Please Help. I need some ideas to isolate this problem :(

[zhiling0229]

Hi guys,

I need your help. I'm actually working on isolating a failure on an FPGA.

From the electrical analysis I was able to identified the failure was caused by a driver or a multiplexer (Circuit.jpg) which was stuck logic 0 instead of driving a 1.

I was able to disable the driver. In hope that when I transmit from another driver which shares the same line it did not pull to zero.

However I could not conclude that the driver is ok as the line connecting to the share line is quite far. I'm not sure a minor bridging to vss near the problematic driver might not be seen at it might be the buffer as the drivers were powerful enough to overcome the minor bridging.

Is there any other way which I can perform to isolate whether it is a multiplexer or a driver fault.

the whole design for the particular area range from diffusion, poly, metal 1 until 5.


Please anyone, can you give me some ideas how to proceed. Kind of desprate here.

 

[dkace]

Re: Please Help. I need some ideas to isolate this problem :

Maybe a silly question, but I want to see if I can follow: Have you checked the multiplexer - driver cell on there own? Does it work as a stand alone circuit?

D.

 

[zhiling0229]

Re: Please Help. I need some ideas to isolate this problem :

hi dkace,

Thanks for replying.

the problem is the driver-multiplexer cannot be tap out directly out of the FPGA.
But I have done the following:

The top eye view of the connection would be:

input > driver-multiplexer > bus line > output

I perform a few test. I changed the driver-multiplexer on the same bus line as seen in the picture while maintaining the same configuration it ables to toggle (Driver-Multiplexer 1). I activate the dummy driver-multiplexer it was pull up to 1. I revert back to that particular driver-multiplexer it fails to toggle.

So i conclude it is a driver-multiplexer failure.

What do you think?

 

[nan_ishan]

not able to open circuit.jpg

 

[dkace]

Re: Please Help. I need some ideas to isolate this problem :

I will continue with the dumm questions since I can't figure it out...

It fails to toggle we you activate the dummy?

You show 3 cells in a row and a dummy.

Is the one in the middle failing?

D.

 

[zhiling0229]

Re: Please Help. I need some ideas to isolate this problem :

hi dkace,

i show 4 cells in the row. The driver-multiplexer between the dummy and driver-multiplexer1 is failing.

The dummy cannot be toggle as I cannot connect it to an input (it has no fan-in) since it is just a dummy or a redundant driver-multiplexer.


any more queries to give u a full idea on the problem?

 

[dkace]

Re: Please Help. I need some ideas to isolate this problem :

I think I can't help specifically on this but I can keep hitting you with questions in case you miss any check.
I.E. Does the dummy creates a voltage drop or any other kind of error in the line that will cause the first cell not to toggle? My thinking is that if the dummy is shown as a resistor (let's say that this is the issue) , then the first cell in parallel with the dummy will not have enough current to move to toggle 1.

I am not saying this is the issue, it is quite likely it is not, but have you checked it? is it valid ?
D.