0
\$\begingroup\$

I have a device (VSC8540) driving another device (LS1046). The driver specifies output leakage current as +/- 85uA. The receiver specifies input current as +/-50uA.

First of all, is the input current specified for the receiver (LS1046) the same as leakage current? And also, I am trying to evaluate this to see if this is a problem. I see the driver can source and sink 1mA if needed so its more than enough for the input receiver.

Can anyone help me understand whether the leakage output current and input current are compatible (ie there is no problem with this part of the interface?)

Thanks! Transmitter Device: https://ww1.microchip.com/downloads/en/DeviceDoc/VMDS-10515.pdf

Receiver Device: https://www.nxp.com/docs/en/data-sheet/LS1046A.pdf

\$\endgroup\$

1 Answer 1

0
\$\begingroup\$

These devices are maybe DC isolated by a hybrid-transformer so DC leakage is not transferred to each other. But in the direct connect "reduced ... interface mode" implies the leakage from output to 1V on the interfaces for various supply voltages and not the modulated differential current.

Also, the conditions for Tx leakage are defined as ;

"The following table shows the DC specifications for the pins referenced to VDDMAC, VDDIO, and VDDMDIO when it is set to 2.5 V. The specifications listed in the following table are valid only when VDD1 = 1.0 V, VDD1A = 1.0 V, and VDD25A = 2.5 V."

The Tx also has an internal R to limit the voltage with this DC leakage. This appears to be the current near the midswing of the output. from <= 0.4 to >=2.0

\$\endgroup\$
3
  • \$\begingroup\$ Stewart - How do you know that there is a hybrid transformer that is isolating these two devices? \$\endgroup\$
    – Matty
    Apr 26 at 20:44
  • \$\begingroup\$ I assume you want maximum CMRR and be compliant to IEEE 802.3-2008 and maybe IEEE 802.3az-2010 but if you don't then I would expect some shift in Vcm (common mode) \$\endgroup\$ Apr 26 at 20:53
  • \$\begingroup\$ Sorry I made a bad assumption. the Reduced gigabit media-independent interface (RGMII) has sufficiently low DC leakage to stay within a linear common mode range. This leakage current implies a high common mode impedance rather than the differential current control impedance. But clocking arrangements and baseband patterns are changed to accommodate this, I believe. \$\endgroup\$ Apr 26 at 21:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.