This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

AM26LV32: AM26LV32I - RS-485

Part Number: AM26LV32

Hello,

I'm looking for a Quad RS-485 receiver with low tHL/tLH<~20ns.

I found the AM26LV32I. There is a conflict between the parameters mentioned in different places in the site.

1. Is this component available? (with the "I" at the end).

1. Is this component fully compatible for RS-485 communication standard, and what is the termination required in multi drop usage (from one transmitter to X receivers)?

2. What is the allowed number of receivers to be connected on a single lane? (one place says 32, other 10, not mentioned in the datasheet).

3. Is the SN65LBC173A preferred? (although it seems to be older, and it has SOIC package only).

Thanks!

 

  • Hi Max,

    Yes, this component is available. I am reviewing your other technical questions and plan to get back to you tomorrow (Monday).

    Regards,
    Max
  • Max,

    I have double-checked the specification, and it looks like AM26LV32I can meet most requirements of RS-485 but may be limited in the common-mode voltage range supported. The RS-485 standard specifies operation across a range from -7 V to +12 V, but the AM26LV32I limits the bus voltages to -4 V (minimum) / +8 V (maximum).

    The maximum number of receivers is limited by signal integrity concerns (primarily cable-dependent and application specific) as well as the loading each receiver places on the bus (device-dependent). RS-485 drivers are typically specified based on an assumed loading of 54 Ohms, equivalent to two termination resistances and the parallel combination of up to 32 "unit loads" (representing receivers). You can read more about the RS-485 unit load here: e2e.ti.com/.../rs-485-basics-how-to-calculate-unit-loads-and-the-maximum-number-of-nodes-on-your-network. The AM26LV32I device has a minimum input resistance of 7 kOhms, so it could be considered equivalent to ~1.7 unit loads. This means that 18 of them could be supported by a standard RS-485 driver. Of course, with stronger drivers (i.e., higher output voltage specs or equivalent output voltages for higher loading) more nodes could be supported. More nodes could also be supported with single termination rather than double termination. In this case, the total number of parallel receivers could be as high as 71 without exceeding the standard requirements.

    Choice of termination scheme would be determined at the overall system level and would not be limited by the receiver used. A general rule of thumb is that double termination is beneficial for multi-point buses (i.e., each node can send or receive) since signaling is bidirectional and can originate from any node. For multi-drop/"broadcast" applications, then single termination (placed at the furthest receiver) is generally OK. (This is the standard system configuration for RS-422. For more information, you may find this app note helpful: www.ti.com/.../slla070d.pdf.)

    The 'LBC173A device isn't necessarily preferred over AM26LV32I. Assuming availability of both devices, you could go with whichever one's specifications suit your need.

    Best regards,
    Max
  • Hi Max,

    Thanks for the detailed answer.

    1. I'm not sure I understood the calculation right.

    If I'm using a single 120ohm termination, I get:

    (7k/N)||(120/2)=54 ==> N=142

    [I'm using 100ohm termination due to 100ohm Zo (transmission lines) ==> N=119]

    2. Anyway it's more than enough,  I need only 12-16 receivers on the bus.

    Other issue, if I'm not confusing, is that the -4 V to +8 V you have mentioned, is actually −0.3-V to 5.5-V (Common-Mode Range) - by the datasheet that I'm looking at (www.ti.com/.../am26lv32.pdf). Is this good enough to treat this device as RS-485 compatible?

    Thanks!

    Max

  • Max,

    Good point on the common mode range - it is best to follow those limits given in the Recommended Operating Conditions section. The range I gave before is the Absolute Maximum range, over which the device will operate (without becoming damaged) but may not meet all specifications. I would say that this limited range means that the solution is not fully RS-485 compliant, but would be compatible with RS-485 as long as the common-mode range stays within these bounds. (Practically, this limits the amount of ground shift that could be tolerated between different nodes on the network.)

    I double-checked the calculations. It looks like if N = 142 then (7k/N)||(120/2) yields 27 Ohms rather than 54 Ohms. Please let me know if I am missing something obvious here (it's surely possible). Anyway, if you are using a single termination resistance then there shouldn't be a need to divide it by two in the calculation. So, in your case you could use the equation (7k/N) || (100) = 54 and solve for N.

    Best regards,
    Max