hi,all. I am realize pci target core in fpga(cyclone),but there are some questions. 1,I don't know what type pin should pci_clk be,normal io or clk of fpga? 2,I series 50ohm resistors between pci connector and fpga,does this bring some problems? the goal is to make it suitable for 3.3v and 5v. thanks.
realize pci in fpga
Started by ●February 6, 2006
Reply by ●February 6, 20062006-02-06
eehinjor wrote:> hi,all. > I am realize pci target core in fpga(cyclone),but there are some > questions. > 1,I don't know what type pin should pci_clk be,normal io or clk of > fpga? > 2,I series 50ohm resistors between pci connector and fpga,does this > bring some problems? the goal is to make it suitable for 3.3v and 5v. > thanks. >Cyclone is not compatible with PCI 5v, the voltage transients exceed 5v and will damage the device. You will need to use IDT Quickswitch or similar device to limit the input voltage to your Cyclone in a 5v scenario. Google for PCI and quickswitch, you should find a Xilinx app's note on the subject. Alan
Reply by ●February 6, 20062006-02-06
Alan Myler wrote:> > Cyclone is not compatible with PCI 5v, the voltage transients > exceed 5v and will damage the device. > > You will need to use IDT Quickswitch or similar device to > limit the input voltage to your Cyclone in a 5v scenario. > > Google for PCI and quickswitch, you should find a Xilinx > app's note on the subject. >... or look a the schematics for the Altera Cyclone PCI development kits. -hpa
Reply by ●February 7, 20062006-02-07
H. Peter Anvin wrote:> ... or look a the schematics for the Altera Cyclone PCI development kits. > > -hpaFinding is better than looking.... ftp://ftp.altera.com/outgoing/devkit/PCI_DK_2C35/schem_cii_pcidevbd_b.pdf Karl.
Reply by ●February 8, 20062006-02-08
Thanks all.>From the datasheet of cyclone,it is compatible with pci-5v by diode andrisistors. Can somebody help me to solve the first question?
Reply by ●February 9, 20062006-02-09
it should trace through to the clock pin this allows better us in the FPGA fabric, we have applied this to our development boards so that the PCI clock can be used as the system clock if the user so wishes REGARDS IAN Enterpoint Ltd: Home of the Low-Cost development board RAGGEDSTONE1 Website: www.enterpoint.co.uk "eehinjor" <eehinjor@163.com> wrote in message news:1139408642.434006.277270@g43g2000cwa.googlegroups.com...> Thanks all. >>From the datasheet of cyclone,it is compatible with pci-5v by diode and > risistors. > > Can somebody help me to solve the first question? >
Reply by ●February 9, 20062006-02-09
Thank you.Ian Muncaster. By the way, In the spec:the length of CLK is 2.5inches +-0.1inch. if there is a qs3861 or resistor between connector and fpga,how can I calculator the length of PCB-trace?
Reply by ●February 9, 20062006-02-09
RIGHT this is some calculation: Wouldn't advise using a resistor? For the QS381 the maximum propagation is 250ps, using a signal propagation of 1ns equals 12inches for lightly loaded line or 3inches for a loaded line (using bus switch most likely means lightly loaded) we can work out the equivalent trace length the bus switch would represent. Doing this you would find that this fails the PCI spec, we have seen no problems if you just make sure your clock trace is longer than your worst case for the rest of the bus. This means that the setup and hold times are met. "eehinjor" <eehinjor@163.com> wrote in message news:1139485190.614959.181580@o13g2000cwo.googlegroups.com...> Thank you.Ian Muncaster. > By the way, > In the spec:the length of CLK is 2.5inches +-0.1inch. > if there is a qs3861 or resistor between connector and fpga,how can I > calculator the length of PCB-trace? >
Reply by ●February 9, 20062006-02-09
Thank you for your response so soon. 250ps is about 3inches,but the length of clk can not be longer than 2.5+0.1inches. how can that meet the spec?
Reply by ●February 9, 20062006-02-09
It doesnt meat the spec thats what im saying, this area is one of the major problems with PCI and 3.3v systems such as your FPGA, as i said in our development boards we make sure the rest of the PCI bus meets there specification and then make sure the trace lengths of the clock signal either side of the bus switch is longer than your worst case by 0.5inches or greater. This works for our systems and we have seen no problems, making the trace longer ensures you meet the set up and hold times for the rest of the bus. This is no given thing it what we have seen working. REGARDS IAN "eehinjor" <eehinjor@163.com> wrote in message news:1139487456.386069.198790@g47g2000cwa.googlegroups.com...> Thank you for your response so soon. > 250ps is about 3inches,but the length of clk can not be longer than > 2.5+0.1inches. > how can that meet the spec? >






