need help with ADM2484E

Have received a problem at use ADM2484E - the starting bit turns out with  "step". The circuit drawing and signal photo in the appendix. In the photo:

- blue signal (3) - transmission enable - pin 5 DE ADM2484E

- green (2) - TxD signal on 6 pin of ADM2484E

- yellow (1) - RS485 line signal (DATA+ pin 14 ADM2485E)

Thanks!

attachments.zip
  • 0
    •  Analog Employees 
    on Nov 3, 2011 6:08 PM over 9 years ago

    Hello Yurik,

    Thank you for the schematic and oscilloscope plot.

    Could I ask what data+/data- were connected to for the test condition? Were there any termination resistors, pull-up/pull-down on the bus, other nodes or cable?

    Also, does the half/full signal into the circuit switch between ADM2587E/ADM2484E at the time of the test, or will it have been stable (ADM2484E active) for some time before this?

    Regards,

    Conal

  • Hello Conal,

    the signal HALF/FULL are stable during test, the  pull-up/down resistors

    does  not present. Terminations resistors (120 om) are installed in both

    ends of net. 

    Test RS-485 network drawing attached.

     

    In the point-to-point network  when we use terminal resistor connected to

    TAU-G DATA+ signal are correct.

     

    Regards,

    Yuri Kolbasin

  • 0
    •  Analog Employees 
    on Nov 7, 2011 5:57 PM over 9 years ago

    Hello Yuri,

    If I understand correctly then, the behaviour in the oscilloscope plot you provided is not present if the circuit is connected only to the bus without the other two nodes? This would suggest the ADM2484E is operating correctly.

    Is it possible to check if this occurs with only the TAU circuit and one of the other nodes connected (and termination resistors) and not the other node. And vice-versa as well?

    One of the nodes does have an "automatic flow control" feature, perhaps the unexpected "high" is due to an error in this? (unexpected bus contention)

    Regards,

    Conal

  • Hello Conal,

    i don't agree with your conclusion. Now we hold device TAU-G testing - the

    first stage was a point-to-point network. We connected TAU-G directly to

    ADAM server, between contacts DATE + and DATE  have connected Rt=120 ohm.

    TAU-G has fulfilled correctly.

    The second stage we have connected TAU-G  to a working network from two

    nodes - ADAM  server and ICPCON device. This network before connection TAU-G

    worked correctly.

    In my opinion a problem that the isolated power supply ADM2587E at an exit

    from a 'sleeping mode' can't develop sufficient capacity at a first time.

    Probably, it is necessary to load an isolated power supply  in advance,

    prior to the beginning of transfer of starting bit.

    In a case unexpected bus contention we would have all bits of transferred

    sequence amazed (drawn to DATE +).

     

    Regards,

     

    Yuri Kolbasin

  • 0
    •  Analog Employees 
    on Nov 14, 2011 10:03 PM over 9 years ago

    Hello Yuri,

    Apologies for the delay in getting back to you.

    The ADM2587E does not have a "sleep mode". With the ADM2587E driver and receiver inactive, the digital isolation (signal and power) components of the device remain active.

    Your earlier oscilloscope plot only has DATE+ captured, can you confirm that DATE- is not also high during the "error" interval? (Ideally could the same plot be captured with DATE-?)

    I mentioned any potential auto-transmit/auto-direction feature of the other nodes previously, as such a feature may not have been designed for compatibility with a 3.3V powered transceiver. The ADM2484E driver output levels are lower than for a 5 V powered transceiver (e.g. ADM2491E) and correspond to the 3.3V operation on Vdd2.

    Regards,

    Conal