TXB0108PWR & TXS0108EPWR Oscillation

Michael Blevins

Other Parts Discussed in Thread: TXS0108E, AM26LV32

I am using TI's TXS0108EPWR to translate 5V encoder signals to 3.3V CMOS inputs. The encoder signals are from a HEDS 9100 encoder. 4.7K pull-ups are used on the 5V (B) side of the IC. The 3.3V (A) side is connected to a PMD I/O chip. The Enable line is pulled up to 3.3V through a 4.7K resistor. Whenever the encoder signals are high there is a 23.5 MHz rough square wave oscillating and causing extra encoder signals to be read by the PMD chip. The signal goes away when the TI chip is replaced by step down resistors. A buffer is still desired between the external encoder connector and PMD chip. I have tried the TXB0108PWR but this did not work either. A drop in buffer/receiver is still desired to salvage the board layout if available. I have also tried different size pull-ups/pull-downs and no pull-ups.

over 11 years ago

0 Wei Lang over 11 years ago

TI__Expert 6430 points

For TXS0108E, there is 10k internal pull-up resistor on each I/O. Could you remove external pull-up resistor and test again?

Thanks

0 Michael Blevins over 11 years ago in reply to Wei Lang

Prodigy 55 points

Hello,

I have tried removing the pull-ups and also different size pull-ups with no improvement. I have removed all oscillators from the boards as well. The oscillations seem to be coming from the TI chip because when this is removed the oscillations go away. Any other ideas or drop in replacements?

Thanks

0 Michael Blevins over 11 years ago in reply to Wei Lang

Prodigy 55 points

PMD recommends an AM26LV32 line receiver. This is not a drop in, but has worked on other designs for them, so we are planning to use this.

Thanks,

0 Wei Lang over 11 years ago in reply to Michael Blevins

TI__Expert 6430 points

Thanks for your feedback.

For TXB/TXS serial, if output load is greater, the devices could be abnormal. there is another requirement for TXB/TXS serial, input transition edge should be less 30ns/V (TXB) and 10ns/V(TXS).

Thanks

Wei

0 Carl Kosmerl over 11 years ago in reply to Wei Lang

Prodigy 30 points

Hello I am having the exact same problems that Winston has stated about a month ago. I was wondering if either Winston or TI Support knows of the resolution to this?

Thanks

Carl

0 Michael Blevins over 11 years ago in reply to Carl Kosmerl

Prodigy 55 points

Hello Carl,

I ended up using a network of voltage dividers to transition from 5V to 3.3V. I was trying to transition a 20 kHz signal that had way to slow of a rise time for this device. The AM26LV32 also works for differential inputs... receives 5V level inputs and is powered by 3.3V for the outputs.

0 Wei Lang over 11 years ago in reply to Michael Blevins

TI__Expert 6430 points

Hi Carl,

Could you give more detailed information. Such as input network and output nework.

Thanks

Logic

Logic forum

TXB0108PWR & TXS0108EPWR Oscillation