• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 21 subscribers
  • Views 332 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.

SN74LV8T245: Issue with clamp diodes

Alex Bertin
Prodigy 10 points
Part Number: SN74LV8T245
Other Parts Discussed in Thread: SN74AHC541, SN74LV541A

Tool/software:

Hello. I have a question regarding the SN74LV8T245 octal bus transceiver. We are using this device to translate from +5V (B-side) to +3.3V (A-side) I/O. The circuit diagram is below:

This device is actually on an adaptor that sits on a host board. On the host board the D0~D7 lines are pulled up to +5V with 10K pull-up resistors. The setup is shown below (one channel):

The problem we are finding with this setup is that the VCCO (+3.3V) voltage rail is being pulled up to +4.2V. It appears that the combination of the pull-up resistor and the upper clamp diode of the A-side output buffer provide a path from the +5V rail to the +3.3V (VCCO) rail.

Please could you comment on this. 

The SN74LV8T245 voltage translator VT1B is correctly configured for translation from +5V to +3.3V. Pulling up a databus line to +5V with a 10Kohm resistor is typical practice.

I notice that TI have other voltage transceivers. If I selected one without clamp diodes i.e. the SN74AHC245RKSR or the SN74LV245ARKSR, would this solve the problem?

Thanks.

  • 0
    Guru 317370 points

    This device does not have clamp diodes at the inputs. What is the marking? Is the DIR input really at GND? Can you confirm that a current is flowing through the pull-up resistors?

  • 0 in reply to Clemens Ladisch
    Prodigy 10 points

    Hi Clemens, Thanks for your reply. You are right the SN74LV8T245 only has a pull down clamp diode at the input and pull-up and pull-down clamp diodes at the outputs:

    The marking on the device is:

    In my diagram above I have added them for the A buffer. The problem comes in that the A buffer is un-enabled (as DIR is GND (I need to check this) and I am translating from B to A) but the clamp diodes are still present. This gives the problem I have outlined above.

    As another experiment I have isolated the adaptor from the host board and powered it from +5V. I have then pulled D0 up to +5V with a 10K resistor and the +3.3V supply on the adaptor rises to +4.2V.

    Certainly I could add in a ammeter to see whether a current was actually flowing through the 10K pull-up resistor.

  • 0 in reply to Alex Bertin
    Guru 317370 points

    What is the voltage drop over the pull-up? Can you detect these apparent diodes with the diode test mode of the multimeter?

    You do not actually need a transceiver. I'd guess you would not have this problem with a plain buffer, e.g., SN74AHC541.

  • 0 in reply to Clemens Ladisch
    Prodigy 10 points

    Hi Clemens, Ok. I will need to do some measurements. You are right: in this instance I don't actually need a transceiver. The SN74AHC541 is a good suggestion, only problem is we are space constrained so it's package offerings are far too big. Thanks.

  • 0 in reply to Alex Bertin
    Guru 317370 points

    The SN74LV541A is available in QFN packages.

  • 0 in reply to Clemens Ladisch
    Prodigy 10 points

    Yes, this device is available in the RKS package, which we could use. Thanks.

  • 0 in reply to Alex Bertin
    Prodigy 10 points

    Hi Clemens, Ok, so I have measured the current flow through the 10K resistor, it is 13uA. So the voltage drop across the resistor is 0.13V. I have also measured the DIR input pin on the SN74LV8T245 and it is GND.

    Please let me know your thoughts. Thanks.

  • 0 in reply to Alex Bertin
    Guru 317370 points

    So the voltage drop over the diode is 0.67 V, which is plausible. And I guess that 13 µA also is plausible for all devices connected to VCCB. As you've suspected, this looks exactly like a diode.

    I'd guess that this is the output diode, and that it does not get disabled even when the output is disabled by DIR. Figure 8-1 might be technically correct, but is misleading for a transceiver where all pins are both inputs and outputs.

    As far as I know, a device with the Ioff feature would disable the output diodes.

  • 0 in reply to Clemens Ladisch
    Prodigy 10 points

    Ok. So to solve my issue with the SN74LV8T245 I need to change it for either the SN74AHC245RKSR or the SN74LV245ARKSR.

    I notice in the SN74LV245ARKSR datasheet (p6: Section 6.6 Electrical characteristics and section 8.1 (page 10)) Ioff is specified. I guess this would be the replacement part to go for?

  • +1 in reply to Alex Bertin
    Guru 317370 points

    Yes. That Ioff value is specified for VCC = 0 V, but it shows that there cannot be a diode.