UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

cancel
Showing results for 
Search instead for 
Did you mean: 
Newbie roymirabal
Newbie
750 Views
Registered: ‎02-15-2017

XC7S50-1FFGA484I Backdriving

Jump to solution

I am currently integrating a design that utilizes legacy Virtex II devices with a newer XC7S50 device.  I had to implement power sequencing separate from the legacy design, specifically 1.0V, 1.8V & 3.3V,  but the 7 series device interfaces with a different 3.3V rail.  This legacy 3.3V rail appears to backdrive the new 3.3V rail for the 7 series device.  I isolated the JTAG chain so that is not the source.  Is there any recommendations as to where this backdrive may be coming from?

0 Kudos
1 Solution

Accepted Solutions
Newbie roymirabal
Newbie
666 Views
Registered: ‎02-15-2017

Re: XC7S50-1FFGA484I Backdriving

Jump to solution

Thanks for the response.  I believe this closes this SR.


@klumsde wrote:

It is not violated since you are not exceeding the TVCCO2VCCAUX time. 

 

That is not to say that this is a good situation. You are turning on the clamp diode. 

Do you know we have a spec called Iin. 

There is a limit to the amount of current that can flow into this pin when the clamp is forward biased. 10mA maximum. 

If this is on more than 1 pin in the bank the limit is 200mA per bank. 

 

 Also your DC-DC converter. Is it able to cope with a current flowing back into it's input in this way?

 


 

0 Kudos
8 Replies
Moderator
Moderator
724 Views
Registered: ‎04-18-2011

Re: XC7S50-1FFGA484I Backdriving

Jump to solution
You are going to have to look closely at the design and see what on the Virtex II device is connected. Specifically the 3V3 IO.
If you have a bank on the spartan 7 that is not powered then if a 3V3 volt IO drives this unpowered IO bank you will turn on a clamp diode to VCCO in the IO.
This will back power the rail to a diode drop below the input.
Perhaps the IO on the virtex are pulling up during the boot of the virtex device. There is usually an option to just let the IO go into trí state during configuration. Try this and see is the back powering still going on.

Since this is 3.3v then you have to be careful. In 7 series vcco is not allowed to be more than 2.625V above VCCAUX for longer than TVCCO2VCCAUX. Repeatedly violating this will reduce the lifetime of the device.
-------------------------------------------------------------------------
Don’t forget to reply, kudo, and accept as solution.
-------------------------------------------------------------------------
0 Kudos
Newbie roymirabal
Newbie
719 Views
Registered: ‎02-15-2017

Re: XC7S50-1FFGA484I Backdriving

Jump to solution

Thanks for the feedback.  Could you please explain the 2.625V above VCCAUX for longer than TVCCO2VCCAUX requirement?  This is exactly what I am concerned with.

0 Kudos
Newbie roymirabal
Newbie
717 Views
Registered: ‎02-15-2017

Re: XC7S50-1FFGA484I Backdriving

Jump to solution

In addition, the VCCO - VCCAUX is <60ms

 

0 Kudos
Newbie roymirabal
Newbie
706 Views
Registered: ‎02-15-2017

Re: XC7S50-1FFGA484I Backdriving

Jump to solution

It appears from the datasheet that I am not exceeding the worst case with Tj = 125C which corresponds to 300ms max of VCCO exceeding VCCAUX.  Am I interpreting this correctly?

0 Kudos
Moderator
Moderator
686 Views
Registered: ‎04-18-2011

Re: XC7S50-1FFGA484I Backdriving

Jump to solution

yes 300ms is the time allowed VCCO to be above VCCAUX at this temperature point. 

It is longer at lower temperatures. 

 

However this spec is not a one time thing. So for 85C if you violated it for longer than 800ms then you won't damage it immediately. 

The spec is based on repeatedly violating it over a number of power cycles. 

The higher the VCCO is will reduce this number of power cycles. 

 

 

-------------------------------------------------------------------------
Don’t forget to reply, kudo, and accept as solution.
-------------------------------------------------------------------------
0 Kudos
Newbie roymirabal
Newbie
679 Views
Registered: ‎02-15-2017

Re: XC7S50-1FFGA484I Backdriving

Jump to solution

I am attaching 2 plots.  The 1st one shows the 1.0V & the 1.8V.  The 2nd shows the 3.3V with the back-drive voltage present.  As you can see from the plots, the overall time that VCCO exceeds VCCAUX is approximately 25ms.  This would be for every power up.  What I am up against is I am not allowed to change the programming files for the other 2 Virtex II devices.  Is this a violation based on the datasheet as this time does not exceed the TVCCO2VCCAUX time?  I will never exceed 25ms of back-drive the way this design is set  up.

6.5v vs 3.3v Scaler.jpg
0 Kudos
Moderator
Moderator
669 Views
Registered: ‎04-18-2011

Re: XC7S50-1FFGA484I Backdriving

Jump to solution

It is not violated since you are not exceeding the TVCCO2VCCAUX time. 

 

That is not to say that this is a good situation. You are turning on the clamp diode. 

Do you know we have a spec called Iin. 

There is a limit to the amount of current that can flow into this pin when the clamp is forward biased. 10mA maximum. 

If this is on more than 1 pin in the bank the limit is 200mA per bank. 

 

 Also your DC-DC converter. Is it able to cope with a current flowing back into it's input in this way?

 

-------------------------------------------------------------------------
Don’t forget to reply, kudo, and accept as solution.
-------------------------------------------------------------------------
0 Kudos
Newbie roymirabal
Newbie
667 Views
Registered: ‎02-15-2017

Re: XC7S50-1FFGA484I Backdriving

Jump to solution

Thanks for the response.  I believe this closes this SR.


@klumsde wrote:

It is not violated since you are not exceeding the TVCCO2VCCAUX time. 

 

That is not to say that this is a good situation. You are turning on the clamp diode. 

Do you know we have a spec called Iin. 

There is a limit to the amount of current that can flow into this pin when the clamp is forward biased. 10mA maximum. 

If this is on more than 1 pin in the bank the limit is 200mA per bank. 

 

 Also your DC-DC converter. Is it able to cope with a current flowing back into it's input in this way?

 


 

0 Kudos