve" is placed
    between the chamber and the vacuum pump. A vacuum gauge reads out the
    chamber pressure and provides a 0-5 V output proportional to the
    pressure (linear). I wish to use this signal to automatically tweak
    the throttle valve so as to maintain a set pressure. The valve is
    a "butterfly" type (6" dia.) and the valve shaft can easily be
    rotated using two fingers. During the process, the control range
    required for the shaft is less than 10 degrees of rotation.
    Adjustments are made every 15 - 30 sec. by hand and a typical
    adjustment requires a shaft rotation of 2-3 degrees. This will
    maintain a gauge output voltage typically, of say, 50 mV plus or
    minus 1 mV. This is what I would like to accomplish with inexpensive
    automation such as the stamp or a pc. Dr. Diode suggests a pc/USB I/O
    controller/ DC motor controller/ DC motor with appropriate mechanical
    linkage (gearing, etc.). Thanks for this info – I'll check out the
    hardware you suggested. AMG AMG suggests going the stamp route –
    thanks for your comments also. If my revised description of what I'm
    trying to do is any more understandable than before and precipitates
    further comments from the group I'd love to hear them. Thanks for
    your patience with this control challenged plasma nerd.

    ps - can this be true?

    Someone named Rich Cook said "Programming today is a race between
    software engineers striving to build bigger and better idiot-proof
    programs, and the Universe trying to produce bigger and better
    idiots. So far, the Universe is winning."


    --- In basicstamps@yahoogroups.com, PH <drdiode2002@y...> wrote:
    > Hi cskjdk,
    >
    > Your description is a touch vague, but I think you can
    > accomplish what you're looking for. The main
    > consideration is the feedback-loop speed. But, since
    > you mentioned adjusting the valve by hand, I guess
    > loop speed is not a big issue. The BASIC Stamp would
    > work for what you want by measuring the 0-5 volt
    > output and acting accordingly. But, acting accordingly
    > begins the tough (and potentially expensive) part.
    > First, I think you have to determine the electrical
    > requirements for the valve control. Could this be done
    > with an R/C servo? If not, I've seen motorized valves
    > in Product Design and Developement magazine (search
    > the web). But this is industrial type stuff and not
    > cheap. Would it be possible to attach a DC
    > gear-reduction motor to the valve shaft? These motors
    > are on eBay for around $10.00. Once the determination
    > is made about the valve control, then comes the signal
    > to control the valve. This would likely involve a
    > motor controller and appropriate power supply and
    > control signal. From there, a series of if-then
    > statements, plus adequate dead-band time (to avoid
    > "hunting") is what's required.
    >
    > You mentioned PC and VB control. You could use a PC,
    > loaded with a simple VB application in conjunction
    > with a USB I/O controller to do the job. Look into the
    > LabJack U12 controller. It is plug-n-play USB
    > connected, had 8 x analog inputs, and 2 x analog
    > outputs along with 20 x digital I/O channels and
    > timers. You could easily establish a closed-loop
    > control system with this. I use one for a robot and it
    > is slick. At $119, it's a bargain and is compact and
    > robust. With this type of digital controller, you
    > could monitor the 0-5 volts directly, and use a 0-5
    > volt output to drive a Curtis DC motor controller
    > (eBay @ $47) to drive the DC motor. Very doable.
    >
    > Dr. Diode
    > --- cskjdk <jkostura@s...> wrote:
    > > Hi everyone,
    > >
    > > I'm new to the group and just learning about the
    > > basic stamp. Before
    > > going further with this technology, I'd like your
    > > opinion on wheather
    > > or not the basic stamp is capable of doing what I
    > > intend.
    > >
    > > I need an inexpensive way to automatically control
    > > the pressure
    > > within a vacuum chamber. Process gas is introduced
    > > into the chamber
    > > and the pressure is controlled by adjusting a
    > > "throttle" valve
    > > between the chamber and the vacuum pump (currently
    > > done by hand with
    > > reference to the pressure guage). The guage has a
    > > 0-5V linear output
    > > which I monitor with a DVM. I would like to use a pc
    > > and Visual Basic
    > > to provide the brains but don't know where to go
    > > from there. Stepper
    > > motor to actuate the valve?, motor controller
    > > required?,other
    > > hardware needed?
    > >
    > > Does anyone know of a similar application that has
    > > been developed
    > > that is published on the web or somewhere else? Any
    > > help would be
    > > greatly appreciated.
    > >
    > > Thanks
    > >
    > >
    > > To UNSUBSCRIBE, just send mail to:
    > > basicstamps-unsubscribe@yahoogroups.com
    > > from the same email address that you subscribed.
    > > Text in the Subject and Body of the message will be
    > > ignored.
    > >
    > >
    > > Your use of Yahoo! Groups is subject to
    > > http://docs.yahoo.com/info/terms/
    > >
    > >
    >
    >
    > __________________________________
    > Do you Yahoo!?
    > The New Yahoo! Search - Faster. Easier. Bingo.
    > http://search.yahoo.com
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-27 15:03
    I think there called Proportional Valves, I have used them years ago with 4
    to 20 ma inputs. This should not be a huge problem, also if your pump works
    at a specific rate maybe a needle valve in the gas line set at the correct
    flow rate and a solenoid or ball valve to open the line would work.
    Larry Gaminde


    Original Message
    From: "cskjdk" <jkostura@s...>
    To: <basicstamps@yahoogroups.com>
    Sent: May 26, 2003 9:46 PM
    Subject: [noparse][[/noparse]basicstamps] Re: closed loop control using basic stamp


    I agree - my original question was more than a touch vague. I'll try
    again: Initially, the vacuum chamber is evacuated to around 10E-3
    Torr (760 Torr = 14.7psi or one atmosphere pressure). Process gasses
    are then let into the chamber at a very low flowrate with the vacuum
    pump operating. If the gas is introduced faster than it can be pumped
    away, the chamber pressure rises, but the system is designed to allow
    the necessary gas inflow and still maintain a low pressure (vacuum)
    in the chamber. The chamber pressure varies by plus or minus 5% of
    setpoint due to the process taking place within the chamber (a
    metal/gas plasma). In order to control the chamber pressure with a
    given (constant) gas inflow rate, a "throttle valve" is placed
    between the chamber and the vacuum pump. A vacuum gauge reads out the
    chamber pressure and provides a 0-5 V output proportional to the
    pressure (linear). I wish to use this signal to automatically tweak
    the throttle valve so as to maintain a set pressure. The valve is
    a "butterfly" type (6" dia.) and the valve shaft can easily be
    rotated using two fingers. During the process, the control range
    required for the shaft is less than 10 degrees of rotation.
    Adjustments are made every 15 - 30 sec. by hand and a typical
    adjustment requires a shaft rotation of 2-3 degrees. This will
    maintain a gauge output voltage typically, of say, 50 mV plus or
    minus 1 mV. This is what I would like to accomplish with inexpensive
    automation such as the stamp or a pc. Dr. Diode suggests a pc/USB I/O
    controller/ DC motor controller/ DC motor with appropriate mechanical
    linkage (gearing, etc.). Thanks for this info - I'll check out the
    hardware you suggested. AMG AMG suggests going the stamp route -
    thanks for your comments also. If my revised description of what I'm
    trying to do is any more understandable than before and precipitates
    further comments from the group I'd love to hear them. Thanks for
    your patience with this control challenged plasma nerd.

    ps - can this be true?

    Someone named Rich Cook said "Programming today is a race between
    software engineers striving to build bigger and better idiot-proof
    programs, and the Universe trying to produce bigger and better
    idiots. So far, the Universe is winning."


    --- In basicstamps@yahoogroups.com, PH <drdiode2002@y...> wrote:
    > Hi cskjdk,
    >
    > Your description is a touch vague, but I think you can
    > accomplish what you're looking for. The main
    > consideration is the feedback-loop speed. But, since
    > you mentioned adjusting the valve by hand, I guess
    > loop speed is not a big issue. The BASIC Stamp would
    > work for what you want by measuring the 0-5 volt
    > output and acting accordingly. But, acting accordingly
    > begins the tough (and potentially expensive) part.
    > First, I think you have to determine the electrical
    > requirements for the valve control. Could this be done
    > with an R/C servo? If not, I've seen motorized valves
    > in Product Design and Developement magazine (search
    > the web). But this is industrial type stuff and not
    > cheap. Would it be possible to attach a DC
    > gear-reduction motor to the valve shaft? These motors
    > are on eBay for around $10.00. Once the determination
    > is made about the valve control, then comes the signal
    > to control the valve. This would likely involve a
    > motor controller and appropriate power supply and
    > control signal. From there, a series of if-then
    > statements, plus adequate dead-band time (to avoid
    > "hunting") is what's required.
    >
    > You mentioned PC and VB control. You could use a PC,
    > loaded with a simple VB application in conjunction
    > with a USB I/O controller to do the job. Look into the
    > LabJack U12 controller. It is plug-n-play USB
    > connected, had 8 x analog inputs, and 2 x analog
    > outputs along with 20 x digital I/O channels and
    > timers. You could easily establish a closed-loop
    > control system with this. I use one for a robot and it
    > is slick. At $119, it's a bargain and is compact and
    > robust. With this type of digital controller, you
    > could monitor the 0-5 volts directly, and use a 0-5
    > volt output to drive a Curtis DC motor controller
    > (eBay @ $47) to drive the DC motor. Very doable.
    >
    > Dr. Diode
    > --- cskjdk <jkostura@s...> wrote:
    > > Hi everyone,
    > >
    > > I'm new to the group and just learning about the
    > > basic stamp. Before
    > > going further with this technology, I'd like your
    > > opinion on wheather
    > > or not the basic stamp is capable of doing what I
    > > intend.
    > >
    > > I need an inexpensive way to automatically control
    > > the pressure
    > > within a vacuum chamber. Process gas is introduced
    > > into the chamber
    > > and the pressure is controlled by adjusting a
    > > "throttle" valve
    > > between the chamber and the vacuum pump (currently
    > > done by hand with
    > > reference to the pressure guage). The guage has a
    > > 0-5V linear output
    > > which I monitor with a DVM. I would like to use a pc
    > > and Visual Basic
    > > to provide the brains but don't know where to go
    > > from there. Stepper
    > > motor to actuate the valve?, motor controller
    > > required?,other
    > > hardware needed?
    > >
    > > Does anyone know of a similar application that has
    > > been developed
    > > that is published on the web or somewhere else? Any
    > > help would be
    > > greatly appreciated.
    > >
    > > Thanks
    > >
    > >
    > > To UNSUBSCRIBE, just send mail to:
    > > basicstamps-unsubscribe@yahoogroups.com
    > > from the same email address that you subscribed.
    > > Text in the Subject and Body of the message will be
    > > ignored.
    > >
    > >
    > > Your use of Yahoo! Groups is subject to
    > > http://docs.yahoo.com/info/terms/
    > >
    > >
    >
    >
    > __________________________________
    > Do you Yahoo!?
    > The New Yahoo! Search - Faster. Easier. Bingo.
    > http://search.yahoo.com


    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject and
    Body of the message will be ignored.


    Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-27 21:26
    Try <A HREF="www.valvestore.com">www.valvestore.com</A>
    They have all sorts of neat (and expensive) stuff. You need a good valve if
    you look at what it will cost you if the valve gets stuck open/closed. What
    kind of damage can be done?
    The stamp can monitor a sensor and operate a valve without any help. If you
    want to use VB and a PC then forget the stamp and buy an I/O board for the PC
    and do it directly.

    Alan Bradford
    Plasma Technologies


    In a message dated 5/26/2003 9:35:16 PM Eastern Daylight Time,
    controlsdude2000@j... writes:

    >
    >
    > I don't really think you need a PC and VB. Based on your description,
    > the stamp should have all the "brains" you require.
    >
    > I will do some looking thru some info at work... might have something on
    > a servo operated valve. I know I have something on a motor actuated ball
    > valve. What are the flow requirements, medium, temperature, etc. etc?
    >
    >
    > On Mon, 26 May 2003 06:59:47 -0000 "cskjdk" <jkostura@s...>
    > writes:
    >
    > >I need an inexpensive way to automatically control the pressure
    > >within a vacuum chamber. Process gas is introduced into the chamber
    > >and the pressure is controlled by adjusting a "throttle" valve
    > >between the chamber and the vacuum pump (currently done by hand with
    > >reference to the pressure guage). The guage has a 0-5V linear output
    > >which I monitor with a DVM. I would like to use a pc and Visual
    > >Basic to provide the brains but don't know where to go from there.
    > Stepper
    > >motor to actuate the valve?, motor controller required?,other
    > >hardware needed?
    >
    > ________________________________________________________________
    > The best thing to hit the internet in years - Juno SpeedBand!
    > Surf the web up to FIVE TIMES FASTER!
    > Only $14.95/ month - visit www.juno.com to sign up today!
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject and
    > Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
    >
    >
    >



    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-27 23:04
    Since you are using such a large valve, I assume you are moving quite
    a bit of gas !

    My approach would be to add a bypass loop and a needle style valve.
    these are considderably easier to control, and adjustments induce
    considderably less system upset.

    Since you are monitoring pressure, there is a very simple psuedo PID
    contorl loop that the stamp can run will easily along with your
    monitoring.


    If you are forced to use the butterfly valve, considder your trig.
    the farther your control point is from the valve axis, the more
    movement you must impart to move it a distance.

    a 1 inch lever will move a small fractin of an inch, whereas a 8"
    lever will require over an inch to move the same degree of rotation.

    The psuedo PID I wrote was for a PLC and is less than a few dozen
    line in that format. if you have any programming experiance it would
    be a piece of cake.

    For simplicity, I would use a DC motor and an H-Bridge as the ammount
    of control imparted on the motor is not that great.

    Dave





    --- In basicstamps@yahoogroups.com, "cskjdk" <jkostura@s...> wrote:
    > I agree - my original question was more than a touch vague. I'll
    try
    > again: Initially, the vacuum chamber is evacuated to around 10E-3
    > Torr (760 Torr = 14.7psi or one atmosphere pressure). Process
    gasses
    > are then let into the chamber at a very low flowrate with the
    vacuum
    > pump operating. If the gas is introduced faster than it can be
    pumped
    > away, the chamber pressure rises, but the system is designed to
    allow
    > the necessary gas inflow and still maintain a low pressure (vacuum)
    > in the chamber. The chamber pressure varies by plus or minus 5% of
    > setpoint due to the process taking place within the chamber (a
    > metal/gas plasma). In order to control the chamber pressure with a
    > given (constant) gas inflow rate, a "throttle valve" is placed
    > between the chamber and the vacuum pump. A vacuum gauge reads out
    the
    > chamber pressure and provides a 0-5 V output proportional to the
    > pressure (linear). I wish to use this signal to automatically tweak
    > the throttle valve so as to maintain a set pressure. The valve is
    > a "butterfly" type (6" dia.) and the valve shaft can easily be
    > rotated using two fingers. During the process, the control range
    > required for the shaft is less than 10 degrees of rotation.
    > Adjustments are made every 15 - 30 sec. by hand and a typical
    > adjustment requires a shaft rotation of 2-3 degrees. This will
    > maintain a gauge output voltage typically, of say, 50 mV plus or
    > minus 1 mV. This is what I would like to accomplish with
    inexpensive
    > automation such as the stamp or a pc. Dr. Diode suggests a pc/USB
    I/O
    > controller/ DC motor controller/ DC motor with appropriate
    mechanical
    > linkage (gearing, etc.). Thanks for this info – I'll check out the
    > hardware you suggested. AMG AMG suggests going the stamp route –
    > thanks for your comments also. If my revised description of what
    I'm
    > trying to do is any more understandable than before and
    precipitates
    > further comments from the group I'd love to hear them. Thanks for
    > your patience with this control challenged plasma nerd.
    >
    > ps - can this be true?
    >
    > Someone named Rich Cook said "Programming today is a race between
    > software engineers striving to build bigger and better idiot-proof
    > programs, and the Universe trying to produce bigger and better
    > idiots. So far, the Universe is winning."
    >
    >
    > --- In basicstamps@yahoogroups.com, PH <drdiode2002@y...> wrote:
    > > Hi cskjdk,
    > >
    > > Your description is a touch vague, but I think you can
    > > accomplish what you're looking for. The main
    > > consideration is the feedback-loop speed. But, since
    > > you mentioned adjusting the valve by hand, I guess
    > > loop speed is not a big issue. The BASIC Stamp would
    > > work for what you want by measuring the 0-5 volt
    > > output and acting accordingly. But, acting accordingly
    > > begins the tough (and potentially expensive) part.
    > > First, I think you have to determine the electrical
    > > requirements for the valve control. Could this be done
    > > with an R/C servo? If not, I've seen motorized valves
    > > in Product Design and Developement magazine (search
    > > the web). But this is industrial type stuff and not
    > > cheap. Would it be possible to attach a DC
    > > gear-reduction motor to the valve shaft? These motors
    > > are on eBay for around $10.00. Once the determination
    > > is made about the valve control, then comes the signal
    > > to control the valve. This would likely involve a
    > > motor controller and appropriate power supply and
    > > control signal. From there, a series of if-then
    > > statements, plus adequate dead-band time (to avoid
    > > "hunting") is what's required.
    > >
    > > You mentioned PC and VB control. You could use a PC,
    > > loaded with a simple VB application in conjunction
    > > with a USB I/O controller to do the job. Look into the
    > > LabJack U12 controller. It is plug-n-play USB
    > > connected, had 8 x analog inputs, and 2 x analog
    > > outputs along with 20 x digital I/O channels and
    > > timers. You could easily establish a closed-loop
    > > control system with this. I use one for a robot and it
    > > is slick. At $119, it's a bargain and is compact and
    > > robust. With this type of digital controller, you
    > > could monitor the 0-5 volts directly, and use a 0-5
    > > volt output to drive a Curtis DC motor controller
    > > (eBay @ $47) to drive the DC motor. Very doable.
    > >
    > > Dr. Diode
    > > --- cskjdk <jkostura@s...> wrote:
    > > > Hi everyone,
    > > >
    > > > I'm new to the group and just learning about the
    > > > basic stamp. Before
    > > > going further with this technology, I'd like your
    > > > opinion on wheather
    > > > or not the basic stamp is capable of doing what I
    > > > intend.
    > > >
    > > > I need an inexpensive way to automatically control
    > > > the pressure
    > > > within a vacuum chamber. Process gas is introduced
    > > > into the chamber
    > > > and the pressure is controlled by adjusting a
    > > > "throttle" valve
    > > > between the chamber and the vacuum pump (currently
    > > > done by hand with
    > > > reference to the pressure guage). The guage has a
    > > > 0-5V linear output
    > > > which I monitor with a DVM. I would like to use a pc
    > > > and Visual Basic
    > > > to provide the brains but don't know where to go
    > > > from there. Stepper
    > > > motor to actuate the valve?, motor controller
    > > > required?,other
    > > > hardware needed?
    > > >
    > > > Does anyone know of a similar application that has
    > > > been developed
    > > > that is published on the web or somewhere else? Any
    > > > help would be
    > > > greatly appreciated.
    > > >
    > > > Thanks
    > > >
    > > >
    > > > To UNSUBSCRIBE, just send mail to:
    > > > basicstamps-unsubscribe@yahoogroups.com
    > > > from the same email address that you subscribed.
    > > > Text in the Subject and Body of the message will be
    > > > ignored.
    > > >
    > > >
    > > > Your use of Yahoo! Groups is subject to
    > > > http://docs.yahoo.com/info/terms/
    > > >
    > > >
    > >
    > >
    > > __________________________________
    > > Do you Yahoo!?
    > > The New Yahoo! Search - Faster. Easier. Bingo.
    > > http://search.yahoo.com
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-27 23:27
    --- In basicstamps@yahoogroups.com, "Larry Gaminde" <lgaminde@t...>
    wrote:
    > I think there called Proportional Valves, I have used them years
    ago with 4
    > to 20 ma inputs. This should not be a huge problem, also if your
    pump works
    > at a specific rate maybe a needle valve in the gas line set at the
    correct
    > flow rate and a solenoid or ball valve to open the line would work.
    > Larry Gaminde

    ball valves are not good for proportinal control.

    butterfly are nearly as bad, but the needle valve would be a huge
    improvement.

    considder also that in a bypass arrangement, the main range could be
    controlled by the butterfly and the fine control by the much smaller
    needle valve.

    for instance, a 6 inch valve open 1/4 of the way may pass 100 CFM and
    a needle valve selected for 20 CFM would then offer much much higer
    resolution over it's range.

    Dave
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-28 02:04
    --- In basicstamps@yahoogroups.com, "Larry Gaminde" <lgaminde@t...>
    wrote:
    > I think there called Proportional Valves, I have used them years
    ago with 4
    > to 20 ma inputs. This should not be a huge problem, also if your
    pump works
    > at a specific rate maybe a needle valve in the gas line set at the
    correct
    > flow rate and a solenoid or ball valve to open the line would work.
    > Larry Gaminde

    ball valves are not good for proportinal control.

    butterfly are nearly as bad, but the needle valve would be a huge
    improvement.

    considder also that in a bypass arrangement, the main range could be
    controlled by the butterfly and the fine control by the much smaller
    needle valve.

    for instance, a 6 inch valve open 1/4 of the way may pass 100 CFM and
    a needle valve selected for 20 CFM would then offer much much higer
    resolution over it's range.

    Dave
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-28 07:05
    Thanks all for the overwhelming response to my question. Some
    comments I received questioned the suitability of a 6" butterfly
    valve for this application. Actually, in a high vacuum situation a
    butterfly valve works fine. Remember that at 10E-3 Torr pressure
    (about 1 millionth of an atmosphere) there aren't very many
    (relatively speaking) gas molecules going past the valve so small
    pressure changes can be handled quite well. I've decided to go with
    the pc/I/O board concept suggested by Dr Diode. I checked out the
    LabJack U12 controller and it looks like a promising candidate.

    Thanks to all again for your help.

    John (cskjdk)
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-29 07:16
    cskjdk wrote:
    > I'm new to the group and just learning about the basic stamp. Before
    > going further with this technology, I'd like your opinion on wheather
    > or not the basic stamp is capable of doing what I intend.
    >
    > I need an inexpensive way to automatically control the pressure
    > within a vacuum chamber. Process gas is introduced into the chamber
    > and the pressure is controlled by adjusting a "throttle" valve
    > between the chamber and the vacuum pump (currently done by hand with
    > reference to the pressure guage). The guage has a 0-5V linear output
    > which I monitor with a DVM. I would like to use a pc and Visual Basic
    > to provide the brains but don't know where to go from there. Stepper
    > motor to actuate the valve?, motor controller required?,other
    > hardware needed?
    >
    > Does anyone know of a similar application that has been developed
    > that is published on the web or somewhere else? Any help would be
    > greatly appreciated.

    IMO, you are attempting to create a "servovalve" the hard way.
    Search for "servovalve" (and Pneumatics or vacuum) for some
    technology hints.

    OOC... Is this for a job, or a personal project? If for a job,
    you'll probably find that once you've factored in your cost of time,
    parts, assembly, etc, to reinvent the servovalve from scratch,
    it would probably have been much cheaper to simply buy an off
    the shelf one. (Hobby projects are different animals, though. :-)

    I'll assume for now that you're still going to try it this way.

    Your biggest problem if you try it this way is definitely
    a HORRIBLY slow loop speed. Unless you are talking a LONG
    time, or this system has a LOT of inertia, you could easily
    end up with an oscillating pressure.

    DVMs: First off, DVMs are IMO not the best way to measure this.
    They take a long time to settle, sample, then EVENTUALLY
    send it out serial. The date rate is LOW, and by the time
    you've read it in, the latency for the data is significant.

    Heck, you already have a 0-5V sensor. You are MUCH better
    off directly interfacing your transducer to the micro via
    a 0-5V A/D converter. You can always use your DVM as a
    "local display" of the value for diagnostic purposes. <grin>
    See the Stamp appnotes on parts and how to do that.

    Next...

    Under certain circumstances, instead of trying to make
    a servo loop open and close a proportional valve, you MAY
    be able to save money by using PWM and an on/off (bang-bang)
    valve. This works very well for certain values of flow
    rates, system inertial capacities, etc.. YMMV though
    depending on the application.

    To use a proportional valve, you'll need to mechanically
    interface a servo to it. Otherwise, you'll ALSO have the
    additional challenge of adding feedback to THAT, and
    setting up an "inner" servo loop JUST for valve positioning
    control, further slowing the system's response time.
    BTW, If you DO decide to add a servo to a standard valve,
    your programming may get a boost if you take the time to
    measure and create an "opening command vs air flow rate"
    lookup table in a test rig.

    Lastly, to program this you should learn about PID Control
    Loop technology, to understand how control loops work.

    Bottom line:

    IMHO, a Stamp / DVM / and motor driven valve will only work
    if you're talking on the order of seconds to minutes loop
    response times. If you need something faster (and aren't
    buying a servovalve off the shelf), my advice is to use a
    micro with a built-in A/D, and a good PID algorithm already
    written and available in their applications library.

    Good luck.

    - Keith Mc.
    --
    "Bees aren't that busy, they just can't buzz any slower..."
  • ArchiverArchiver Posts: 46,084
    edited 2003-05-29 18:50
    Dude, take a look at pneucleus.com . . . Tell em the cadguy sent ya!
    Good luck with it.

    --- In basicstamps@yahoogroups.com, "Keith Mc." <acti@P...> wrote:
    > cskjdk wrote:
    > > I'm new to the group and just learning about the basic stamp.
    Before
    > > going further with this technology, I'd like your opinion on
    wheather
    > > or not the basic stamp is capable of doing what I intend.
    > >
    > > I need an inexpensive way to automatically control the pressure
    > > within a vacuum chamber. Process gas is introduced into the
    chamber
    > > and the pressure is controlled by adjusting a "throttle" valve
    > > between the chamber and the vacuum pump (currently done by hand
    with
    > > reference to the pressure guage). The guage has a 0-5V linear
    output
    > > which I monitor with a DVM. I would like to use a pc and Visual
    Basic
    > > to provide the brains but don't know where to go from there.
    Stepper
    > > motor to actuate the valve?, motor controller required?,other
    > > hardware needed?
    > >
    > > Does anyone know of a similar application that has been developed
    > > that is published on the web or somewhere else? Any help would be
    > > greatly appreciated.
    >
    > IMO, you are attempting to create a "servovalve" the hard way.
    > Search for "servovalve" (and Pneumatics or vacuum) for some
    > technology hints.
    >
    > OOC... Is this for a job, or a personal project? If for a job,
    > you'll probably find that once you've factored in your cost of time,
    > parts, assembly, etc, to reinvent the servovalve from scratch,
    > it would probably have been much cheaper to simply buy an off
    > the shelf one. (Hobby projects are different animals, though. :-)
    >
    > I'll assume for now that you're still going to try it this way.
    >
    > Your biggest problem if you try it this way is definitely
    > a HORRIBLY slow loop speed. Unless you are talking a LONG
    > time, or this system has a LOT of inertia, you could easily
    > end up with an oscillating pressure.
    >
    > DVMs: First off, DVMs are IMO not the best way to measure this.
    > They take a long time to settle, sample, then EVENTUALLY
    > send it out serial. The date rate is LOW, and by the time
    > you've read it in, the latency for the data is significant.
    >
    > Heck, you already have a 0-5V sensor. You are MUCH better
    > off directly interfacing your transducer to the micro via
    > a 0-5V A/D converter. You can always use your DVM as a
    > "local display" of the value for diagnostic purposes. <grin>
    > See the Stamp appnotes on parts and how to do that.
    >
    > Next...
    >
    > Under certain circumstances, instead of trying to make
    > a servo loop open and close a proportional valve, you MAY
    > be able to save money by using PWM and an on/off (bang-bang)
    > valve. This works very well for certain values of flow
    > rates, system inertial capacities, etc.. YMMV though
    > depending on the application.
    >
    > To use a proportional valve, you'll need to mechanically
    > interface a servo to it. Otherwise, you'll ALSO have the
    > additional challenge of adding feedback to THAT, and
    > setting up an "inner" servo loop JUST for valve positioning
    > control, further slowing the system's response time.
    > BTW, If you DO decide to add a servo to a standard valve,
    > your programming may get a boost if you take the time to
    > measure and create an "opening command vs air flow rate"
    > lookup table in a test rig.
    >
    > Lastly, to program this you should learn about PID Control
    > Loop technology, to understand how control loops work.
    >
    > Bottom line:
    >
    > IMHO, a Stamp / DVM / and motor driven valve will only work
    > if you're talking on the order of seconds to minutes loop
    > response times. If you need something faster (and aren't
    > buying a servovalve off the shelf), my advice is to use a
    > micro with a built-in A/D, and a good PID algorithm already
    > written and available in their applications library.
    >
    > Good luck.
    >
    > - Keith Mc.
    > --
    > "Bees aren't that busy, they just can't buzz any slower..."
Sign In or Register to comment.