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: 
Observer ahmedmohamed85
Observer
1,618 Views
Registered: ‎07-02-2015

5V IC connected to Artix7

Jump to solution

HI

I have designed a PCB that connects the output of IC HI-8588 to Artix7 FPGA to be used as an input, the HI-8588 level is 4.5 V but as i understand from the datasheet it's logic output drive current at logic one is maximum -0.8 ma and at logic zero is typically 5.6 ma, i checked the Artix7 Maximum current through any pin and i found it to be 10 ma, is it safe to continue using this PCB?

Best regards

0 Kudos
1 Solution

Accepted Solutions
Scholar u4223374
Scholar
2,412 Views
Registered: ‎04-26-2015

Re: 5V IC connected to Artix7

Jump to solution

@ahmedmohamed85 Realistically, in this case, the HI-8588 would be unable to pull the pin up to 5V. The 10mA current through the clamp diode would exceed what the output could provide. However, the HI-8588 datasheet does not provide enough information to determine what would happen in this case.

 

First, the maximum output current is not defined (because the high-level currents are listed as negative, this would actually be in the "min" column). It would probably not be more than 5mA, but we can't be sure.

 

Second, we don't know what the characteristics are as the output gets pulled below 4.6V. Suppose we provide a low enough resistance to pull the pin down to 4V. Will the pin now output 30mA? Will the HI-8588 output driver be destroyed? Will the HI-8588 break so badly that it takes down the entire ARINC 429 bus? Or will it just provide 6mA at 4V? We don't know. Even testing it doesn't really help - there's no guarantee that all chips (even with the same part number) will behave the same, nor is there a guarantee that what works once will continue to work in the long-term.

 

The Artix side is similar. If you feed an Artix 7 pin 4.6V, my guess is that it'll work. For at least a few hours, probably at least a week. After that, who knows? After a few months you might get units coming back with random faults - one totally dead, one with a moderate number of errors on the ARINC interface, etc. The engineering time to identify this problem, redesign the boards, and replace all units already in service will be massive.

 

On a practical note, ARINC 429 is an avionics interface (as I'm sure you're aware). This seems like a situation where relying on undocumented behaviour is an exceptionally bad idea, especially if you're intending to have the product certified.

 

Just stick either a resistor voltage divider or an active level translator in there.

6 Replies
Xilinx Employee
Xilinx Employee
1,607 Views
Registered: ‎08-01-2008

Re: 5V IC connected to Artix7

Jump to solution

No, none of the pins are 5V tolerant, under any circumstances.

Please use level translation circuits, either a simple resistor divider (OK for unidirectional I/O) or a dedicated level shifter.

 

you can refer these documents for more details and ratings

https://www.xilinx.com/support/documentation/data_sheets/ds181_Artix_7_Data_Sheet.pdf

https://www.xilinx.com/support/documentation/user_guides/ug483_7Series_PCB.pdf

Thanks and Regards
Balkrishan
--------------------------------------------------------------------------------------------
Please mark the post as an answer "Accept as solution" in case it helped resolve your query.
Give kudos in case a post in case it guided to the solution.
0 Kudos
Observer ahmedmohamed85
Observer
1,606 Views
Registered: ‎07-02-2015

Re: 5V IC connected to Artix7

Jump to solution

But the maximum current is 0.8 ma, this means no current more than 10 ma will go through the FPGA PIN, what would be the danger?

 

0 Kudos
Moderator
Moderator
1,553 Views
Registered: ‎04-18-2011

Re: 5V IC connected to Artix7

Jump to solution
You've misread the datasheet a little bit here.
There is a note about Iin that says something like 10mA when forward biasing the clamp diode in the IO.
So yes when you forward bias the clamp diode you are required to limit the current.
But you've missed the other part of the puzzle and that is the absolute maximum VIN, which is the max voltage you can apply to the pin. 5V will violate this
-------------------------------------------------------------------------
Don’t forget to reply, kudo, and accept as solution.
-------------------------------------------------------------------------
Highlighted
Observer ahmedmohamed85
Observer
1,543 Views
Registered: ‎07-02-2015

Re: 5V IC connected to Artix7

Jump to solution

Even if the driving current is low ?

0 Kudos
Scholar u4223374
Scholar
2,413 Views
Registered: ‎04-26-2015

Re: 5V IC connected to Artix7

Jump to solution

@ahmedmohamed85 Realistically, in this case, the HI-8588 would be unable to pull the pin up to 5V. The 10mA current through the clamp diode would exceed what the output could provide. However, the HI-8588 datasheet does not provide enough information to determine what would happen in this case.

 

First, the maximum output current is not defined (because the high-level currents are listed as negative, this would actually be in the "min" column). It would probably not be more than 5mA, but we can't be sure.

 

Second, we don't know what the characteristics are as the output gets pulled below 4.6V. Suppose we provide a low enough resistance to pull the pin down to 4V. Will the pin now output 30mA? Will the HI-8588 output driver be destroyed? Will the HI-8588 break so badly that it takes down the entire ARINC 429 bus? Or will it just provide 6mA at 4V? We don't know. Even testing it doesn't really help - there's no guarantee that all chips (even with the same part number) will behave the same, nor is there a guarantee that what works once will continue to work in the long-term.

 

The Artix side is similar. If you feed an Artix 7 pin 4.6V, my guess is that it'll work. For at least a few hours, probably at least a week. After that, who knows? After a few months you might get units coming back with random faults - one totally dead, one with a moderate number of errors on the ARINC interface, etc. The engineering time to identify this problem, redesign the boards, and replace all units already in service will be massive.

 

On a practical note, ARINC 429 is an avionics interface (as I'm sure you're aware). This seems like a situation where relying on undocumented behaviour is an exceptionally bad idea, especially if you're intending to have the product certified.

 

Just stick either a resistor voltage divider or an active level translator in there.

Xilinx Employee
Xilinx Employee
1,467 Views
Registered: ‎10-11-2007

Re: 5V IC connected to Artix7

Jump to solution

Asserting a higher voltage on a Artix HR pin than speced is not a good idea will cause long term degradation and eventually failure. Current or current limit doesn't matter. There is an effect in this and later technologies called HCS (Hot Carrier Injection). This can degrade the gate dielectric, causing electron and hole traps to form. Time to failure depends on process, temperature and Vin. Without going into further details, you can imagine that that is not good. You can probably google it if you are curious. 

0 Kudos