Xilinx Spartan II and 5V PCI

Christian,

Yes.  After I sent the first email, I realized that you were using 
Spartan II, not 3.  Sigh.  Next time I will be more careful on the send 
button.  Engage brain.

Virtex and Spartan II are what I call "Classic" FPGA parts:  they do 5V 
PCI, they have hot insertion tri-state behavior, etc.  In a way, the 
last of the great simple FPGAs.  Everything since is based on the 
process technology getting more complicated (not to mention more 
internal features).  In order to stay with high yields, we had to drop 
the floating outputs (not supported by foundries, so if anything goes 
wrong, like ESD -- you are on your on, and you own the wafers! which 
leads to higher costs to customers) as well as other nice features (like 
0.35u IO transistors which are bulletproof).  But a V1000 is old news, 
with a million "gates" becoming a small part today.  The 2VP7 is now 
considered a very small part, with 3S1000 being the median member of the 
Soartan 3 family.

S3 by the way has (finally) passed through the difficult times (first 
was process a year ago, next was demand was >>> supply).  All parts 
(except one, again due to unprecedented demand) are in stock, and the 
last two largest family members should be out shortly (3S5000 is just 
shy of a Virtex 2V6000, or a Virtex II Pro 2VP50 in "gates").

Consult your disti,

Austin

Christian E. Boehme wrote:
> Austin Lesea wrote:
> 
>> You might try reading the app notes?
>>
>> http://www.xilinx.com/bvdocs/appnotes/xapp653.pdf
> 
> 
> That one I did read and considered it for a while but
> dropped later due to lack of availability of the part.
> 
>> http://www.xilinx.com/bvdocs/appnotes/xapp646.pdf
> 
> 
> This one I read also.  Too bad that it's not applicable
> to my problem (namely _5V_ PCI tolerance ;).
> 
> 
> -Chris
> 

Christian,

Yes, power would be the least of your worries:  you would be sinking 
huge currents from the 5V rail to the 3.3V rail.  Probably break something.

Austin

Christian E. Boehme wrote:

> Austin Lesea wrote:
> 
>> Spartan II is identical with Virtex, which means that the clamp diodes 
>> can be programmed.  Thus the device can operate just fine in the 5V 
>> environment, and does meet the overshoot requirement of the PCI test 
>> without any concerns.
> 
> 
> Clamping diodes only work in 3.3V PCI signaling environments.  Since
> these programmable diodes clamp to Vcco and Vcco is 3.3V it's not too
> hard to imagine what a 5V steady level would do to the power department ;)
> 
> 
> -Chris
> 

All,

The zener structure is the ESD protection.  It is not a simple zener. 
It is a complex SCR-zener ESD structure that was designed to pass the 
PCI requirement, yet also protect the device from ESD.  Verrrry tricky. 
  But it works, and gets tested, or it doesn't ship for Spartan II and 
Virtex.

Eric is right:  we stand behind the 5V and 3.3V PCI compliance on S2 and 
Virtex.

Austin

Christian E. Boehme wrote:

> Paul Fulghum wrote:
> 
>> The Spartan II datasheet, page 2 of module 2:
>>
>> Two forms of over-voltage protection are provided,
>> one that permits 5V compliance, and one that does not.
>> For 5V compliance, a zener-like structure connected
>> to ground turns on when the output rises to
>> approximately 6.5V.
> 
> 
> Probably slipped through because it says nothing about
> PCI, hmm ...
> 
>> Page 31 of module 2:
>>
>> I/Os configured for the PCI, 33 MHz, 5V standard are also
>> 5V-tolerant.
> 
>   ^^^^^^^^^^^
> 
> Exactly that was the reason for the concern.  I have been dealing
> with ``5V-tolerant'' standard logic (LVCMOS etcpp.) lately where
> the meaning of that term is technically something completely
> different from what I would expect from a ``5V-PCI-compliant''
> device built in 3.3V technology.
> 
> However, apart from the fuzzy 5V compliance and tolerance
> verbiage in the data sheet, that seems to answer my initial
> question.
> 
> 
> Thanks & regards,
> Christian Boehme
> 

Austin Lesea wrote:

> The zener structure is the ESD protection.  It is not a simple zener. It 
> is a complex SCR-zener ESD structure that was designed to pass the PCI 
> requirement, yet also protect the device from ESD.  Verrrry tricky.  But 
> it works, and gets tested, or it doesn't ship for Spartan II and Virtex.

The datasheet calls the zener 'overvoltage protection'.
Does it work as both overvoltage and ESD protection?

The IBIS file for Spartan-II in the
power clamp section of the PCI33_5 model shows
that little current (~2 nA) is passed when the
pin voltage goes over 6.5V

The ground clamp section shows significant
current when pin voltage drops below -0.5V
as expected for the ground clamp diode.

How does the zener structure protect from overvoltage
if it does not pass significant current?

Is the IBIS file not accounting for the zener structure?
Am I misinterpreting the IBIS file?

Thanks,
Paul

--
Paul Fulghum
paulkf@microgate.com

Christian E. Boehme wrote:
> However, apart from the fuzzy 5V compliance and tolerance
> verbiage in the data sheet, that seems to answer my initial
> question.

I definately agree the description is fuzzy ;-)
Lots of good info in the datasheet, but it sort
of dances around that particular issue.

--
Paul Fulghum
paulkf@microgate.com

Eric Crabill wrote:

> If you are building a universal card, the "right" way
> to do it is have two designs/bitstreams only differing
> in what SelectIO Mode is used.  Then, at power-on, use
> an analog comparator to compare the slot's VIO with
> ~4.15v (make it with a voltage divider...) and load one
> of two bitstreams based on the result.

Interestingly, that's very simlar to what I actually do
right now:  I am using two biased comparators comparing
PCI 5V with PCI VIO such that even with maximum allowable
voltage tolerances the result would be predictably correct.
Until now I'd written the first word in config ROM with some
well-known header telling the kind of config written to the
ROM and warned the user (some LED or whatever) if there was
some mismatch with the actual environment.  But since there's
ample space in the config ROM for at least two complete streams
your suggestion is the _definitive_ way to go.  It seems that
I wasn't dreaming up too unusual things here ;-)


Thanks & regards,
Christian Boehme

Austin Lesea wrote:

> Virtex and Spartan II are what I call "Classic" FPGA parts:  they do 5V 
> PCI, they have hot insertion tri-state behavior, etc.  In a way, the 
> last of the great simple FPGAs.

Spartan II is just perfect for my projects.  The one limitation that gave
me some headaches and that was the number of different Vcco for IO banks
available in PQFP versions (namely 1 ;).  But that was about it already.
I am actually more constrained by the software tools than by the target
hardware.


-Chris