This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Team,
My customer has a question.
We are searching for a way to translate a 125Mbps LVDS differential signal from a POF (plastic optical fiber) transceiver (500-900mV differential output), to a LVPECL PHY input (800mV differential input minimum). These voltage ranges align with the below definitions of LVDS/LVPECL from TI’s app note. Note these are defined as single-ended.
We are interested in TI’s SN65LVDx series of parts. However, I’m confused by the voltage ranges presented in the datasheet. I might be misunderstanding single ended vs. differential, etc. Why would a part claiming to have LVPECL output, give only up to 750mV differential output?
Can you shed some light for us?
Thanks,
Tom
Hi Tom,
The app note mentioned shows a typical voltage swing for LVPECL. SN65LVDT101 is still outputs a valid LVPECL signal just with a lover overall swing. The key is to make sure the LVPECL receiver is compatible with the the output voltage swing from SN65LVDT101.
Hi Malik,
From customer:
"This answer tells me that this part isn’t going to work in our application. Our LVPECL receiver requires a minimum input voltage swing of 800mVpp differential, whereas the SN65LVDT101 specifies it will output 750mVpp max.
Is there any other part you or the team can recommend for this level-shifting LVDS-LVPECL translation?"
We have SN65LVCP23 but this device may not fit nicely in the existing customer design if only one channel is used. This is a two channel device.
Hi Malik,
The customer feels they need a higher voltage swing than what you have offered with this part… I’m afraid the min value for V_OD is still too low. The ENET PHY has a V_IDIFF of 800mVpp min. The signal they need to translate has a 500mV to 900mV Diff Output Voltage (pk-pk). I sent you a separate email with datasheet snippets for the requirements.
Please let me know if there is any TI solution which can solve this problem.
Thanks,
Tom