• State
  • +1 person also asked this people also asked this
  • Replies 6 replies
  • Subscribers 36 subscribers
  • Views 97 views
  • Users 0 members are here
Support feedback
Options
Options
Related

PCF8574: PCF8574 very high P port transient pullup current

Mathias
Prodigy 10 points

Part Number: PCF8574

Hello,

Vcc = 5.1V
I2C freq = 5kHz

We use PCF8574DWR's P ports as inputs, 3 of them are pulled low while we first write the ports high, as per datasheet: "the I/Os should be high before being used as inputs". 

During the acknowledgment of WRITE HIGH to all ports, we have historically measured approximately 4mA drawn per P port pulled LOW (P port transient pullup current). 

Multiple samples of PCF8574DWR with batch marking "5BA2VKT" tested recently draw more than 60mA per P port pulled low. This overloads our power supply which causes Vcc to drop and cause functional issues. Samples with batch marking "58AE4ET" exhibit the historical behaviour. 

The datasheet only specifies a typical P port transient pullup current of -1mA at Vcc 2.5V. 

Is the observed increase in P port transient current a consequence of PCN20240930002.1 or is it more likely that we are dealing with a defective/damaged batch of components? We need to figure out if our designs need to evolve following PCN20240930002.1. 

Thank you

  • 0
    TI__Guru 59431 points

    Hi Mathias,

    We are still looking into this and will follow up shortly.

    Regards,

    Jack

  • 0
    TI__Guru 62928 points

    Hi Mathias, 

    Do you have a schematic to show how the device is being used? 

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    Prodigy 10 points

    Hi Tyler,

    thank you for your reply, sorry I cannot share a schematic. I can say that we have been using this device on the design in quesiton for decades with no issues. 

    The P ports all have a 47k pullup to Vcc and may be pulled down by open-collector devices or N-MOS open drains, those devices are on separate power supplies from PCF8574. The transient current, which lasts 1 SCL high duration during the PCF8574 ACK, is proportional to how many P ports are pulled low. The 5.1V PCF8574 power supply is current limited to 10mA, bypass capacitors provide the rest of the transient current. We used to have a worst-case 0.5V voltage drop across the bypass capacitors caused by the transient current. Now with the 15x increase in current, more power supply current or capacitance is required to keep PCF8574 powered during the transient. But we are not planing on accomodating the devices with this behaviour, concerned that the abs max current ratings are being exceeded.  

    Regards,

    Mathias

  • 0 in reply to Mathias
    TI__Guru 62928 points

    Hi Mathias,

    This looks to be related to the PCN. The device underwent a process change, so I think this probably had an effect on the transient current seen. 

    If is possible to adapt your system to the PCN device? 

    Regards,

    Tyler

  • 0 in reply to Tyler Townsend
    Prodigy 10 points

    Hi Tyler,

    thank you for your response. I am concerned that this exceeds the part's absolute maximum ratings: "Continuous output high current max -4mA" and "continuous current through Vcc +100mA", if only for 100µs (half clock cycle). Do you concur? 

    If it is now intended for us to limit the transient current externally, then no we can't adapt our systems ot the PCN device. We have several other use cases that will require too many resources to update.  

    Would you please be able to confirm the typical transient current of PCN devices? the datasheet still states -1mA at Vcc=2.5V, but it seems it would be 15 to 30 times higher now. 

    Regards,

    Mathias

  • 0 in reply to Mathias
    TI__Guru 62928 points

    Hi Mathias,

    How is the transient current being measured at the ACK bit? Do you have a scope capture that shows the current waveform? This information may be useful to our design team to help analyze the situation. 

    Regards,

    Tyler