• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 234 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Original question:

TRS3232E-Q1: TRS3232EQPWRQ1 Part Unused channel pin connect in schematic.

TRS3232E-Q1: TRS3232WQPWRQ1

Scott Morrill
Prodigy 70 points
Part Number: TRS3232E-Q1
Other Parts Discussed in Thread: TRS3232,

Tool/software:

I have many boards with TRS3232EQPWR component already installed with DIN2 pin 10 floating, I think I miss-read your data sheet suggesting unused inputs are tied down, but it appears its only the RIN pins.  I noticed in the blogs the unused DIN pin must be tied down, why? Is it just for EMI, the other TRS3232 channel won't fail eventually will it, in other words what the hazard in leaving a DIN pin floating?

BTW, if it is a must have, then why isn't it stated in the data sheet not to leave the DIN1or2 pin floating?

BR,

Scott

  • 0
    TI__Guru 73971 points

    Hi Scott,

    Scott Morrill said:
    I noticed in the blogs the unused DIN pin must be tied down, why? Is it just for EMI, the other TRS3232 channel won't fail eventually will it, in other words what the hazard in leaving a DIN pin floating?

    No, the device itself shouldn't fail or become damaged if the DIN pin is left floating. Its not really an EMI issue either with RS232 devices since the slew rate of the Dout pin is controlled to be less than 30V/uS (part of the standard). The main reason for this is it will consume more current if the Din pin floats to a value between Vcc and GND (worst case is Vcc/2). What I mean by this is the internal CMOS input will be half biased where the PFET will turn on slightly and the NFET will turn on slightly resulting in more leakage current. In the past, when I studied this for our I2C devices I saw around 30uA per floating input. 

    Scott Morrill said:
    BTW, if it is a must have, then why isn't it stated in the data sheet not to leave the DIN1or2 pin floating?

    This isn't really a must have in end equipments where the power consumption isn't important because it's powered through a plug. End equipments that need lower power consumption are battery powered ones where they probably shouldn't float any inputs at all. This situation isn't just for RS232 TTL inputs but for all TTL high-Z inputs. 

    -Bobby

  • +1
    Guru 317370 points

    When you allow in input to float, you will get large shoot-through currents (which can be many mA, depending on the device), and the device might fail or get damaged. See [FAQ] How does a slow or floating input affect a CMOS device?

  • +1 in reply to Clemens Ladisch
    Prodigy 70 points

    Hello, I'm confused you Bobby and you Clemens don't seem to be on the same page with answers.  BTW of scoped a few boards at this point and the input DIN2 seems to be consistently at ground, almost like it is being pulled low, but the input impedance is measuring open with my DMM proving that to be not the case.  Is there something in the structure of the device that would help pull the floating input low?  To be sure a little extra supply current to the device do to the floating input does not bother me.  What I'm trying to decide is if I need to rework the boards to tie the DIN2 input to DIN1 with a solder blob between the pins to avoid any damage to the device?

  • +1 in reply to Scott Morrill
    Guru 317370 points

    The input is symmetrical and will float at random voltages. The input capacitance is extremely low, so the input impedance of your DMM will pull the input to the reference voltage; when you measure the voltage between the input and VCC, it should measure high.

    Here is an example where a floating unused channel made an inverter work like a non-inverting buffer. The TRS3232E-Q1 might or might not be more robust.

  • +1 in reply to Scott Morrill
    TI__Guru 73971 points

    Hi Scott,

    Sometimes adding a scope/DMM can add leakage current into the node your looking at (usually very low leakage though). The device itself may have some small level of leakage that could pull it to GND though it can vary between devices as well. You may also see the voltage fluctuate in higher noise environments. 

    I personally don't think one floating input here will cause a complete failure since the shoot through current will be small and won't result in much heat build up. Maybe if there were a bunch of floating inputs here, the additional leakage could cause noticeable heat build up but with one pin, I think it would be very minimal. 

    I think if you have to respin the board for other reasons, you can include biasing the input to a known value but I don't think it's necessary to do a complete spin just to bias a floating pin if you aren't concerned about the leakage it's generating. 

    -Bobby