+ Post New Thread
Results 1 to 13 of 13
  1. #1
    Junior Member level 1
    Points: 83, Level: 1

    Join Date
    Jun 2019
    Location
    Chennai, India
    Posts
    15
    Helped
    0 / 0
    Points
    83
    Level
    1

    Problems on Sending RS232 Signals through 3 Meter Core Cable

    Hello Everyone,
    Greetings. I have developed a circuit board with master slave concept. Master sending RS-232 Data( I am using SP-3232 DIP IC from Exar- Bought from Mouser). Slave just Receive the data. Most of time communication is fine. But sometime Communication is not happening

    We are using 5 core Shielded cable which is in 4.5 Meter length.
    1-----> 15VDC( 5A)
    2----->GND
    3-----> RX
    4----->TX
    5----->OE( To Enable Communication @3v3 Volt)

    We are manufacturing this product more than a year. Out of 100 boards, 15 number of boards are making problems. If We change IC it is working for some times. I am wondering some time If I touch over the IC, communication is ok..

    What would be the reason??
    Environment is so simple. Just We have a AC motor( 3 Phase). Induction wise no problem I thing . Everything is properly earthed. To power up our Device We using Meanwell SMPS( NES-350-15)

  2. #2
    Super Moderator
    Points: 29,576, Level: 41
    andre_teprom's Avatar
    Join Date
    Nov 2006
    Location
    Brazil
    Posts
    8,837
    Helped
    1114 / 1114
    Points
    29,576
    Level
    41
    Blog Entries
    9

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Does this communication occur between periphreal board and a PC desktop/laptop? If so, is it done through a USB adapter or something else? Do you have a diagram of the electrical scheme you can show?

    Just We have a AC motor( 3 Phase). Induction wise no problem I thing .
    Carrying 3v3 across a 4,5m cable, how can you be sure? BTW, it would be advisable think of using logical address to each device through serial protocol instead of a low level signal acting as device 'selector'. The RS-232 electrical standard is more suitable to deal with devices appart each other.
    --------------------------------------------------------------------------------------------------
    Part of the world that you live in, You are the part that you're giving ( Renaissance )


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  3. #3
    Super Moderator
    Points: 77,026, Level: 67
    Achievements:
    7 years registered
    Awards:
    Most Frequent Poster 3rd Helpful Member

    Join Date
    Apr 2014
    Posts
    15,607
    Helped
    3556 / 3556
    Points
    77,026
    Level
    67

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Hi,

    Show the exact circuit...from master side...which IC, which pin drives which line of which cable...and the complete slave receiver circuit.
    ESD protection including wiring and PCB layout.
    Connectors, if used....
    All GND (and Earth) connections to recognize a loop.
    Why 5A? What's the return path of the 5A? Hopefully not the signal_GND of the RS232....

    How is the motor involved?
    How is the power supply involved?
    Why the headline talks about 3m while the text says 4.5m?

    What exactly means: "communication is not happening"? Fault at the master side or slave side? How do the signals look like? Idle level, active level...

    Klaus
    Please donīt contact me via PM, because there is no time to respond to them. No friend requests. Thank you.


    1 members found this post helpful.

  4. #4
    Junior Member level 1
    Points: 83, Level: 1

    Join Date
    Jun 2019
    Location
    Chennai, India
    Posts
    15
    Helped
    0 / 0
    Points
    83
    Level
    1

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Quote Originally Posted by andre_teprom View Post
    Does this communication occur between periphreal board and a PC desktop/laptop? If so, is it done through a USB adapter or something else? Do you have a diagram of the electrical scheme you can show?



    Carrying 3v3 across a 4,5m cable, how can you be sure? BTW, it would be advisable think of using logical address to each device through serial protocol instead of a low level signal acting as device 'selector'. The RS-232 electrical standard is more suitable to deal with devices apart each other.
    Hello! Thanks for your prompt response.
    Communication is happening between Two MCU boards. There is no more Laptop/computer involved. We are fetching some data from SD card and We sending the data(Binary<2058 Bits/One RPM, Max speed 120 RPM speed). Slave board just receive the data and it will engage 2048 Solenoids as per the data)Attachment 153945

    We have no idea about how to initiate the data communication. So that We have used One IO directly to start communication. Yet now We have not faced any problem on this part. So we feel ok with that Circuit. Kindly refer some links to do this jon without any IO through RS232. Else you can guide me other logical circuits to send the TTL data over the lengthy cable(4.5 mtr)

    Is there any problem will be if we send RS-232 data through over 5 Meter cable? Cables are properly shielded. I am not saying every problem is making trouble. But out of 100 few making serious issues..
    We using SP3232 DIP IC due to lowest cost. It is not having any internal ESD protection diodes. Will it cause any trouble. Through one RMC connector( 5 Pin Header) pin We are connecting the data cables.
    Let me know if there will be any Spikes will cause these problems..

    I am explaining strange problem happening with my product. If We touch the IC through our finger It is working without any problem.. I am getting doubt on 3 things.

    1. ESD problem( If so Please suggest some ESD protection diodes I will add VCC and Data lines)
    2. Capacitor We have used. 2 Cpacitors are 0.1uF. Other 3 are 0.47uF.. I have seen everyone is using 0.1uF. But in data sheet they have recommended 0.47uf for VCC 3-5.5V
    3. Grounding miss match( Does metal body should be connected to DC ground). Slave board is power by master. We sending power through Data cable

    - - - Updated - - -

    Quote Originally Posted by KlausST View Post
    Hi,

    Show the exact circuit...from master side...which IC, which pin drives which line of which cable...and the complete slave receiver circuit.
    ESD protection including wiring and PCB layout.
    Connectors, if used....
    All GND (and Earth) connections to recognize a loop.
    Why 5A? What's the return path of the 5A? Hopefully not the signal_GND of the RS232....

    How is the motor involved?
    How is the power supply involved?
    Why the headline talks about 3m while the text says 4.5m?

    What exactly means: "communication is not happening"? Fault at the master side or slave side? How do the signals look like? Idle level, active level...

    Klaus
    Attachment 153946

    I have attached circuit. Master and slave circuits are same. No changes in schematic or components used. We have a RMC connector to connect data cable. Female connector is directly soldered with 5 Core Cable and of the cable is soldered with DB-9 Male connector. So that it will be detachable. Mother board having DB9 female connector and Slave board having RMC Attachment 153947

    We have not done anything for ESD protection. Data cable used to send Power as well as RS-232 signals. No separate GND used for RS-232 since all grounds are looped in PCB lay out.
    We are not using 5A exactly. Mother board is just consuming 500mA current. Slave board feeds the power to Mother board.
    Motor is not used for our purpose it is supposed to actuate mechanical levers. Separate Lines are used for power up the Motor.

    We using Mean well SMPS for Power up our device. it is 23.3A/15V DC.. Since We using many solenoids maximum current needed. Solenoids to getting the power from same SMPS..
    I apologize for the wrong info about cable. Cable is 4.5 Meters in length. Shielded 5 core Cable and Shield is left free

    I have no idea about checking the Signal status. If We use CRO shall We measure?
    I am sure communication is not happening. If it is happening We have some LEDs to check out the status..

    Master will send data. Slave will receive. It is One way communication. Communication will happen after the OE pin is getting Low( OE pin means We have dedicated IO to start sending the data



  5. #5
    Super Moderator
    Points: 259,183, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    45,250
    Helped
    13766 / 13766
    Points
    259,183
    Level
    100

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    May be just a typo, C39 (MAX232 V-) is reversed in the schematic.



    •   AltAdvertisement

        
       

  6. #6
    Junior Member level 1
    Points: 83, Level: 1

    Join Date
    Jun 2019
    Location
    Chennai, India
    Posts
    15
    Helped
    0 / 0
    Points
    83
    Level
    1

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Hello,
    Thanks FvM.. You are exactly correct. I will correct in PCB manually and I will let you know..
    For your information I am using 5V VCC.. If I use 0.47uF capacitor don't be problem right???
    Otherwise Shall I use 0.1uF AEC disc capacitors as Everyone using.. I hope baud rate is not big for our application. Just We sending 2100 Bits/RPM(120 RPM max speed)

    Previously We have used 0.1uF capacitors but We could not find any difference in that. Suggest some links to Test( Testing procedure or Test must do) the PCB before delivering to client end



    •   AltAdvertisement

        
       

  7. #7
    Super Moderator
    Points: 77,026, Level: 67
    Achievements:
    7 years registered
    Awards:
    Most Frequent Poster 3rd Helpful Member

    Join Date
    Apr 2014
    Posts
    15,607
    Helped
    3556 / 3556
    Points
    77,026
    Level
    67

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Hi,

    It's all very confusing ... at least to me...

    You gave just a little snippet of information/schematic of what I asked for in post #3.
    This makes helping difficult/impossible ....
    I assume you have good reasons not to give some informations...thus I don't ask again.
    ***

    Just one example:
    We are not able to check the voltage drop across the GND line of the cable, because:
    * we don't know cable type and wire diameter
    * return path of power
    * we are confused if there is 5A, or 0.5A. Average or pulsed?
    * whether the connectors are suitable

    With your last post some more questions arise:
    * Why do you use a shielded cable, but don't connect the shield?
    * you give a variable data rate, depending in RPM.... don't you use a fixed baud rate?
    * if no motor is used, why do you talk about RPM at all? (As far as I understand: No motor --> no RPM --> no data transfer)
    * how the solenoids are involved
    * why that big 23A SMPS?
    * your schematic shows unconnected input pins at the RS232 driver...are you sure they are not floating?

    Klaus

    From your other post:
    Why do you say "It is not having any internal ESD protection diodes", while tha datasheet on the first page and other pages specify ESD rating. Maybe you use other manufacturer than SIPEX?
    Please donīt contact me via PM, because there is no time to respond to them. No friend requests. Thank you.


    1 members found this post helpful.

  8. #8
    Junior Member level 1
    Points: 83, Level: 1

    Join Date
    Jun 2019
    Location
    Chennai, India
    Posts
    15
    Helped
    0 / 0
    Points
    83
    Level
    1

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Quote Originally Posted by Prabhakarankft View Post
    Hello,
    Thanks FvM.. You are exactly correct. I will correct in PCB manually and I will let you know..
    For your information I am using 5V VCC.. If I use 0.47uF capacitor don't be problem right???
    Otherwise Shall I use 0.1uF AEC disc capacitors as Everyone using.. I hope baud rate is not big for our application. Just We sending 2100 Bits/RPM(120 RPM max speed)

    Previously We have used 0.1uF capacitors but We could not find any difference in that. Suggest some links to Test( Testing procedure or Test must do) the PCB before delivering to client end
    Hello FvM,
    Kindly find attached image. SP3232 IC pin configuration is slightly differs from MAX3232 Check at 6th PinClick image for larger version. 

Name:	SIPEX.PNG 
Views:	5 
Size:	92.5 KB 
ID:	153949



  9. #9
    Super Moderator
    Points: 259,183, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    45,250
    Helped
    13766 / 13766
    Points
    259,183
    Level
    100

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    I don't believe that any datasheet shows the V- bypass capacitor reversed, your schematic however does:

    Click image for larger version. 

Name:	c39.PNG 
Views:	0 
Size:	11.2 KB 
ID:	153950


    1 members found this post helpful.

  10. #10
    Junior Member level 1
    Points: 83, Level: 1

    Join Date
    Jun 2019
    Location
    Chennai, India
    Posts
    15
    Helped
    0 / 0
    Points
    83
    Level
    1

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Quote Originally Posted by KlausST View Post
    Hi,

    It's all very confusing ... at least to me...

    You gave just a little snippet of information/schematic of what I asked for in post #3.
    This makes helping difficult/impossible ....
    I assume you have good reasons not to give some informations...thus I don't ask again.
    ***

    Just one example:
    We are not able to check the voltage drop across the GND line of the cable, because:
    * we don't know cable type and wire diameter
    * return path of power
    * we are confused if there is 5A, or 0.5A. Average or pulsed?
    * whether the connectors are suitable

    With your last post some more questions arise:
    * Why do you use a shielded cable, but don't connect the shield?
    * you give a variable data rate, depending in RPM.... don't you use a fixed baud rate?
    * if no motor is used, why do you talk about RPM at all? (As far as I understand: No motor --> no RPM --> no data transfer)
    * how the solenoids are involved
    * why that big 23A SMPS?
    * your schematic shows unconnected input pins at the RS232 driver...are you sure they are not floating?

    Klaus

    From your other post:
    Why do you say "It is not having any internal ESD protection diodes", while tha datasheet on the first page and other pages specify ESD rating. Maybe you use other manufacturer than SIPEX?
    Hello!
    I am thanking you for thinking about my issue. I will explain about your doubt as my level best

    We could not measure the voltage drop as of my knowledge Since Master and slaves placed with 5 meter distance. Anyhow We will bring back the slave device and We will measure if there is any drop over the GND lines.. Is there any need to check voltage drop on all the lines?

    Specification of Cable: 14/36 Core Outer Dia: 1.6mm.

    Return path of power is same Core cable. VCC and GND lines are separately given. I am not getting exactly what you mean by Return path of power. But for power We use only two lines

    Power consumption is ~500mA and it is continues current. But We have given extra room space for more current flow( We meant Cable thickness of core cable. But not sure about this)

    Is there any problems with RMC connectors? I am not sure about the connectors.. 80% Okay based upon my present experience

    24576 baud Rate

    Motors are used for different purpose. For easy understanding For every rotation We will send data from Master to Slave

    23.3A SMPS used for energies the solenoids and 120 Solenoids are used. Solenoids will consume ~0.2A( At 15VDC). Solenoids are connected to another board. We will send just data to that board. Slave will take care of that part

    By reading the data sheet of SIPEX IC it is having inbuilt ESD controller circuit.So shall We confirm this is not due to ESD??



  11. #11
    Junior Member level 1
    Points: 83, Level: 1

    Join Date
    Jun 2019
    Location
    Chennai, India
    Posts
    15
    Helped
    0 / 0
    Points
    83
    Level
    1

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Quote Originally Posted by FvM View Post
    I don't believe that any datasheet shows the V- bypass capacitor reversed, your schematic however does:

    Click image for larger version. 

Name:	c39.PNG 
Views:	0 
Size:	11.2 KB 
ID:	153950
    OMG! I will correct my circuit.. Sorry for my bad understanding



  12. #12
    Super Moderator
    Points: 77,026, Level: 67
    Achievements:
    7 years registered
    Awards:
    Most Frequent Poster 3rd Helpful Member

    Join Date
    Apr 2014
    Posts
    15,607
    Helped
    3556 / 3556
    Points
    77,026
    Level
    67

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Hi,

    We could not measure the voltage drop
    No urgent need to measure...but measurement gives exact values with your exact conditions...

    Return path of power is same Core cable. VCC and GND lines are separately given. I am not getting exactly what you mean by Return path of power. But for power We use only two lines
    If 5A (or 0.5A) goes the one way, it has to come back somehow...this is the return path.

    Usually you should differ "no current" GND wires - as signal reference ... and current carrying power GND wires.
    The signal (GND) wires should carry not much current, thus there should be ignorable voltage drop.
    While at the power path there will be unavoidable voltage drop.

    A 14/36 cable has about 50mOhms/m.
    Multiply this with 5m, then you get about 250mOhms for 5m per wire.
    Multiply this with 0.5A and get 125mV per wire.
    This 125mV is not critical as long as it is DC....but in case it is pulsed current you have to calculate with peak currents.

    Is there any problems with RMC connectors?
    Not as long as they comply with your requirements...current, resistance, voltage, voltage drop, thermal...
    You need to read the datasheet.

    24576 baud Rate
    Should work.
    But I wonder why no standard baud rate...

    With a standard baud rate you could simply use any PC or Laptop with (USB to) RS232 to debug the communication.

    So shall We confirm this is not due to ESD??
    ESD grade depends on your environment ... we can not say what may happpen...

    Can you be sure the costomer never connects/disconnects the cable when power is supplied.
    If this may happen, then you need to decide that one contact will connect earlier than any other.
    Often the worst case is when VCC is first connected...then a signal line....so the signal line works (temporarily) as return path and may carry a lot of (peak) current. ...while it usually is not designed for this situation.

    To avoid this problem there are connectors with leading GND contacts...to ensure GND is connected at first.

    Klaus
    Please donīt contact me via PM, because there is no time to respond to them. No friend requests. Thank you.



    •   AltAdvertisement

        
       

  13. #13
    Advanced Member level 2
    Points: 5,076, Level: 16
    Achievements:
    7 years registered
    ZASto's Avatar
    Join Date
    Jan 2008
    Location
    Messy Corner, 44°47'31.56"N, 20°28'9.16"E
    Posts
    605
    Helped
    121 / 121
    Points
    5,076
    Level
    16

    Re: Problems on Sending RS232 Signals through 3 Meter Core Cable

    Quote Originally Posted by Prabhakarankft View Post
    Hello FvM,
    Kindly find attached image. SP3232 IC pin configuration is slightly differs from MAX3232 Check at 6th PinClick image for larger version. 

Name:	SIPEX.PNG 
Views:	5 
Size:	92.5 KB 
ID:	153949
    If you are using SP3232, the capacitors could be ceramic. They are non polarized.
    Been there, done that. (0805 0.1 uF).
    Make no mistake between my personality and my attitude.
    My personality is who I am.
    My attitude depends on who you are.



--[[ ]]--