PCI Noise

Hi,

I would like to make a simple PCI device.  I already own an FPGA evaluation
board and I also own a blank PCI prototyping card.  I was planning on
connecting the FPGA board to the PCI prototyping card via ribbon cables.
Would this cause problems with noise?  Would it help if I kept the ribbon
cables quite short.

Thanks for any info,

"Ted" <ted@ted.com> wrote in message
news:chf9eq$frj$1@newsg2.svr.pol.co.uk...
> Hi,
>
> I would like to make a simple PCI device.  I already own an FPGA
evaluation
> board and I also own a blank PCI prototyping card.  I was planning on
> connecting the FPGA board to the PCI prototyping card via ribbon cables.
> Would this cause problems with noise?  Would it help if I kept the ribbon
> cables quite short.

They'd have to be very short. 8-)

Leon

Ted wrote:

> I would like to make a simple PCI device.  I already own an FPGA evaluation
> board and I also own a blank PCI prototyping card.  I was planning on
> connecting the FPGA board to the PCI prototyping card via ribbon cables.
> Would this cause problems with noise?

Well, what's noise ;-)  There's more to the PCI signaling environment
than you may be aware of.  I strongly suggest obtaining the PCI spec or
if that isn't feasible this one may be of considerable (but not complete)
help:

http://www.aw-bc.com/catalog/academic/product/0,1144,0201309742,00.html


Cheers,
Chris

generous coupling to ground with frequency matched capacitors will help you
lengthen signal lines without noise degradation.

"Orbit" <orbit@dunno.net> wrote in message
news:ANydnUUxBYdvzabcRVn-tw@giganews.com...
> generous coupling to ground with frequency matched capacitors will help
you
> lengthen signal lines without noise degradation.

The lengths of the lines from the PCI bus connector are critical, and are
quite short.


Leon
-- 
Leon Heller, G1HSM
http://www.geocities.com/leon_heller

"Orbit" <orbit@dunno.net> wrote in message
news:ANydnUUxBYdvzabcRVn-tw@giganews.com...
> generous coupling to ground with frequency matched capacitors will help
you
> lengthen signal lines without noise degradation.

???

Please enlighten us.

Bob

I had a silimar problem in a different area (DDS's and A/D converters). The 
only solution was to throw away the ribbon and make a small pc board that 
interconnects the two eval boards, using a group plane beneth signal trance 
wires, and the approapriate connectors

"Ted" <ted@ted.com> wrote in message 
news:chf9eq$frj$1@newsg2.svr.pol.co.uk...
> Hi,
>
> I would like to make a simple PCI device.  I already own an FPGA 
> evaluation
> board and I also own a blank PCI prototyping card.  I was planning on
> connecting the FPGA board to the PCI prototyping card via ribbon cables.
> Would this cause problems with noise?  Would it help if I kept the ribbon
> cables quite short.
>
> Thanks for any info,
>
> 

"Ted Lechman" <eastwood132@yahoo.com> wrote in message 
news:CaednQ1g-ZMpPqbcRVn-hw@adelphia.com...
>I had a silimar problem in a different area (DDS's and A/D converters). The 
>only solution was to throw away the ribbon and make a small pc board that 
>interconnects the two eval boards, using a group plane

I meant "ground plane" beneath the signal trace lines


beneth signal trance
> wires, and the approapriate connectors
>
> "Ted" <ted@ted.com> wrote in message 
> news:chf9eq$frj$1@newsg2.svr.pol.co.uk...
>> Hi,
>>
>> I would like to make a simple PCI device.  I already own an FPGA 
>> evaluation
>> board and I also own a blank PCI prototyping card.  I was planning on
>> connecting the FPGA board to the PCI prototyping card via ribbon cables.
>> Would this cause problems with noise?  Would it help if I kept the ribbon
>> cables quite short.
>>
>> Thanks for any info,
>>
>>
>
> 

http://tinyurl.com/42x6o
as to the construction of the ribbon cable - you must solder your
connections and wash the board of any flux and keep the length as short as
practically possible.

Have I done it with PCI? no. Is is possible? probably. Does it meet IEE
standards? of course not, but this is a one-off board.

good luck.

Hi,

If you study the PCI specs, you will see that there is a maximum lenght
specified for the pcb traces (1.5"  and 2.5 " for the clock line) . This is
quite important because the signals and switching points rely on the
reflected wave principle of the PCI bus.!!!

I suggest you obtain the PCI specs. Your io ports on your FPGA must also be
pci compliant and you really should take care of the right timing
constraints to and from the PCI io pads, otherwise you might experience a
lot of problems on the PCI bus.


regards


ron proveniers



"Ted" <ted@ted.com> schreef in bericht
news:chf9eq$frj$1@newsg2.svr.pol.co.uk...
> Hi,
>
> I would like to make a simple PCI device.  I already own an FPGA
evaluation
> board and I also own a blank PCI prototyping card.  I was planning on
> connecting the FPGA board to the PCI prototyping card via ribbon cables.
> Would this cause problems with noise?  Would it help if I kept the ribbon
> cables quite short.
>
> Thanks for any info,
>
>