• State Resolved
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 35 subscribers
  • Views 91 views
  • Users 0 members are here
Support feedback
Options
Options
Related

GD75232: GD75232 power sequencing

B. Hussain
Prodigy 30 points

Part Number: GD75232
Other Parts Discussed in Thread: GD65232, TRS3243E, TMUX7462F

I note that the datasheet for GD75232 says in section 5.2 Recommended Operating Conditions, Note 1:

When powering up the GD65232 and GD75232, the following sequence should be used:
VSS, VDD, VCC, I/Os
Applying VCC before VDD may allow large currents to flow, causing damage to the device. When powering down the GD65232 and GD75232, the reverse sequence should be used

Could some elaboration be provided on the potential issues caused, and the reason why?

In my design I may have an infrequent usage scenario where the GD75232 is powered by VCC (with +5V) when VDD and VSS supplies are not present. The scenario is when a firmware update is being performed on an embedded microcontroller, where +5V is being supplied to VCC by USB, but the regular PSU is not connected and thus not delivering +12V and -12V to VDD and VSS (that is, they will be floating, not at 0V).

I also do not see any such note or caution in the datasheets of other pin-compatible 75232 devices from other manufacturers, so I am wondering if this issue is specific to TI, or these other manufacturers simply do not mention it.

If such a scenario is definitely not recommended, can any suggestions be made as to how I can gate VCC to the GD75232 in the presence/absence of VDD and VSS?

  • 0
    TI__Mastermind 29602 points

    Hi Hussain,

    What made you select this device?

    In general, we don't recommend this RS-232 transceiver for new designs as we have much newer and improved transceivers with internal charge pump circuitry where only the VCC supply pin is need to supply the device:

    https://www.ti.com/product-category/interface/rs-232-transceivers/products.html#286=1%3B3&287=5%3B5&-1=trs%3Bfalse& 

    If possible, I would try to switch to something like the TRS3243E, which only has the VCC input pin. That way power sequencing is no longer an issue. 

    B. Hussain said:
    I also do not see any such note or caution in the datasheets of other pin-compatible 75232 devices from other manufacturers, so I am wondering if this issue is specific to TI, or these other manufacturers simply do not mention it.

    Unfortunately, due to this particular device's age, we do not have documentation on it anymore, nor the exact reasoning for the power sequencing. Based on the nature of the RS-232 output voltage being higher than the VCC input voltage, my guess is that this large current flow issue is common across all RS-232 transceivers that require the external positive/negative supplies. 

    Best,

    Ethan

  • 0 in reply to Ethan Sempsrott
    Prodigy 30 points

    I chose it because I have +12/-12V supplies available, so I figured why not use it? Avoids the extra component requirements of a single-supply transceiver, and eliminates any EMI/RFI concerns associated with a charge pump. It's also cheaper than the TRS3243E - at least, from my usual supplier. :)

    It's disappointing that there is no documentation that can be referenced - I would have expected TI to maintain resources for all still-active products.

    Would there be any mitigation against this potential excess current flow in the fact that during my given scenario, VDD and VSS supplies will be physically disconnected and thus floating, and the RS-232 connection will also be disconnected?

  • 0 in reply to B. Hussain
    Guru 326810 points

    The internal schematics are shown in figures 7-3 and 7-4. They do no show a connection between VCC and VDD/VSS, but I suspect that there are diodes between them (probably formed from P/N junctions between the components and the substrate).

    You could use a device like the TMUX7462F to gate the VCC and input signals of the GDxxx (the switches open when its supplies are floating).

  • 0 in reply to Clemens Ladisch
    TI__Mastermind 29602 points

    Hi Hussain,

    Because this device is before my time, it is also possible we never had documentation in the first place. 

    But Clemens' comments are accurate. 

    Best,

    Ethan

  • 0 in reply to Ethan Sempsrott
    Prodigy 30 points

    I think rather than complicate things by trying to comply with the power sequencing requirements by adding extra circuitry and components, I decided to re-evaluate and solve the problem another way, and make the GD75232 totally unpowered during my scenario.

    I've implemented an arrangement where, for firmware update during which the USB is acting as a power input (i.e. device mode, as opposed to usually as a power output, i.e. host mode), the user has to reposition a jumper that redirects USB +5V to only the microcontroller. In normal use the regular +5V supply will come to the microcontroller via a diode, and in update mode the diode will prevent the redirected +5V powering everything else.

    So, the GD75232 will be totally unpowered, and nor will I need to worry about any parasitic supply through I/O lines interfaced with the GD75232, as nothing else directly interfacing will be powered either.