TXU0104: TXU0104PWR 25Mhz/100Mhz clock translation

Darshan Shah

Prodigy 180 points

Part Number: TXU0104
Other Parts Discussed in Thread: SN74AXC4T245

We are using TXU0104PWR for 3.3V to 1.2V translation.

Waveform attached shows Input (Yellow waveform) and Output (Green Waveform) of the translator.

Output green waveform don't swing from 0V to 1.2V but swing from 0.35V to 1.2V. It is above VIH of the input device.

When we connected 3.3V 100Mhz clock to input of the translator, we got flat 0V line at the output of the translator.

When we connected 3.3V 32Khz clock to the input of the translator, we got perfect 0 to 1.2V 32Khz clock at the output of the translator.

Can someone confirm if we can use this translator for 25Mhz and 100Mhz data/clock signal?

Also, please confirm why we are seeing low swing at 0.35V for 25Mhz signal instead of 0V?

over 2 years ago

0 Clemens Ladisch over 2 years ago

Guru 318700 points

These waveforms cannot be correct: the blue waveform start rising or falling before the yellow waveform. The TXU cannot see into the future.

Please show the complete schematic of your test circuit.

0 Darshan Shah over 2 years ago in reply to Clemens Ladisch

Prodigy 180 points

Yellow waveform (P&N) are swapped. You can see that Ground of the yellow is on the high side. There will be close to 20ns of tPD propagation delay from A to B and hence you are seeing that waveform.

Attaching translator circuit page.

TranslatorCircuit.pdf

0 Michael Ikwuyum over 2 years ago in reply to Darshan Shah

TI__Guru 56767 points

Hi Darshan,

Please note that, your VOL of 0.35V for 3.3V to 1.2V translation is common similar to section 7.5 of the datasheet. If VOL of 0.35V will not work for your application, try a lower IOL or help clarify the type of signal being used for the communication and the VOL requirement to help for a better recommendation, as it is typical for push-pull applications to have VOL>0V.

Also, the TXU device is specified for 100 MHz for 5.5V translation as indicated in the datasheet per section 7.6.

For your 1.2V case, 100 MHz is not feasible and would recommend SN74AXC4T245. I have also gone ahead to verify your 3.3V<->1.2V translation for up to 120MHz using the AXC EVM, thanks.

Best Regards,

Michael.

0 Darshan Shah over 2 years ago in reply to Michael Ikwuyum

Prodigy 180 points

We have UART interface that runs at 3Mbps but we have clock that runs at 100Mhz to be translated from 3.3V to 1.2V.

Do you think, SN74AXCT245 will work fine?

0 Michael Ikwuyum over 2 years ago in reply to Darshan Shah

TI__Guru 56767 points

Hi Darshan,

Yes, the AXC device will work fine for UART interfaces running at 3Mbps and 100MHz 3.3V to 1.2V.

Be mindful of external components such as pull-ups >30 kohms as well as any excessive loading, thanks.

Best Regards,

Michael.

0 Darshan Shah over 2 years ago in reply to Michael Ikwuyum

Prodigy 180 points

We don't need pull-ups as such and have not connected any pull-ups.

Do you have 1.2V Ibis model for AXC? I only see 0.7V and 1.8V models.

0 Michael Ikwuyum over 2 years ago in reply to Darshan Shah

TI__Guru 56767 points

Hi Darshan,

Please see SN74AVC4T774: SN74AXC4T774 ibis model support 1.2Vcc, thanks.

Best Regards,

Michael.

Logic

Logic forum

TXU0104: TXU0104PWR 25Mhz/100Mhz clock translation