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.

SN74AXC8T245: SN74AXC8T245 RGMII interface for 250Mbps, translation in 1.2V <-> 1.8V

Part Number: SN74AXC8T245

Hello,

I have noticed a problem with the SN74AXC8T245 device when implemented for an RGMII interface working at 250 Mbps (each RGMII line). My application implements translation form 1V2 and 1V8 domains:

VCCB = 1.8V, VCCA = 1.2V

I have seen that the signal waveforms observed at port A (1.2V logic) have a logic level low at about 500 mV. However the problem does not exist when the interface works at 25 Mbps - then it can be seen the waveform spans form ~0V to ~1.2V.

It has also been tested in an FPGA bench consisted of an FPGA <-> SN74AXC8T245 <-> PHY where the interface works correctly at 25 Mbps and stops working at 250 Mbps.

Could anyone advice please whether the SN74AXC8T245 device has been characterized and tested for the RGMII interface working at 250 Mbps? Providing the translation happens in case: VCCA, VCCB 1.2V and 1.8V

Thank you,

Kind Regards

Marcin

  • Hey Marcin,

    This device is not characterized for any communication interface due to it being a generic level shifter. The output may not be able to drive 250 Mbps effectively when supplied at 1.2 V. However, this would be the best device for the application. The sweet spot for this device is 1.8 V to 3.3 V (also 2.5 V to 3.3 V). 

  • Hey Dylan, 

    Thank you for your reply. I looked at SCEA065B "2.5.1 Voltage Translation for RGMII" where there is an example of RGMII interface with SN74AXCT8245 being implemented on an application requiring translation form RGMII MAC at 1.05V to Ethernet PHY at 3.3V. Having that I've assumed translation in 1.2V to 1.8V would work as well, similarly to the example mentioned above.

    However, having seen your reply and my experience form a test bench should I assume that the SN74AXC8T245 device is not a good choice for RGMII (250 Mbps) in translation: 1.2V <-> 1.8V?

    Thanks,

    Marcin

  • Hey Marcin,

    Most interfaces have ranging frequencies that can be used so its possible a device can suite an interface for certain cases, but not all cases. These are generic so it would be up to the system designer to assess if the device meets their system parameters. Based on your results, I would say this would not work in this particular case.