SN74LVC8T245: how about its output status before its input side is powered on?

John11513

Part Number: SN74LVC8T245

Hi team,

My customer is using competitor's 74LVC4245A. They found a technical issue with it.

The part is configured to transfer 3.3V logic from side B (VCCB=3.3V) to side A 5V logic (VCCA=5V). The 3.3V rail is stepped down from 5V. Such the 3.3V is powered up 15ms latter than 5V. The /OE input is pulled up to 5V before 3.3V powering up. Through this design they expecting the side A outputs could be disabled (hi-z) until 3.3V rail settled. While the test result cannot meet this expectation: they found the side A output is actively pulled down until side B (3.3V) is powered up. Bellow is the power up waveform:

I am wondered if our SN74LVC8T245 has similar issue? Can it be used to solve this problem?

Thank you,

John

over 6 years ago

0 Emrys Maier over 6 years ago

TI__Guru* 96591 points

Hi John,
I have asked our translation expert to take a look at this and comment.

0 ShreyasRao over 6 years ago

TI__Guru 56070 points

Hi John,

To summarize ,
Vccb is powered on , Vcca is turned on 15ms later going through the LDO. OE is tied to Vccb? A_out has external pullup ?

can you provide a schematic with this ?

0 John11513 over 6 years ago in reply to ShreyasRao

TI__Expert 5360 points

Hi Shreyas,

Thank you for reply.
The condition is: DIR=0, Vcca (5V) is powered on first, Vccb (3.3V) is on 15ms later. A_ out has external pullup. OE is controled by POWER GOOD signal of 3.3V DC/DC circuit which is pull up to 5V until 3.3V settled. The design is intended to keep A_ out high before Vccb settled. Whereas the test result shows the OE pin cannot control A_ out before Vccb is turned on. The test was done with NXP's 74lvc4245, I wondered how about our device?

Regards,
John

0 ShreyasRao over 6 years ago in reply to John11513

TI__Guru 56070 points

John ,

www.ti.com/.../scea021a.pdf
The app note talks about the power up considerations for the device.
I would suggest considering the sn74LVC8t245 instead which works for across the power rails from 1.65 to 5.5V and is p2p with the lvcc4245a .

www.ti.com/.../avclvcdircntrl-evm
consider using the EVM for evaluating the devices which supports both devices .

0 John11513 over 6 years ago in reply to ShreyasRao

TI__Expert 5360 points

Thank you for information.

But the app note does not answer my question. The power sequence of the ckt is completely correct.

Regards,

John

0 ShreyasRao over 6 years ago in reply to John11513

TI__Guru 56070 points

John ,

If you are following the power sequence as per the app note, i do not anticipate problems.
i cannot comment on the NXP device performance.

0 John11513 over 6 years ago in reply to ShreyasRao

TI__Expert 5360 points

Hi Shreyas,

Thank you, we have tested SN74LVC8T245. Yes, it does solve the problem. Our part do not have the same problem. This update my knowledge about logic device: the devices with same logic functionality in fact do not behavior completely same way. Below is the power up waveform with LVC8T245:

Regards,

John

0 ShreyasRao over 6 years ago in reply to John11513

TI__Guru 56070 points

Appreciate your inputs and providing the feedback on the device.
Although the parametric are very comparable , the internal architecture will be different and hence the performance behavior especially on the power up and power down.

Logic

Logic forum

SN74LVC8T245: how about its output status before its input side is powered on?