HEATING PROBLEM of Artix-7 XC7A100T FPGA Development Board

[relpro]

Hi,

My Artix-7 XC7A100T FPGA Development Board gets hot just as I plugin the power cable.
I have attached a pic showing the temperature. How can I fix this problem?

 

[KlausST]

Hi,

we know very little about your application.

So generally speaking: Reduce the current drawn by the FPGA.

High current my come from
* high clock speed
* heavy processing (many switching nodes)
* short circuited outputs
* outputs fighting against each other
* floating inputs
....

and so on....

****
so maybe do a supply current test with no code or with all I/Os programmed as weak pullup or pulldown, just to get a clue about expectable idle current.

Klaus

 

[stenzer]

Hi,

it is not unusual for high speed devices to get quite hot, so what is the clock frequency? Further are you measureing an internal temperature of the FPGA, or an external one? According to the signal name I would assume it's an external one.

• So where is the temperature sensor located at?
• What's the steady state temperature?
• Is the board performing its intended operation, or is it not working properly?

BR

 

[relpro]

Hi,

we know very little about your application.

So generally speaking: Reduce the current drawn by the FPGA.

High current my come from
* high clock speed
* heavy processing (many switching nodes)
* short circuited outputs
* outputs fighting against each other
* floating inputs
....

and so on....

****
so maybe do a supply current test with no code or with all I/Os programmed as weak pullup or pulldown, just to get a clue about expectable idle current.

Klaus

The FPGA is not loading anything from flash. Even with a heat sink it gets to 74 degrees.
[Moderator action: removed link to external file server]

 

[stenzer]

Hi,

but I assume some kind of code is running on it (at which frequency), or how are you are measuring the temperature?

BR

 

[FvM]

The OP is apparently assuming that we already know every detail about the board, e.g. how the temperature measurement is working, so he won't need to give meaningful information. Unfortunately that's not the case.

I read the report so that the board has FPGA independent supervisor functions that give the temperature measurement. We should also know if it's a board or FPGA chip temeperature.

My first question is, has the board already been operated before, or is this the first attempt to power it? One would also check if all power supplies are up with nominal voltage.

 

[barry]

Since OP is giving zero information, let me make a guess: They’re measuring FPGA internal temperature. In that case 75 degrees is not unreasonable.

 

[andre_luis]

relpro, unless you provide environmental temperature nearby this board ( inside the enclosure, if the case ), giving only the absolute temperature of whatever you're measuring is useless at all.