TXS0108E: Oscillation on input side of level shifter (side B)

Prodigy 20 points

Replies: 3

Views: 52

Part Number: TXS0108E

The 5 volt side B inputs are connected to Atmel ATF750C and some lines are connected to HCPL2400 with totem pole output.  Signals on these lines oscillate at a high rate (~5 nSec pulses 10 to 20 nSec apart) in bursts of several hundred uSec.  There are no pull ups on these lines and the trace lengths between parts are less than 10 inches.  The normal data rates are in the order of few MHZ.  If I remove the level shifters, every thing works normal.  There are some lines that seem to behave that are connected to data bus that has pull ups.  Any suggestion as to what the cause may be or how to fix the issue.

Thanks

3 Replies

  • Hey Rajen,

    Do you have any scope shots of this issue? This is usually due to heavy capacitive loading causing the transition time to be slow which would make the device oscillate from the one shots continuously firing in each direction.

    Check out our new AXC family of Voltage Translators! 

    Watch the Introduction to AXC Family to find out more about the family.

  • Hi Dylan

    I do not have scope shots handy.  If i describe them, they are 3 to nSec pulses going from hi to low for about 15 to 20 nSec.  They come in bursts of 3 to 5 hundred uSec at a time randomly.  Looking at the pulses, the 3.3 v side leads the transition by few nSec as what is seen on the 5 V side.  So, the oscillations seem to start from the 3.3 V side.  We are using the MAX10 from Altera and perhaps the pin attribute is incorrectly set on the FPGA.  We selected weak pull up for the pin.  What should it be set for so it matches the side "A" input charctercstics of the level shifter.  The FPGA is 10m04scu169c8g from Altera and connected directly to the level shifter side "A" within 1 to 2 inches.

    Thanks,

    Rajen

  • In reply to Rajen Sud:

    Hey Rajen,

    I will be honest, helping debug with no schematic or scope shots will be quite difficult and I will only be able to give generic responses for the leading causes for this type of behavior. This device has internal pull-ups so external pull-ups will not be necessary. The leading component to this would be heavy capacitive loading leading to slow edge transition rates. Are these oscillations happening in a steady state with the device just idle or are they occurring during a transition?

    Check out our new AXC family of Voltage Translators! 

    Watch the Introduction to AXC Family to find out more about the family.