Shop OBEX P1 Docs P2 Docs Learn Events
1-Wire Weather and the BS2p — Parallax Forums

1-Wire Weather and the BS2p

ArchiverArchiver Posts: 46,084
edited 2003-04-04 03:24 in General Discussion
Hi everyone,

I have a BS2p and and three 1-Wire sensors (humidity, temperature and
rain). The BS2p and sensors are about 30 feet from each other on cat 5E
cable. Just to get them to work I had to lower the pull-up resistor to 500
ohms and ground 1 unused pair in the cat 5. I still get a lot of bad data at
times. I plan to add the weather station from AAG in the near future. Anyone
have any suggestions on how to clean up the data. Using a 1k pull up it
wouldn't work at all.

Thanks in advance,
Bill

Comments

  • ArchiverArchiver Posts: 46,084
    edited 2003-04-02 04:16
    Hi Bill:

    Had the exact problem when doing a network of 16 DS1820's on 300' of cable.
    A lot has to do with the way you tap off the cat 5 backbone cable and the
    reflections that are caused on the OWB. If your tap is greated than a few
    feet reflections cause unreliable results. One cure is something I
    discovered in the mounds of tech data and that was installing a 100 ohme in
    series with the data lead only at the stub connection to the backbone. It
    is detailed on Tech Note from Dallas. I believe the title was OWN
    reliability (Yea Sure) . When you first read about the OWB it sounds great
    but after you get more than a few feet away and have nultiple sensors
    everything goes to pot. It took me months to overcome this problem and their
    tech help never returned my messages. Consequently when I found the problem
    I will remember the fix forever.

    Eric

    Good Luck
    Original Message
    From: "Bill Cramer" <billcramer@a...>
    To: <basicstamps@yahoogroups.com>
    Sent: Tuesday, April 01, 2003 8:15 PM
    Subject: [noparse][[/noparse]basicstamps] 1-Wire Weather and the BS2p


    > Hi everyone,
    >
    > I have a BS2p and and three 1-Wire sensors (humidity, temperature and
    > rain). The BS2p and sensors are about 30 feet from each other on cat 5E
    > cable. Just to get them to work I had to lower the pull-up resistor to 500
    > ohms and ground 1 unused pair in the cat 5. I still get a lot of bad data
    at
    > times. I plan to add the weather station from AAG in the near future.
    Anyone
    > have any suggestions on how to clean up the data. Using a 1k pull up it
    > wouldn't work at all.
    >
    > Thanks in advance,
    > Bill
    >
    >
    > 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 2003-04-02 04:34
    Someone could write a book on methods to clean up signals. Oh wait, 32,768
    people already did!

    Seriously though, you haven't told us much about your circuit so I'll toss
    out a few ideas, you may already know/have done them.

    First, is the cat-5 shielded? If so, have you grounded the stamp end of the
    shield? (Dont connect the far end to ground). This will help keep EMFs from
    interfering with the data too much.

    Second, are the three sensors sharing a single cable? If so, they could be
    inducing currents in each other and messing up the data (shot in the dark)

    What do you have the circuit laid out on? Carrier board, breadboard, wire
    wrap, PCB?

    My 'lectronic Newb site http://members.shaw.ca/pmeloy/
    Original Message
    From: "Bill Cramer" <billcramer@a...>
    To: <basicstamps@yahoogroups.com>
    Sent: Tuesday, April 01, 2003 5:15 PM
    Subject: [noparse][[/noparse]basicstamps] 1-Wire Weather and the BS2p


    > Hi everyone,
    >
    > I have a BS2p and and three 1-Wire sensors (humidity, temperature and
    > rain). The BS2p and sensors are about 30 feet from each other on cat 5E
    > cable. Just to get them to work I had to lower the pull-up resistor to 500
    > ohms and ground 1 unused pair in the cat 5. I still get a lot of bad data
    at
    > times. I plan to add the weather station from AAG in the near future.
    Anyone
    > have any suggestions on how to clean up the data. Using a 1k pull up it
    > wouldn't work at all.
    >
    > Thanks in advance,
    > Bill
    >
    >
    > 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 2003-04-02 13:48
    Good morning,

    Thanks Pat and Eric for the response.
    To answer Pat's questions, no the cat 5 is not shielded, and the sensors are
    from Jim Jennings (www.jjware.com). One is a temp/humidity sensor and the other
    is a general purpose counter module. The cat 5 runs from the Stamp about 30 feet
    to the temp/humidity module. Then a 2 foot cat 5 jumper connects to the counter
    which is connected to my rain guage. As I told you the pull-up is at 500 ohms
    and one unused pair of the cat 5 is grounded at the Stamp.

    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-04-02 15:07
    Bill,

    Do you have filter caps sprinkled on the board?

    Jonathan

    www.madlabs.info

    Original Message
    From: "Bill Cramer" <billcramer@a...>
    To: <basicstamps@yahoogroups.com>
    Sent: Wednesday, April 02, 2003 4:48 AM
    Subject: [noparse][[/noparse]basicstamps] 1-wire weather and the BS2p


    > Good morning,
    >
    > Thanks Pat and Eric for the response.
    > To answer Pat's questions, no the cat 5 is not shielded, and the sensors
    are from Jim Jennings (www.jjware.com). One is a temp/humidity sensor and
    the other is a general purpose counter module. The cat 5 runs from the Stamp
    about 30 feet to the temp/humidity module. Then a 2 foot cat 5 jumper
    connects to the counter which is connected to my rain guage. As I told you
    the pull-up is at 500 ohms and one unused pair of the cat 5 is grounded at
    the Stamp.
    >
    > [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 2003-04-03 00:10
    Ok, I'm just a newb at this too but I've been doing a lot of reading. I'm
    dealing with Analog sensors with currents down in the femto-amperes (berry
    berry small!) so signal conditioning and noise-harm reduction is what I've
    been concentrating on for the last while. Haven't really thought about how
    these things could affect digital signals so keep that in mind.

    The biggies:

    Seperate cables for the sensors may help, though since you're only talking
    to one sensor at a time it might be useless - who cares if sensor two has
    induced currents if you aren't talking to it at the time?

    Shield the cable - Probably a good idea. RS232 cables off a computer have a
    practical length limit around 25 feet (depends on lot on the environment).
    Any longer and you need "booster" boxes. Shielding keeps RF from messing up
    signals (very big effect on my femto-ampere sensor, digital I dunno). Just
    remember to connect the shield ground at the MCU end and leave the sensor
    end un-connected (If using an ADC, make sure the shield is grounded at the
    same point your REF- is connected). Sometimes you want to use a powered
    shield, +5v or whatever. Haven't figured out when that is required yet.
    Still learning!

    Decouple the power at your chips - Whenever a chip does something it draws
    power. When its not doing anything it draws very little power. The
    transition between the two states causes a power pulse, this pulse causes EM
    emissions which can induce noise into the rest of your circuit. The longer
    the power trace/wire, the bigger the transmitting antenna it becomes. Place
    a 0.1 or 1.0uF cap right by each chip's power pin. The power supply will
    charge the cap and that section of the wiring won't have the pulses. When
    the chip draws more power, the Cap will supply the current - since it is
    nice and close to the chip your transmitting antenna is nice and small.

    Place a low-pass filter on the signal line to cancel out 60hz noise (or 50hz
    if you're in Europe/Asia). For 60hz you can use a 0.1uF tantalum cap and 80
    ohm resistor in series to ground. There are also high-pass and inductance
    filters which I haven't researched yet.

    Use a single ground point - The sensors and chip should be hooked to ground
    at the same point. Its possible to get different potentials if the grounds
    are far apart. The Pic needs (I think) 1.2V or higher to consider the pin
    "high". If you have a voltage drop of 3.7v and the grounds are 0.2v apart,
    you'd only get 1.1v and the stamp wouldn't acknowledge the high state.

    The smaller resistor isn't all that unusual. 1k-10k pullups are common but
    those assume the transmitter is on the same PCB - only a few inches away.

    If worst comes to worse, you can also add a 1 x gain op-amp before the
    stamp. Some op-amps have a built in frequency rejection and they can clean
    up square waves real nice. If your signal voltages come in too low you can
    add some gain to get it into the right range as well (Only read about this a
    couple of days ago).

    Thats all I know so far.


    My 'lectronic Newb site http://members.shaw.ca/pmeloy/
    Original Message
    From: "Bill Cramer" <billcramer@a...>
    To: <basicstamps@yahoogroups.com>
    Sent: Wednesday, April 02, 2003 4:48 AM
    Subject: [noparse][[/noparse]basicstamps] 1-wire weather and the BS2p


    > Good morning,
    >
    > Thanks Pat and Eric for the response.
    > To answer Pat's questions, no the cat 5 is not shielded, and the sensors
    are from Jim Jennings (www.jjware.com). One is a temp/humidity sensor and
    the other is a general purpose counter module. The cat 5 runs from the Stamp
    about 30 feet to the temp/humidity module. Then a 2 foot cat 5 jumper
    connects to the counter which is connected to my rain guage. As I told you
    the pull-up is at 500 ohms and one unused pair of the cat 5 is grounded at
    the Stamp.
    >
    > [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 2003-04-03 00:15
    Oop, just to be clear - for the decoupling cap, its in parallel, not series,
    with the power supply, In other words, power goes directly to the chips and
    the cap goes from the power line to ground. You probably know that but, just
    in case.

    My 'lectronic Newb site http://members.shaw.ca/pmeloy/
    Original Message
    From: "Pat M" <pmeloy@s...>
    To: <basicstamps@yahoogroups.com>
    Sent: Wednesday, April 02, 2003 3:10 PM
    Subject: Re: [noparse][[/noparse]basicstamps] 1-wire weather and the BS2p
  • ArchiverArchiver Posts: 46,084
    edited 2003-04-03 23:28
    Some observations:

    For 30 feet on a single wire, make sure that a good
    ground reference is present to all devices. Also
    follow the advice about appropriate amounts of filter
    capacitors for all devices to ensure the supply
    voltage is DC.

    Are the drive levels from the sensors sufficient to
    travel the entire cable distance from the BS2P and to
    all the sensors? Even though only one device is
    talking on the line, the signal still has to travel to
    each sensor just the same; thus adding further wire
    length between a sensor to the Stamp. I would consider
    having each sensor on a separate line. This will be
    more of an issue at 500 feet as each sensor creates a
    load on the wire. That is why you are having to change
    your termination resistor value from the reference
    design.

    Having _shielded_ Cat 5 cable is not a necessity, but
    why not take advantage of the twisted pair
    characteristic of the Cat 5 cable (noise immunity) and
    use differential transcievers such as 75176s for
    RS-485 or similar for RS-422? RS-485 and RS-422 are
    rated to about 4000 feet between nodes for slow speed
    communications. RS-485 can also be used for multipoint
    wiring applications. Wiring Tip and Ring leads of each
    pair is not recommended as this does something
    different (not necessarily nice) for the signalling
    you are trying to put through on the cable.

    In your application, where are both devices going to
    be located? If they are going to be both mounted on a
    mast, why not located the Stamp out there also. This
    would keep the 1-wire links short, and longer haul
    techinques such as RS-422 or RS-485 drivers can be
    employed to bring the data down to a host PC. Use
    dedicated I/O pins for each device to cut down on
    loading or the I/O pins. Power for all devices can be
    put down 2 of the pairs in the Cat 5 group - 1 for
    +%VDC and the other for return (ground). To be in
    spec, the +5VDC should be higher to compensate for
    Ohm's Law losses in the cable run and the actual
    devices under power. This technique is similar to what
    is being consider for the PoE (Power over Ethernet)
    standard.

    kevin / kc6pob

    --- Bill Cramer <billcramer@a...> wrote:
    > Hi everyone,
    >
    > I have a BS2p and and three 1-Wire sensors
    > (humidity, temperature and
    > rain). The BS2p and sensors are about 30 feet from
    > each other on cat 5E
    > cable. Just to get them to work I had to lower the
    > pull-up resistor to 500
    > ohms and ground 1 unused pair in the cat 5. I still
    > get a lot of bad data at
    > times. I plan to add the weather station from AAG in
    > the near future. Anyone
    > have any suggestions on how to clean up the data.
    > Using a 1k pull up it
    > wouldn't work at all.
    >
    > Thanks in advance,
    > Bill
    >
    >
    > 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! Tax Center - File online, calculators, forms, and more
    http://tax.yahoo.com
  • ArchiverArchiver Posts: 46,084
    edited 2003-04-04 03:10
    Hi Bill,

    If you or anyone else has BS2P code for running the (newish) AAG Version
    3.0 One Wire weather station, please let me know. The old article in
    Nuts & Volts magazine was not for V3.0.

    Thanks,
    Chris

    <<>>

    Bill Cramer <billcramer@a...> wrote:
    > Hi everyone,
    >
    > I have a BS2p and and three 1-Wire sensors
    > (humidity, temperature and
    > rain). The BS2p and sensors are about 30 feet from
    > each other on cat 5E
    > cable. Just to get them to work I had to lower the
    > pull-up resistor to 500
    > ohms and ground 1 unused pair in the cat 5. I still
    > get a lot of bad data at
    > times. I plan to add the weather station from AAG in
    > the near future. Anyone
    > have any suggestions on how to clean up the data.
    > Using a 1k pull up it
    > wouldn't work at all.
    >
    > Thanks in advance,
    > Bill

    ________________________________________________________________
    Sign Up for Juno Platinum Internet Access Today
    Only $9.95 per month!
    Visit www.juno.com
  • ArchiverArchiver Posts: 46,084
    edited 2003-04-04 03:24

    Original Message
    From: <cdundorf@j...>
    To: <basicstamps@yahoogroups.com>
    Sent: Thursday, April 03, 2003 9:10 PM
    Subject: [noparse][[/noparse]basicstamps] Re: 1-Wire Weather and the BS2p


    > Hi Bill,
    >
    > If you or anyone else has BS2P code for running the (newish) AAG Version
    > 3.0 One Wire weather station, please let me know. The old article in
    > Nuts & Volts magazine was not for V3.0.

    Chris

    I haven't purchased the weather station from AAG yet since I'm having a lot
    of trouble with my temperature/humidity board and rain guage counter board
    and 1-wire bus. However the software I downloaded from
    http://homepage.mac.com/tbitson/weather/index.html should work.It addersses
    the same 1-wire chips as is in the V3.0 weather station.
Sign In or Register to comment.