FPGARelated.com
Forums

Interfacing Cyclone III to 3.3v LVDS devices

Started by liqi...@gmail.com November 28, 2007
How can i interface the cyclone III with 2.5v LVDS to 3.3v LVDS adc &
dac ?

Thanks
On Nov 29, 6:46 am, "liqi...@gmail.com" <liqi...@gmail.com> wrote:
> How can i interface the cyclone III with 2.5v LVDS to 3.3v LVDS adc & > dac ? > > Thanks
IMHO directly. I've connected 3.3V Stratix LVDS to 2.5V StratixII LVDS in both directions. "The Cyclone III device meets the ANSI/TIA/EIA-644 standard with the following exceptions: =94 The maximum voltage output differential (VOD) is increased to 600 mV. The maximum VOD for ANSI specification is 450 mV. =94 The input voltage range can be reduced to the range of 1.0 V to 1.6 V, 0.5 V to 1.85 V or 0 V to 1.8 V based on different frequency ranges. The ANSI/TIA/EIA-644 specification supports an input voltage range of 0V to 2.4V." See the Cyclone III Device Datasheet: DC and Switching Characteristics of Cyclone III Devices chapter in volume 2 of the Cyclone III Device Handbook for the LVDS I/O standard electrical specifications and check your DAC/ ADC specification for final proof. Digitally yours, Michael Tsvetkov http://www.jpegls.com
liqiyue@gmail.com wrote:
> How can i interface the cyclone III with 2.5v LVDS to 3.3v LVDS adc & > dac ? > > Thanks
If you set 3.3v to a i/o block and Quartus does not allow you to assign LVDS on a pin of that block (w/ an error message when fitting) tell quartus you have 2.5v (despite you will use 3.3v in your board) It will pass compilation and will work great. I had that issue on CycloneII driving clocks to ADCs from Analog devices, I'll bet will be the same on CyIII. lc.
>> How can i interface the cyclone III with 2.5v LVDS to 3.3v LVDS adc & >> dac ? > >If you set 3.3v to a i/o block and Quartus does not allow you to >assign LVDS on a pin of that block (w/ an error message when fitting) >tell quartus you have 2.5v (despite you will use 3.3v in your board) >It will pass compilation and will work great. >I had that issue on CycloneII driving clocks to ADCs from Analog >devices, I'll bet will be the same on CyIII.
For LVDS transmission from 3.3V I/O, if all three matching resistors (at the transmitting end) are 150 ohm then this gives output levels to the LVDS specification and presents 100 ohm to the line. The only disadvantage (compared to 2.5V operation) is an increase in power dissipation.