SN74LVC8T245: /OE pull-up R does not seperate ports during VCCA power down?

Olaf Albert

Other Parts Discussed in Thread: SN74LVC8T245

Hi supporters,

Our design has VCCA = 3.3V and VCCB = 4.7 V. During power-off the /OE gets pulled-up to VCCA by a 10kOhm resistor. VCCA disappears slow but long before VCCB disappears. Around VCCA = ca. 0,7V we get a strong glitch (L -> H -> exponential to L) at Bn ports that would damage our circuitry if we cannot avoid it.

Is this effect known (due to the datasheet, p. 3, it should not happen with a pull-up R)? Is there any remedy against it?

Thank you for your help!

over 8 years ago

0 Nirav Patel over 8 years ago

TI__Expert 6335 points

Olaf,

There is a similar e2e ticket concerning power ramp and output glitch using SN74LVC8T245 (power up instead of power down):

The remedy in that situation is to have pullup resistors on the inputs. Perhaps it is applicable here as well.

Can you give me the exact configuration of the SN74LVC8T245? I think a schematic would be helpful. From there we can try to re-create a similar setup in our lab.

Best Regards,
Nirav

0 Olaf Albert over 8 years ago in reply to Nirav Patel

Intellectual 320 points

Hello Nirav,

Thank you for your fast reply. Pull-up Rs will not do for us (except for the /OE signal of course) because we need to stay the signals down. I enclosed a pdf with schematic and measurement plots for more details.

Best regards,

Olaf

Latch_glitch_SN74LVC8T245.pdf

0 Olaf Albert over 8 years ago in reply to Olaf Albert

Intellectual 320 points

Hello,

Just wanted to write that we removed the SN74LVC8T245 from our design because we could not solve this problem.

Regards,

Olaf

Logic

Logic forum

SN74LVC8T245: /OE pull-up R does not seperate ports during VCCA power down?