This list is for discussion of the design and implementation of field-programmable gate array based processors and integrated systems. It is also for discussion and community support of the XSOC Project (see http://www.fpgacpu.org/xsoc).
|
Austin- > I disagree. That may not be an option, and in that case, the better > solution is to use quickswitches, which do in fact work just fine. The > issue I personally have with it is the larger FPGAs are simply not 5V > tolerant, and the majority of PCI busses are still 5V. I have had long > discussions with Xilinx about this, and I also understand the reasons for > their not supporting 5V tolerance. > > I've laid out all the facts as I know them. I'll add that I speak from a > reasonable amount of PCI experience, having designed over 20 PCI cards (and > quite a few of their PCI interfaces), was on the PCI spec. review committee, > I designed the first PCI interface for Xilinx, and I've done the designs for > four PCI ASICs, so I am pretty experienced when it comes to PCI. I believe > we just have to agree to disagree. Based on this... I just ripped the 220 Rs out of a Mini PCI design and found enough board space for switches. Thanks Austin. -Jeff |
|
|
|
Hi Jeff, > Based on this... I just ripped the 220 Rs out of a Mini PCI > design and found enough > board space for switches. > > Thanks Austin. You're welcome. Keep the trace length between the switches and the IC the shortest you can, and choose your switches wisely ;-) If you are pinning for an FPGA, make sure your FPGA pinout optimizes the routing...and also I'd suggest routing on the outer layers only (minimizing vias, such that you have none on the top layer, and only one to get to the bottom layer) and making sure you watch your trace characteristic impedance (which for PCI is 60-100 ohms...and get a copy of the PCI spec and look over the routing section for trace lengths etc.), which means you (or your PCB fabricator) have to calculate the trace widths, and make sure the PCB manufacturer knows what they are when you order the boards, in fact, I'd suggest asking them to work up a stack-up for you prior to PCB layout. Please feel free to ask me questions with regards to the PCI aspect of this if you have any. Regards, Austin P.S. BTW, are you making a 5V only, or a "universal" card (3.3V and 5V)? If a universal card, there are other issues you may want to be aware of. |
|
|
|
Jeff wrote: > I just ripped the 220 Rs out of a Mini PCI design and found enough > board space for switches. Austin wrote: > P.S. BTW, are you making a 5V only, or a "universal" card (3.3V and 5V)? MiniPCI is 3.3V only. |
|
Hi, Austin and the group gurus, It's very helpful to read this thread about universal PCI implementation (3.3V and 5V) while I am in the middle of design a PTMC card. My card needs to support 3.3V and 5V and I am planing to go with the quickswitchs method. Do you have any suggestion on this switch chip selection? What is the one you have used and prove to be good? My card is pretty dense and I am running out of space quickly on this little PTMC card. To add level translation to about 50 PCI signals does add some stress to the board space, but I guess I have no other choice, right? BTW, I am only supporting 32bit/33MHz PCI. Thank you in advance. Regards, Jane Austin Franklin wrote: >Hi Jeff, > >>Based on this... I just ripped the 220 Rs out of a Mini PCI >>design and found enough >>board space for switches. >> >>Thanks Austin. >> >> > >You're welcome. Keep the trace length between the switches and the IC the >shortest you can, and choose your switches wisely ;-) If you are pinning >for an FPGA, make sure your FPGA pinout optimizes the routing...and also I'd >suggest routing on the outer layers only (minimizing vias, such that you >have none on the top layer, and only one to get to the bottom layer) and >making sure you watch your trace characteristic impedance (which for PCI is >60-100 ohms...and get a copy of the PCI spec and look over the routing >section for trace lengths etc.), which means you (or your PCB fabricator) >have to calculate the trace widths, and make sure the PCB manufacturer knows >what they are when you order the boards, in fact, I'd suggest asking them to >work up a stack-up for you prior to PCB layout. > >Please feel free to ask me questions with regards to the PCI aspect of this >if you have any. > >Regards, > >Austin > >P.S. BTW, are you making a 5V only, or a "universal" card (3.3V and 5V)? If >a universal card, there are other issues you may want to be aware of. > > >To post a message, send it to: >To unsubscribe, send a blank message to: >Yahoo! Groups Links [Non-text portions of this message have been removed] |
|
|
|
Hi Jane, Are you interfacing with an FPGA? If so, typically, you'll have to provide external circuitry to do the detection of the 3.3/5V and provide differing bitstreams, one for each voltage. This is a separate issue from the use of QuickSwitches. I believe QuickSwitch has been bought out by IDT, so I'd check with IDT. Also keep in mind, you've got to accommodate the VIO voltage difference with the switches as well. Regards, Austin > Hi, Austin and the group gurus, > > It's very helpful to read this thread about universal PCI implementation > (3.3V and 5V) while I am in the middle of design a PTMC card. > > My card needs to support 3.3V and 5V and I am planing to go with the > quickswitchs method. Do you have any suggestion on this switch chip > selection? What is the one you have used and prove to be good? My card > is pretty dense and I am running out of space quickly on this little > PTMC card. To add level translation to about 50 PCI signals does add > some stress to the board space, but I guess I have no other choice, right? > > BTW, I am only supporting 32bit/33MHz PCI. > > Thank you in advance. > > Regards, > > Jane > Austin Franklin wrote: > > >Hi Jeff, > > > > > > > >>Based on this... I just ripped the 220 Rs out of a Mini PCI > >>design and found enough > >>board space for switches. > >> > >>Thanks Austin. > >> > >> > > > >You're welcome. Keep the trace length between the switches and > the IC the > >shortest you can, and choose your switches wisely ;-) If you are pinning > >for an FPGA, make sure your FPGA pinout optimizes the > routing...and also I'd > >suggest routing on the outer layers only (minimizing vias, such that you > >have none on the top layer, and only one to get to the bottom layer) and > >making sure you watch your trace characteristic impedance (which > for PCI is > >60-100 ohms...and get a copy of the PCI spec and look over the routing > >section for trace lengths etc.), which means you (or your PCB fabricator) > >have to calculate the trace widths, and make sure the PCB > manufacturer knows > >what they are when you order the boards, in fact, I'd suggest > asking them to > >work up a stack-up for you prior to PCB layout. > > > >Please feel free to ask me questions with regards to the PCI > aspect of this > >if you have any. > > > >Regards, > > > >Austin > > > >P.S. BTW, are you making a 5V only, or a "universal" card (3.3V > and 5V)? If > >a universal card, there are other issues you may want to be aware of. > > > > > > > > > >To post a message, send it to: > >To unsubscribe, send a blank message to: > > >Yahoo! Groups Links > > > > > > > > > > > > > > > > > > > [Non-text portions of this message have been removed] > > To post a message, send it to: > To unsubscribe, send a blank message to: > > Yahoo! Groups Links |