Shop OBEX P1 Docs P2 Docs Learn Events
Linear displacement — Parallax Forums

Linear displacement

ArchiverArchiver Posts: 46,084
edited 2003-03-26 20:58 in General Discussion
I am looking for suggestions for a hardware interface for my stamp. I want to
measure very small displacements (0 to 5mm) with 0.01mm accuracy. I don't
think I want to use an LVDT. Perhaps an optical sensor, Linear Hall effect,
??????????? I have some general ideas, but many heads are better than my one
bald one!

Any suggestions welcomed.

Ken


[noparse][[/noparse]Non-text portions of this message have been removed]
«1

Comments

  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 18:43
    0-5mm in 0.01mm increments is pretty fine and may require some serious
    brainwork to get any repeatability with a hardware solution. You can buy
    digital calipers or digital machine scales and interface them to a Stamp,
    but these are usually only good to 0.001 inches, even though they claim
    0.0005 inches.

    When I worked for a company that manufactured leak detection equipment for
    the service station industry, they used a calibrated piston / cylinder and
    LVDT to detect very minute leaks in plumbing. The engineers claimed nothing
    else worked as well.

    Original Message

    > I am looking for suggestions for a hardware interface for my stamp. I want
    to
    > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I don't
    > think I want to use an LVDT. Perhaps an optical sensor, Linear Hall
    effect,
    > ??????????? I have some general ideas, but many heads are better than my
    one
    > bald one!
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 19:29
    What about decreasing the diameter of the vessel? That would increase the
    resolution of an optical type system.

    What kind of fluid are we talking? Is it always the same density/color?
    What degree of accuracy is required? Is it under any pressure?

    The calibrated piston sounds like the way to go if really fine accuracy is
    needed.

    Just a thought,

    Jonathan

    Original Message
    From: "Rodent" <daweasel@s...>
    To: <basicstamps@yahoogroups.com>
    Sent: Thursday, March 20, 2003 10:43 AM
    Subject: Re: [noparse][[/noparse]basicstamps] Linear displacement


    > 0-5mm in 0.01mm increments is pretty fine and may require some serious
    > brainwork to get any repeatability with a hardware solution. You can buy
    > digital calipers or digital machine scales and interface them to a Stamp,
    > but these are usually only good to 0.001 inches, even though they claim
    > 0.0005 inches.
    >
    > When I worked for a company that manufactured leak detection equipment for
    > the service station industry, they used a calibrated piston / cylinder and
    > LVDT to detect very minute leaks in plumbing. The engineers claimed
    nothing
    > else worked as well.
    >
    >
    Original Message
    >
    > > I am looking for suggestions for a hardware interface for my stamp. I
    want
    > to
    > > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I
    don't
    > > think I want to use an LVDT. Perhaps an optical sensor, Linear Hall
    > effect,
    > > ??????????? I have some general ideas, but many heads are better than my
    > one
    > > bald one!
    >
    >
    >
    >
    > 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 2003-03-20 19:47
    Not sure he was wanting to measure a liquid when the original poster
    mentioned displacement. I only mentioned the fluid measurement because it
    used LVDT as a way to measure minute increments.

    What I was getting at was that with very fine measurements, using mechanical
    means to "amplify" a measurement would be harder than other methods.

    Original Message

    > What about decreasing the diameter of the vessel? That would increase the
    > resolution of an optical type system.
    >
    > What kind of fluid are we talking? Is it always the same density/color?
    > What degree of accuracy is required? Is it under any pressure?
    >
    > The calibrated piston sounds like the way to go if really fine accuracy is
    > needed.

    >
    Original Message

    > > 0-5mm in 0.01mm increments is pretty fine and may require some serious
    > > brainwork to get any repeatability with a hardware solution. You can buy
    > > digital calipers or digital machine scales and interface them to a
    Stamp,
    > > but these are usually only good to 0.001 inches, even though they claim
    > > 0.0005 inches.
    > >
    > > When I worked for a company that manufactured leak detection equipment
    for
    > > the service station industry, they used a calibrated piston / cylinder
    and
    > > LVDT to detect very minute leaks in plumbing. The engineers claimed
    > nothing
    > > else worked as well.
    > >
    > >
    Original Message
    > >
    > > > I am looking for suggestions for a hardware interface for my stamp. I
    > want
    > > to
    > > > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I
    > don't
    > > > think I want to use an LVDT. Perhaps an optical sensor, Linear Hall
    > > effect,
    > > > ??????????? I have some general ideas, but many heads are better than
    my
    > > one
    > > > bald one!
    > >
    > >
    > >
    > >
    > > 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/
    > >
    > >
    > >
    > >
    >
    >
    > 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 2003-03-20 19:55
    How much $$ though? My suggestion is $100-150.

    Original Message

    > Mitutoyo offers a device with resolution of .0001 mm
    > and an accuracy of .002 mm over 5,10,25,50 mm range.
    > 4 um cycle square wave output, RS422 compatible.
    > See "LGB/LGF linear gauges". My info is dated - Current
    > devices may be better.

    > > 0-5mm in 0.01mm increments is pretty fine and may require some serious
    > > brainwork to get any repeatability with a hardware solution. You can buy
    > > digital calipers or digital machine scales and interface them to a
    Stamp,
    > > but these are usually only good to 0.001 inches, even though they claim
    > > 0.0005 inches.
    > >
    > > When I worked for a company that manufactured leak detection equipment
    for
    > > the service station industry, they used a calibrated piston / cylinder
    and
    > > LVDT to detect very minute leaks in plumbing. The engineers claimed
    nothing
    > > else worked as well.
    > >
    > >
    Original Message
    > >
    > > > I am looking for suggestions for a hardware interface for my stamp. I
    want
    > > to
    > > > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I
    don't
    > > > think I want to use an LVDT. Perhaps an optical sensor, Linear Hall
    > > effect,
    > > > ??????????? I have some general ideas, but many heads are better than
    my
    > > one
    > > > bald one!
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 20:13
    In a message dated 3/20/2003 11:49:54 AM Pacific Standard Time,
    daweasel@s... writes:

    > Not sure he was wanting to measure a liquid when the original poster
    > mentioned displacement. I only mentioned the fluid measurement because it
    > used LVDT as a way to measure minute increments.
    >
    > What I was getting at was that with very fine measurements, using
    > mechanical
    > means to "amplify" a measurement would be harder than other methods.

    I will not be measuring liquid. I will be measuring a deflection. I don't
    have any experience with strain guages, but it would seem a strain guage
    would not be real repeatable.????


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 20:18
    In a message dated 3/20/2003 12:00:44 PM Pacific Standard Time,
    daweasel@s... writes:

    > How much $$ though? My suggestion is $100-150.
    >

    As little $$ as possible (of course). I think I may try to hack a caliper.
    Can you tell me a little about the caliper RS232 output. What does it output,
    a lineat voltage, ASCII stuff or????


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 21:24
    From what I have gathered so far, the output of the Mitutoyo and knock-off
    calipers and linear scales requires you to clock each bit of data from the
    unit. A level converter is required as the output voltage is very low due to
    these units running off batteries.

    I have some Stamp code I got from another user thats supposed to work with
    the Mitutoyo calipers, and have messed around with a level converter some,
    but don't have a fully functional unit yet. I have found several sources for
    display heads for these devices -- one is www.wttool.com -- item #
    1289-0465. They also sell the linear scales. I have another source, but I
    need to track down their URL.

    Original Message

    > > How much $$ though? My suggestion is $100-150.
    > >
    >
    > As little $$ as possible (of course). I think I may try to hack a
    caliper.
    > Can you tell me a little about the caliper RS232 output. What does it
    output,
    > a lineat voltage, ASCII stuff or????
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 21:27
    if your target is ferrous you could try an "analog
    proximity sensor".


    --- goflo@p... wrote:
    > Mitutoyo caliper protocol is 3-wire, CLK, DATA, REQ.
    >
    > CLK and DATA are open-collector outputs. 10K
    > pull-ups
    > work fine.
    >
    > Pulling REQ low produces 52 clocks, DATA valid on
    > CLK low, each group of 4 bits represents a hex #,
    > lsb to msb:
    >
    > d0-d3 0Fh
    >
    > d4 sign digit, 0 is positive, 1 negative
    >
    > d5-d10 measurement value, BCD, MSD to LSD,
    >
    > d11 decimal point position in the 6-digit
    > measurement value, counting right to left
    >
    > d12 units, 0=mm, 1=inches
    >
    > IIRC CLK freq is 1562 Hz, or thereabouts.
    >
    > regards, Jack
    >
    > smartdim@a... wrote:
    > >
    > > In a message dated 3/20/2003 12:00:44 PM Pacific
    > Standard Time,
    > > daweasel@s... writes:
    > >
    > > > How much $$ though? My suggestion is $100-150.
    > > >
    > >
    > > As little $$ as possible (of course). I think I
    > may try to hack a caliper.
    > > Can you tell me a little about the caliper RS232
    > output. What does it output,
    > > a lineat voltage, ASCII stuff or????
    >
    >
    > 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!?
    Yahoo! Platinum - Watch CBS' NCAA March Madness, live on your desktop!
    http://platinum.yahoo.com
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 21:28
    Gracias. I had a good code example for the Stamp, but no interface or
    protocol info. I had grafted a standard Molex KK connector on my calipers
    and started working on an interface, but got sidetracked.

    Original Message

    > Mitutoyo caliper protocol is 3-wire, CLK, DATA, REQ.
    >
    > CLK and DATA are open-collector outputs. 10K pull-ups
    > work fine.
    >
    > Pulling REQ low produces 52 clocks, DATA valid on
    > CLK low, each group of 4 bits represents a hex #,
    > lsb to msb:
    >
    > d0-d3 0Fh
    >
    > d4 sign digit, 0 is positive, 1 negative
    >
    > d5-d10 measurement value, BCD, MSD to LSD,
    >
    > d11 decimal point position in the 6-digit
    > measurement value, counting right to left
    >
    > d12 units, 0=mm, 1=inches
    >
    > IIRC CLK freq is 1562 Hz, or thereabouts.
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 21:35
    Mitutoyo offers a device with resolution of .0001 mm
    and an accuracy of .002 mm over 5,10,25,50 mm range.
    4 um cycle square wave output, RS422 compatible.
    See "LGB/LGF linear gauges". My info is dated - Current
    devices may be better.

    Jack

    Rodent wrote:
    >
    > 0-5mm in 0.01mm increments is pretty fine and may require some serious
    > brainwork to get any repeatability with a hardware solution. You can buy
    > digital calipers or digital machine scales and interface them to a Stamp,
    > but these are usually only good to 0.001 inches, even though they claim
    > 0.0005 inches.
    >
    > When I worked for a company that manufactured leak detection equipment for
    > the service station industry, they used a calibrated piston / cylinder and
    > LVDT to detect very minute leaks in plumbing. The engineers claimed nothing
    > else worked as well.
    >
    >
    Original Message
    >
    > > I am looking for suggestions for a hardware interface for my stamp. I want
    > to
    > > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I don't
    > > think I want to use an LVDT. Perhaps an optical sensor, Linear Hall
    > effect,
    > > ??????????? I have some general ideas, but many heads are better than my
    > one
    > > bald one!
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 21:41
    In a message dated 3/20/2003 12:58:27 PM Pacific Standard Time,
    goflo@p... writes:

    >
    > Mitutoyo caliper protocol is 3-wire, CLK, DATA, REQ.
    >
    > CLK and DATA are open-collector outputs. 10K pull-ups
    > work fine.
    >
    > Pulling REQ low produces 52 clocks, DATA valid on
    > CLK low, each group of 4 bits represents a hex #,
    > lsb to msb:
    >
    > d0-d3 0Fh
    >
    > d4 sign digit, 0 is positive, 1 negative
    >
    > d5-d10 measurement value, BCD, MSD to LSD,
    >
    > d11 decimal point position in the 6-digit
    > measurement value, counting right to left
    >
    > d12 units, 0=mm, 1=inches
    >
    > IIRC CLK freq is 1562 Hz, or thereabouts.
    >
    > regards, Jack
    >

    Awesome......Thanks Jack!!!!!!!1


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 21:47
    In a message dated 3/20/2003 1:26:17 PM Pacific Standard Time,
    daweasel@s... writes:

    > From what I have gathered so far, the output of the Mitutoyo and knock-off
    > calipers and linear scales requires you to clock each bit of data from the
    > unit. A level converter is required as the output voltage is very low due
    > to
    > these units running off batteries.
    >
    > I have some Stamp code I got from another user thats supposed to work with
    > the Mitutoyo calipers, and have messed around with a level converter some,
    > but don't have a fully functional unit yet. I have found several sources
    > for
    > display heads for these devices -- one is www.wttool.com -- item #
    > 1289-0465. They also sell the linear scales. I have another source, but I
    > need to track down their URL.
    >

    This is what's so great about this forum!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Lots
    of help from the Stamp Brethren.

    Thanks.....

    Ken


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-20 22:47
    Mitutoyo caliper protocol is 3-wire, CLK, DATA, REQ.

    CLK and DATA are open-collector outputs. 10K pull-ups
    work fine.

    Pulling REQ low produces 52 clocks, DATA valid on
    CLK low, each group of 4 bits represents a hex #,
    lsb to msb:

    d0-d3 0Fh

    d4 sign digit, 0 is positive, 1 negative

    d5-d10 measurement value, BCD, MSD to LSD,

    d11 decimal point position in the 6-digit
    measurement value, counting right to left

    d12 units, 0=mm, 1=inches

    IIRC CLK freq is 1562 Hz, or thereabouts.

    regards, Jack

    smartdim@a... wrote:
    >
    > In a message dated 3/20/2003 12:00:44 PM Pacific Standard Time,
    > daweasel@s... writes:
    >
    > > How much $$ though? My suggestion is $100-150.
    > >
    >
    > As little $$ as possible (of course). I think I may try to hack a caliper.
    > Can you tell me a little about the caliper RS232 output. What does it output,
    > a lineat voltage, ASCII stuff or????
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-21 08:39
    (Warning: semi tongue in cheek reply follows)

    You could attach a mirror to the thing that will be displaced, shine a laser on
    it, and measure the beam deflection on the side of a building several blocks
    away [noparse]:)[/noparse]

    I shouldn't type when it's so long after my bedtime..g'night.



    --- In basicstamps@yahoogroups.com, smartdim@a... wrote:
    > I am looking for suggestions for a hardware interface for my stamp. I want to
    > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I don't
    > think I want to use an LVDT. Perhaps an optical sensor, Linear Hall effect,
    > ??????????? I have some general ideas, but many heads are better than my one
    > bald one!
    >
    > Any suggestions welcomed.
    >
    > Ken
    >
    >
    > [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-21 14:34
    Ken, Keyence makes laser sensors which can accurately sense distances but
    they are a bit expensive. check out their site at www.keyence.com.
    Jim

    Original Message
    From: stan45039 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=0qAqRmBbGI6PGuHkm5scUwMAeR_Jg080NxEsL4AtXZhBLmD-_KL8xwdDnOYcQs7t032sAhBA]slogan@f...[/url
    Sent: Friday, March 21, 2003 3:39 AM
    To: basicstamps@yahoogroups.com
    Subject: [noparse][[/noparse]basicstamps] Re: Linear displacement



    (Warning: semi tongue in cheek reply follows)

    You could attach a mirror to the thing that will be displaced, shine a laser
    on it, and measure the beam deflection on the side of a building several
    blocks away [noparse]:)[/noparse]

    I shouldn't type when it's so long after my bedtime..g'night.



    --- In basicstamps@yahoogroups.com, smartdim@a... wrote:
    > I am looking for suggestions for a hardware interface for my stamp. I want
    to
    > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I don't
    > think I want to use an LVDT. Perhaps an optical sensor, Linear Hall
    effect,
    > ??????????? I have some general ideas, but many heads are better than my
    one
    > bald one!
    >
    > Any suggestions welcomed.
    >
    > Ken
    >
    >
    > [noparse][[/noparse]Non-text portions of this message have been removed]


    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 2003-03-21 18:05
    There was also a company called cyberoptics that did something along this in
    many years ago.

    Original Message
    From: Jim Forkin [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=LJNIW39CqXlbaGG1hwdFuUTYa4MJpNmJNgz1dptCEFZaJh6a-IqGnJoelnlckouyFYn4TSPoTtk]jjf@p...[/url
    Sent: Friday, March 21, 2003 9:35 AM
    To: basicstamps@yahoogroups.com
    Subject: RE: [noparse][[/noparse]basicstamps] Re: Linear displacement

    Ken, Keyence makes laser sensors which can accurately sense distances but
    they are a bit expensive. check out their site at www.keyence.com.
    Jim

    Original Message
    From: stan45039 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=Zkfy_gePNalJ5zBI7azwHcsi2kAGuRYB6dgCk4DHS3_AhroDZQ99SFQ8QCEJWS89FaMCzgn63eY]slogan@f...[/url
    Sent: Friday, March 21, 2003 3:39 AM
    To: basicstamps@yahoogroups.com
    Subject: [noparse][[/noparse]basicstamps] Re: Linear displacement



    (Warning: semi tongue in cheek reply follows)

    You could attach a mirror to the thing that will be displaced, shine a laser
    on it, and measure the beam deflection on the side of a building several
    blocks away [noparse]:)[/noparse]

    I shouldn't type when it's so long after my bedtime..g'night.



    --- In basicstamps@yahoogroups.com, smartdim@a... wrote:
    > I am looking for suggestions for a hardware interface for my stamp. I want
    to
    > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I don't
    > think I want to use an LVDT. Perhaps an optical sensor, Linear Hall
    effect,
    > ??????????? I have some general ideas, but many heads are better than my
    one
    > bald one!
    >
    > Any suggestions welcomed.
    >
    > Ken
    >
    >
    > [noparse][[/noparse]Non-text portions of this message have been removed]


    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/



    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 2003-03-21 18:48
    I have seen a Cahn microbalance that used a similar principle. A light attached
    to the moving arm of the balance shone through a V-shaped slit onto a photocell.
    Amount on light proportional to the displacement. The instrument was capable of
    measuring weights
    down to tenths of milligrams.

    Bob
    nienhuis@w...

    Original Message
    From: Sadler Porter <porter.sadler@s...>
    Reply-To: basicstamps@yahoogroups.com
    Date: Fri, 21 Mar 2003 13:05:12 -0500

    >There was also a company called cyberoptics that did something along this in
    >many years ago.
    >
    >
    Original Message
    >From: Jim Forkin [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=GUfWiyQirPM0cpy5Iy5Mxn9KnU2RkyxFyKkAf7CQ65HteQOXQvxB02RmC7o6gKmumA8biPeK1Q]jjf@p...[/url
    >Sent: Friday, March 21, 2003 9:35 AM
    >To: basicstamps@yahoogroups.com
    >Subject: RE: [noparse][[/noparse]basicstamps] Re: Linear displacement
    >
    >Ken, Keyence makes laser sensors which can accurately sense distances but
    >they are a bit expensive. check out their site at www.keyence.com.
    >Jim
    >
    >
    Original Message
    >From: stan45039 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=rWlAnESxqvXJ_oWcjmAlynM2hQMnKEHz5R8aVk9tgEiR_Hv73-gx7PPaaJ3YJ5xV8btvSbzW1Q]slogan@f...[/url
    >Sent: Friday, March 21, 2003 3:39 AM
    >To: basicstamps@yahoogroups.com
    >Subject: [noparse][[/noparse]basicstamps] Re: Linear displacement
    >
    >
    >
    >(Warning: semi tongue in cheek reply follows)
    >
    >You could attach a mirror to the thing that will be displaced, shine a laser
    >on it, and measure the beam deflection on the side of a building several
    >blocks away [noparse]:)[/noparse]
    >
    >I shouldn't type when it's so long after my bedtime..g'night.
    >
    >
    >
    >--- In basicstamps@yahoogroups.com, smartdim@a... wrote:
    >> I am looking for suggestions for a hardware interface for my stamp. I want
    >to
    >> measure very small displacements (0 to 5mm) with 0.01mm accuracy. I don't
    >> think I want to use an LVDT. Perhaps an optical sensor, Linear Hall
    >effect,
    >> ??????????? I have some general ideas, but many heads are better than my
    >one
    >> bald one!
    >>
    >> Any suggestions welcomed.
    >>
    >> Ken
    >>
    >>
    >> [noparse][[/noparse]Non-text portions of this message have been removed]
    >
    >
    >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/
    >
    >
    >
    >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]
    >
    >
    >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 2003-03-21 19:25
    Along the same lines, laser interferometry has been used with
    microscopes to measure parts of the inner ear responding to sound, with
    (much) below micron resolutions.

    Dennis

    Original Message
    From: stan45039 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=wActtO3e5p7IJM4akEqXqSisY3NYfDXHk4xQ2RkHprJ32fixhYwHFutxMzizU5BVrj2FGvjZQt3A]slogan@f...[/url
    Sent: Friday, March 21, 2003 12:39 AM
    To: basicstamps@yahoogroups.com
    Subject: [noparse][[/noparse]basicstamps] Re: Linear displacement



    (Warning: semi tongue in cheek reply follows)

    You could attach a mirror to the thing that will be displaced, shine a
    laser on it, and measure the beam deflection on the side of a building
    several blocks away [noparse]:)[/noparse]

    I shouldn't type when it's so long after my bedtime..g'night.



    --- In basicstamps@yahoogroups.com, smartdim@a... wrote:
    > I am looking for suggestions for a hardware interface for my stamp. I
    > want to
    > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I
    don't
    > think I want to use an LVDT. Perhaps an optical sensor, Linear Hall
    effect,
    > ??????????? I have some general ideas, but many heads are better than
    my one
    > bald one!
    >
    > Any suggestions welcomed.
    >
    > Ken
    >
    >
    > [noparse][[/noparse]Non-text portions of this message have been removed]


    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 2003-03-21 21:11
    A linear measurement device is avalialbe that is really a 15 turn pot
    with a string and retrun spring.

    5 mm divided by 15 turns is 0.33 mm per revoltion.
    that would offer the resolution you are looking for, but you would
    need to interface it would an ADC.

    http://www.spaceagecontrol.com/ptmain.htm

    There are lots of digital calipers, some for $20.00 (like Harbor
    Freight) and a few places that the interface is defined. one problem
    is getting wires connected.

    Dave

    --- In basicstamps@yahoogroups.com, smartdim@a... wrote:
    > I am looking for suggestions for a hardware interface for my stamp.
    I want to
    > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I
    don't
    > think I want to use an LVDT. Perhaps an optical sensor, Linear
    Hall effect,
    > ??????????? I have some general ideas, but many heads are better
    than my one
    > bald one!
    >
    > Any suggestions welcomed.
    >
    > Ken
    >
    >
    > [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-21 21:15
    Also, if your device is not subject to vibration, a spring can act as
    a capacatance part in an RC loop.

    Beau has posted on that in the past and offered a circuit some time
    ago. I'm not sure of the accuracy, but again, an ADC would probably
    be nessicary.

    Dave



    --- In basicstamps@yahoogroups.com, smartdim@a... wrote:
    > I am looking for suggestions for a hardware interface for my stamp.
    I want to
    > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I
    don't
    > think I want to use an LVDT. Perhaps an optical sensor, Linear
    Hall effect,
    > ??????????? I have some general ideas, but many heads are better
    than my one
    > bald one!
    >
    > Any suggestions welcomed.
    >
    > Ken
    >
    >
    > [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-21 22:47
    Actually the spring functions as a COIL in an a LC loop, C is a fixed value.
    Similar to tuning an old radio (some actually moved a slug in and out of a
    coil for regular tuning) Personally though, I would use an LVDT. I'm not
    exactly sure why the original poster seems to be against using an LVDT, they
    can be extremely accurate. A laser seems to be a bit overkill. I would even
    be as bold as to roll my own LVDT coils for this type of project if it were
    up to me, but I'm odd that way. If you need a starter schematic let me know.

    Well, that's my quarter!
    -Beau Schwabe


    >Also, if your device is not subject to vibration, a spring can act as
    >a capacatance part in an RC loop.
    >
    >Beau has posted on that in the past and offered a circuit some time
    >ago. I'm not sure of the accuracy, but again, an ADC would probably
    >be nessicary.
    >
    >Dave
    >
    >
    >
    >--- In basicstamps@yahoogroups.com, smartdim@a... wrote:
    > > I am looking for suggestions for a hardware interface for my stamp.
    >I want to
    > > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I
    >don't
    > > think I want to use an LVDT. Perhaps an optical sensor, Linear
    >Hall effect,
    > > ??????????? I have some general ideas, but many heads are better
    >than my one
    > > bald one!
    > >
    > > Any suggestions welcomed.
    > >
    > > Ken
    > >
    > >
    > > [noparse][[/noparse]Non-text portions of this message have been removed]
    >
    >
    >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 2003-03-21 23:29
    In a message dated 3/21/2003 3:03:04 PM Pacific Standard Time,
    bschwabe@a... writes:

    > Actually the spring functions as a COIL in an a LC loop, C is a fixed value.
    > Similar to tuning an old radio (some actually moved a slug in and out of a
    > coil for regular tuning) Personally though, I would use an LVDT. I'm not
    > exactly sure why the original poster seems to be against using an LVDT,
    > they
    > can be extremely accurate. A laser seems to be a bit overkill. I would
    > even
    > be as bold as to roll my own LVDT coils for this type of project if it were
    > up to me, but I'm odd that way. If you need a starter schematic let me
    > know.
    >
    > Well, that's my quarter!
    > -Beau Schwabe

    Beau,

    I am not really "against" using an LVDT. I completely understand the
    principle. It appears that is the way to go. So you actually "roll" your own?
    I would like to try the same. If you can get me started in that area, I will
    try it.

    I guess my questions would be where to start. The deflections I am looking
    for are 0 to 5mm. Not sure where to start for excitation voltage and
    frequency. Thinking out loud, I will likely need to amplify the output,
    rectify it, etc, etc.

    I want the output to end up at TTL level so I can use the data with a uP for
    a custom application.

    Any thoughts are appreciated.

    I also may go with the Veneer caliper approach, that is hack/modify the
    caliper and use the serial output of the caliper....?????

    Ken


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-21 23:53
    At 06:29 PM 3/21/03 -0500, you wrote:
    >In a message dated 3/21/2003 3:03:04 PM Pacific Standard Time,
    >bschwabe@a... writes:
    >
    > > Actually the spring functions as a COIL in an a LC loop, C is a fixed
    > value.
    > > Similar to tuning an old radio (some actually moved a slug in and out of a
    > > coil for regular tuning) Personally though, I would use an LVDT. I'm not
    > > exactly sure why the original poster seems to be against using an LVDT,
    > > they
    > > can be extremely accurate. A laser seems to be a bit overkill. I would
    > > even
    > > be as bold as to roll my own LVDT coils for this type of project if it were
    > > up to me, but I'm odd that way. If you need a starter schematic let me
    > > know.
    > >
    > > Well, that's my quarter!
    > > -Beau Schwabe
    >
    >Beau,
    >
    >I am not really "against" using an LVDT. I completely understand the
    >principle. It appears that is the way to go. So you actually "roll" your own?
    > I would like to try the same. If you can get me started in that area, I
    > will
    >try it.
    >
    >I guess my questions would be where to start. The deflections I am looking
    >for are 0 to 5mm. Not sure where to start for excitation voltage and
    >frequency. Thinking out loud, I will likely need to amplify the output,
    >rectify it, etc, etc.
    >
    >I want the output to end up at TTL level so I can use the data with a uP for
    >a custom application.
    >
    >Any thoughts are appreciated.
    >
    >I also may go with the Veneer caliper approach, that is hack/modify the
    >caliper and use the serial output of the caliper....?????
    >
    >Ken

    Let me play with some values over the weekend, but yes I have "rolled" a wide
    variety of coils. 5mm detection isn't a problem. The output is a varying
    frequency that can easily be measured with a uP or Stamp. Off the top of my
    head this would require 2 I/O pins. One for selecting Coil1 or Coil2, the
    other
    for reading the frequency produced by the coil. With the design in mind, you
    could actually use a single coil, but you will experience drift due to
    temperature variations, adding a second coil to the design operating
    mechanically
    180 deg out of phase will negate any drift due to temperature variations
    and the
    values returned will relate to one another as a percentage relative to location
    of a metallic object between the coils.


    -Beau Schwabe
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-22 00:11
    Gentlefolk:

    A question and an idea in one. Would it be possible to cobble up an analog
    of an LVDT using Hall Effect devices and permanent magnets? Mount the magnet
    on the swing arm, suitably damped, and take the varying output of the Hall
    Device as your signal. Set up as a comparator (it is at rest in range +-
    allowable Hysteresis, or it isn't, with the Hall Device being one with a
    comparator output) you get a binary one or zero, but no scalar values. Use a
    linear output Hall Device and you get an analog scalar output. Feed it
    through an ADC and you get digital word output representative of the
    displacement. This eliminates coil winding and separate excitor sources.
    Probably more expensive though. I just think it is a more flexible answer,
    since it obviously lends itself to a lot of different apps.

    Mic
    Original Message
    From: <smartdim@a...>
    To: <basicstamps@yahoogroups.com>
    Sent: Friday, March 21, 2003 6:29 PM
    Subject: Re: [noparse][[/noparse]basicstamps] Re: Linear displacement


    > In a message dated 3/21/2003 3:03:04 PM Pacific Standard Time,
    > bschwabe@a... writes:
    >
    > > Actually the spring functions as a COIL in an a LC loop, C is a fixed
    value.
    > > Similar to tuning an old radio (some actually moved a slug in and out of
    a
    > > coil for regular tuning) Personally though, I would use an LVDT. I'm
    not
    > > exactly sure why the original poster seems to be against using an LVDT,
    > > they
    > > can be extremely accurate. A laser seems to be a bit overkill. I would
    > > even
    > > be as bold as to roll my own LVDT coils for this type of project if it
    were
    > > up to me, but I'm odd that way. If you need a starter schematic let me
    > > know.
    > >
    > > Well, that's my quarter!
    > > -Beau Schwabe
    >
    > Beau,
    >
    > I am not really "against" using an LVDT. I completely understand the
    > principle. It appears that is the way to go. So you actually "roll" your
    own?
    > I would like to try the same. If you can get me started in that area, I
    will
    > try it.
    >
    > I guess my questions would be where to start. The deflections I am looking
    > for are 0 to 5mm. Not sure where to start for excitation voltage and
    > frequency. Thinking out loud, I will likely need to amplify the output,
    > rectify it, etc, etc.
    >
    > I want the output to end up at TTL level so I can use the data with a uP
    for
    > a custom application.
    >
    > Any thoughts are appreciated.
    >
    > I also may go with the Veneer caliper approach, that is hack/modify the
    > caliper and use the serial output of the caliper....?????
    >
    > Ken
    >
    >
    > [noparse][[/noparse]Non-text portions of this message have been removed]
    >
    >
    > 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 2003-03-22 00:14
    Hmm. See my earlier post. Looks like I should have read more before, to talk
    less after! Sorry, folks.

    Mic
    Original Message
    From: Beau Schwabe <bschwabe@a...>
    To: <basicstamps@yahoogroups.com>
    Sent: Friday, March 21, 2003 5:47 PM
    Subject: Re: [noparse][[/noparse]basicstamps] Re: Linear displacement


    > Actually the spring functions as a COIL in an a LC loop, C is a fixed
    value.
    > Similar to tuning an old radio (some actually moved a slug in and out of a
    > coil for regular tuning) Personally though, I would use an LVDT. I'm not
    > exactly sure why the original poster seems to be against using an LVDT,
    they
    > can be extremely accurate. A laser seems to be a bit overkill. I would
    even
    > be as bold as to roll my own LVDT coils for this type of project if it
    were
    > up to me, but I'm odd that way. If you need a starter schematic let me
    know.
    >
    > Well, that's my quarter!
    > -Beau Schwabe
    >
    >
    > >Also, if your device is not subject to vibration, a spring can act as
    > >a capacatance part in an RC loop.
    > >
    > >Beau has posted on that in the past and offered a circuit some time
    > >ago. I'm not sure of the accuracy, but again, an ADC would probably
    > >be nessicary.
    > >
    > >Dave
    > >
    > >
    > >
    > >--- In basicstamps@yahoogroups.com, smartdim@a... wrote:
    > > > I am looking for suggestions for a hardware interface for my stamp.
    > >I want to
    > > > measure very small displacements (0 to 5mm) with 0.01mm accuracy. I
    > >don't
    > > > think I want to use an LVDT. Perhaps an optical sensor, Linear
    > >Hall effect,
    > > > ??????????? I have some general ideas, but many heads are better
    > >than my one
    > > > bald one!
    > > >
    > > > Any suggestions welcomed.
    > > >
    > > > Ken
    > > >
    > > >
    > > > [noparse][[/noparse]Non-text portions of this message have been removed]
    > >
    > >
    > >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/
    >
    >
    >
    > 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 2003-03-23 03:15
    I think Banner makes a Laser Displacement Sensor. Maybe something like
    that would work.

    www.bannerengineering.com (I think.)
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-24 15:20
    > > > Actually the spring functions as a COIL in an a LC loop, C is a fixed
    > > value.
    > > > Similar to tuning an old radio (some actually moved a slug in and out
    > of a
    > > > coil for regular tuning) Personally though, I would use an
    > LVDT. I'm not
    > > > exactly sure why the original poster seems to be against using an LVDT,
    > > > they
    > > > can be extremely accurate. A laser seems to be a bit overkill. I would
    > > > even
    > > > be as bold as to roll my own LVDT coils for this type of project if
    > it were
    > > > up to me, but I'm odd that way. If you need a starter schematic let me
    > > > know.
    > > >
    > > > Well, that's my quarter!
    > > > -Beau Schwabe
    > >
    > >Beau,
    > >
    > >I am not really "against" using an LVDT. I completely understand the
    > >principle. It appears that is the way to go. So you actually "roll" your
    > own?
    > > I would like to try the same. If you can get me started in that area, I
    > > will
    > >try it.
    > >
    > >I guess my questions would be where to start. The deflections I am looking
    > >for are 0 to 5mm. Not sure where to start for excitation voltage and
    > >frequency. Thinking out loud, I will likely need to amplify the output,
    > >rectify it, etc, etc.
    > >
    > >I want the output to end up at TTL level so I can use the data with a uP for
    > >a custom application.
    > >
    > >Any thoughts are appreciated.
    > >
    > >I also may go with the Veneer caliper approach, that is hack/modify the
    > >caliper and use the serial output of the caliper....?????
    > >
    > >Ken
    >
    >Let me play with some values over the weekend, but yes I have "rolled" a wide
    >variety of coils. 5mm detection isn't a problem. The output is a varying
    >frequency that can easily be measured with a uP or Stamp. Off the top of my
    >head this would require 2 I/O pins. One for selecting Coil1 or Coil2, the
    >other
    >for reading the frequency produced by the coil. With the design in mind, you
    >could actually use a single coil, but you will experience drift due to
    >temperature variations, adding a second coil to the design operating
    >mechanically
    >180 deg out of phase will negate any drift due to temperature variations
    >and the
    >values returned will relate to one another as a percentage relative to
    >location
    >of a metallic object between the coils.
    >
    >
    >-Beau Schwabe

    In your original post you expressed a desire for 0.01mm accuracy. Over the
    weekend
    I was able to achieve an approximate .003mm over the range of 1cm. below
    are my
    results. Instead of having a secondary coil 180 out of phase, the only
    difference
    is that the secondary coil is used only as a reference for the sense
    coil. However
    with the same circuitry, you could still arrange the coils so that they are
    both
    operational.

    http://www.angelfire.com/wizard/y2kbc/LVDT/

    -Beau Schwabe
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-24 16:34
    In a message dated 3/24/2003 7:31:03 AM Pacific Standard Time,
    bschwabe@a... writes:

    > In your original post you expressed a desire for 0.01mm accuracy. Over the
    > weekend
    > I was able to achieve an approximate .003mm over the range of 1cm. below
    > are my
    > results. Instead of having a secondary coil 180 out of phase, the only
    > difference
    > is that the secondary coil is used only as a reference for the sense
    > coil. However
    > with the same circuitry, you could still arrange the coils so that they are
    >
    > both
    > operational.
    >
    > http://www.angelfire.com/wizard/y2kbc/LVDT/
    >
    > -Beau Schwabe
    >

    Beau,

    AWESOME!!!!!!!!!!!!!!

    Very nice. Thank you very much. You provided more than enough to get me
    going.

    Thank you very much!!

    Ken


    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-26 02:16
    Great stuff! I'm puzzled about the resolution figures at
    the bottom of the page - Any chance there's an extra zero
    in there? 300 counts per cm is 30000 counts per meter, or
    1 count per .0000333 meter, which is .0333 mm. What am I
    missing?

    best regards, Jack

    Beau Schwabe wrote:
    > ...
    > http://www.angelfire.com/wizard/y2kbc/LVDT/
  • ArchiverArchiver Posts: 46,084
    edited 2003-03-26 16:01
    At 06:16 PM 3/25/03 -0800, you wrote:
    >Great stuff! I'm puzzled about the resolution figures at
    >the bottom of the page - Any chance there's an extra zero
    >in there? 300 counts per cm is 30000 counts per meter, or
    >1 count per .0000333 meter, which is .0333 mm. What am I
    >missing?
    >
    >best regards, Jack
    >
    >Beau Schwabe wrote:
    > > ...
    > > http://www.angelfire.com/wizard/y2kbc/LVDT/

    Ohhh Rats! I hate it when I do that! Jack, your right it
    should be cm instead of mm.

    However, here is a case where it might be better to use the COUNT
    command over the PULSIN command. Consider the original frequency
    data with the addition of a new coil I wound with 30 turns.

    Test# Option Pressed Released Coil
    1 C 194.2Hz 192.9Hz 100T
    2 HCT 529.1Hz 401.6Hz 100T
    3 HC 507.6Hz 397.6Hz 100T
    4 HCT 1724.0Hz 714.3Hz 30T (New Test)


    Return Values for PULSIN:
    Test# PULSIN PULSIN Delta Res/cm
    1 2574 2592 18 .0555
    2 945 1245 300 .0033
    3 985 1257 272 .0037
    4 290 699 409 .0024

    Return Values for a COUNT duration of 125mS:
    Test# COUNT COUNT Delta Res/cm
    1 24 24 1 1
    2 66 50 16 .0625
    3 63 49 14 .0714
    4 215 89 126 .0079

    Return Values for a COUNT duration of 1000mS:
    Test# COUNT COUNT Delta Res/cm
    1 194 192 2 .5
    2 529 401 128 .0078
    3 507 397 110 .0091
    4 1724 714 1010 .001

    Return Values for a COUNT duration of 2000mS:
    Test# COUNT COUNT Delta Res/cm
    1 388 385 3 .3333
    2 1058 803 255 .0039
    3 1015 795 220 .0045
    4 3448 1428 2020 .0005

    -Beau Schwabe



    [noparse][[/noparse]Non-text portions of this message have been removed]
Sign In or Register to comment.