Don't have a Xilinx account yet?

  • Choose to receive important news and product information
  • Gain access to special content
  • Personalize your web experience on Xilinx.com

Create Account

Username

Password

Forgot your password?
XClose Panel
Xilinx Home
Reply
Newbie
oppie
Posts: 2
Registered: ‎03-04-2011
0

5V tolerant I/O?

I'm new to Xilinx and could use some recommendations please.

Are there any current CPLD devices (recommended for new design) that have 5V tolerant I/O?

 

Looking to replace an Atmel ATF1508AS (128 macrocell) device that is getting hard to find. Have been getting that device in a PLCC84 pkg so it can be socketed.

Expert Contributor
gszakacs
Posts: 7,135
Registered: ‎08-14-2007
0

Re: 5V tolerant I/O?

The XC9500 series is a true 5V part and has a 108-macrocell version in the PLCC84

package.  I'm not sure if these will go away soon, but they still show up as stock

items on DigiKey.

 

The 3.3V XC9500XL series is getting old, but still very much in use.  It can handle 5V on

its pins. 

 

The CoolRunner II series is not 5V tolerant.

 

Also at the moment you can still get Spartan 2 FPGA's that are 5V tolerant (not the 2e),

but I believe this will not be available in the long term.  Overall 5V tolerance is harder

to find, and eventually will go away altogether.  There are quite a few appnotes on

interfacing 5V logic to non-5V-tolerant parts.

 

-- Gabor

-- Gabor
Newbie
oppie
Posts: 2
Registered: ‎03-04-2011
0

Re: 5V tolerant I/O?

Looks like 5V compliant is not going to be around long term....

Part was originally chosen as an I/O buffer with options, going off-board to a 5V world. Socketed, so it could be replaced easily if damaged.

I'm going to lobby our chief designer to use one of the newer 3.3V I/O devices and use simple level shifting buffers to the outside world. I can probably sell it alone on the fact that the lower voltage devices use less silicon real estate and thus cost less.

  Cheers

Expert Contributor
gszakacs
Posts: 7,135
Registered: ‎08-14-2007
0

Re: 5V tolerant I/O?

Yes.  Large socketable parts are on the way out also.  At least for embedded systems

the big push is to get more functionality in a smaller footprint.  Also note that large

non-BGA packages can have very large lead inductance which is not good for high-

speed logic.  So even big square packages like PQFP's are on the way out.  In any

case you can probably save money using a small-footprint high-densiy modern

logic device even after adding a number of jelly-bean level-shifters.  And then the

level-shifters become the replaceable parts if  you ESD the board.

 

Regards,

Gabor

-- Gabor
Xilinx Employee
barriet
Posts: 2,467
Registered: ‎08-13-2007

Re: 5V tolerant I/O?

As Gabor said, the last of the Xilinx FPGAs that were natively 5V tolerant were Virtex (the original) and Spartan-II.

 

The 95XL is 5V tolerant - when powered. This is potentially easy to miss - check the absolute max specs. This could be potentially important depending on power sequencing, when interfacing with signals coming from off-board, etc.

 

Here's some useful app notes:

http://www.xilinx.com/support/documentation/application_notes/xapp429.pdf (5V Tolerance Techniques for CoolRunner-II Devices)
http://www.xilinx.com/support/documentation/application_notes/xapp459.pdf (Eliminating I/O Coupling Effects when Interfacing Large-Swing Single-Ended Signals to User I/O Pins on Spartan-3 Generation FPGAs)
http://www.xilinx.com/support/documentation/application_notes/xapp646.pdf (Connecting Devices in the Virtex and Spartan Families to a 3.3V or 5V PCI Bus)

 

Virtex-6 is not even 3.3V tolerant  - the necessary tradeoff for extremely fast LVDS and DDR3 interfacing:

http://www.xilinx.com/support/documentation/application_notes/xapp899.pdf  (Interfacing Virtex-6 FPGAs with 3.3V I/O Standards)

Cheers,

bt

Visitor
kzietlow
Posts: 13
Registered: ‎08-21-2010
0

Re: 5V tolerant I/O?

 


timpe wrote: 

The 95XL is 5V tolerant - when powered. This is potentially easy to miss - check the absolute max specs. This could be potentially important depending on power sequencing, when interfacing with signals coming from off-board, etc.

 

Cheers,

bt


 

I have seen this note in the data sheet which you seem to be referring to:

 

External I/O voltage may not exceed VCCINT by 4.0V.

 

However, it doesn't say what happens if you do. Does it blow a gate oxide or will it start cross conducting from the I/O to the supply pins?

 

Thanks for any clarifications you can provide on this.

 

Klaus

 

 

Expert Contributor
eteam00
Posts: 8,350
Registered: ‎07-21-2009
0

Re: 5V tolerant I/O?

Does it blow a gate oxide or will it start cross conducting from the I/O to the supply pins?

"Cross conducting" is likely to result in significant localised heating of the die.  Net result is same:  poof! (technical term)

 

-- Bob Elkind

SIGNATURE:
README for newbies is here: http://forums.xilinx.com/t5/New-Users-Forum/README-first-Help-for-new-users/td-p/219369

Summary:
1. Read the manual or user guide. Have you read the manual? Can you find the manual?
2. Search the forums (and search the web) for similar topics.
3. Do not post the same question on multiple forums.
4. Do not post a new topic or question on someone else's thread, start a new thread!
5. Students: Copying code is not the same as learning to design.
6 "It does not work" is not a question which can be answered. Provide useful details (with webpage, datasheet links, please).
7. You are not charged extra fees for comments in your code.
8. I am not paid for forum posts. If I write a good post, then I have been good for nothing.
Visitor
kzietlow
Posts: 13
Registered: ‎08-21-2010
0

Re: 5V tolerant I/O?

 


eteam00 wrote:

Does it blow a gate oxide or will it start cross conducting from the I/O to the supply pins?

"Cross conducting" is likely to result in significant localised heating of the die.  Net result is same:  poof! (technical term)

 

-- Bob Elkind


 

Not necessarily. If it was a protection circuit that behaved a bit like a 4 V zener then a 5V I/O voltage could just lift the supply to 1 V. At this low a voltage the supply might not draw a whole lot of current  and this could be safe. Also the driver would likely be current limited or could be current limited with a simple series resistor.

Overall this limitation seems a bit lame considering the claim of 5V tolerant I/Os made in the data sheet, quoted below:

 

5V Tolerant I/Os
The I/Os on each XC9500XL device are fully 5V tolerant
even though the core power supply is 3.3 volts. This allows
5V CMOS signals to connect directly to the XC9500XL
inputs without damage. In addition, the 3.3V VCCINT power
supply can be applied before or after 5V signals are applied
to the I/Os. In mixed 5V/3.3V/2.5V systems, the user pins,
the core power supply (VCCINT), and the output power supply
(VCCIO) may have power applied in any order. This
makes the XC9500XL devices immune to power supply
sequencing problems.

 

Klaus

 

 

Expert Contributor
eteam00
Posts: 8,350
Registered: ‎07-21-2009
0

Re: 5V tolerant I/O?

[ Edited ]

 


kzietlow wrote:

 


eteam00 wrote:

Does it blow a gate oxide or will it start cross conducting from the I/O to the supply pins?

"Cross conducting" is likely to result in significant localised heating of the die.  Net result is same:  poof! (technical term)

 

-- Bob Elkind


Not necessarily.


Klaus!   I can tell you are a wild and crazy gambling man!!

 

In what planet or dimension would 'it might self-destruct on first plug-in, but I can't guarantee it' be a plausible design review response?

 

-- Bob Elkind

SIGNATURE:
README for newbies is here: http://forums.xilinx.com/t5/New-Users-Forum/README-first-Help-for-new-users/td-p/219369

Summary:
1. Read the manual or user guide. Have you read the manual? Can you find the manual?
2. Search the forums (and search the web) for similar topics.
3. Do not post the same question on multiple forums.
4. Do not post a new topic or question on someone else's thread, start a new thread!
5. Students: Copying code is not the same as learning to design.
6 "It does not work" is not a question which can be answered. Provide useful details (with webpage, datasheet links, please).
7. You are not charged extra fees for comments in your code.
8. I am not paid for forum posts. If I write a good post, then I have been good for nothing.
Expert Contributor
eteam00
Posts: 8,350
Registered: ‎07-21-2009
0

Another level-shifter option

I'm going to lobby our chief designer to use one of the newer 3.3V I/O devices and use simple level shifting buffers to the outside world.

An alternative to the level shifter are clamp/isolation FETs.  Here's an example clipped from the SP605 development board schematics.

forums isolation.jpg

 

The NDS331N nFET keeps the FPGA (CPLD in your case) connections (to the left of the nFET) from exceeding 2.5V supply voltage.

 

Not only does this circuit provide voltage limiting at the FPGA/CPLD, it also provides hot-swap isolation.  The nFET is essentially cut off (FPGA/CPLD is isolated) until power (2.5V supply) is applied.

 

In your case, you would probably use a 3.3V supply instead of 2.5V.  The nFET would allow active push-pull signal drive up to the gate voltage.  If you want true 5V swings for CPLD outputs driving the 5V circuitry, you would need to add 5V pullups on the 5V load side (the right side of the nFETs).  In sum:  active high and low drive in both directions up to 3.3V, with full 5V swing provided by a pullup R.

 

The clamp diodes probably aren't needed -- for 5V circuitry on the same board -- unless you are passing signals through long cables.

 

I don't think this would perform well above 100Mb/sec bandwidth, but it would certainly work well up to at least 400Kb/sec.

 

For cost and power consumption, this isn't a bad little circuit.  I am cheerfully stealing this idea from Xilinx' dev board designs for use in my own boards (substituting a lower-cost, lower current nFET in place of the NDS331n).

 

-- Bob Elkind

SIGNATURE:
README for newbies is here: http://forums.xilinx.com/t5/New-Users-Forum/README-first-Help-for-new-users/td-p/219369

Summary:
1. Read the manual or user guide. Have you read the manual? Can you find the manual?
2. Search the forums (and search the web) for similar topics.
3. Do not post the same question on multiple forums.
4. Do not post a new topic or question on someone else's thread, start a new thread!
5. Students: Copying code is not the same as learning to design.
6 "It does not work" is not a question which can be answered. Provide useful details (with webpage, datasheet links, please).
7. You are not charged extra fees for comments in your code.
8. I am not paid for forum posts. If I write a good post, then I have been good for nothing.