MAX-II input b-scan cell capability lost with "Input tri-stated" unused pin setting

Question

Hello All,

We used EPM2210F256C4 in a design which can not tolerate default internal pull-ups in a boundary-scan test.

A dummy design was made with QII 12.0 web-ed.

One pin used only and in Device and Pin options unused pins was set to "As input tri-stated" to get rid of the pull-up.

After programming the device successfully (via SVF file) this can be confirmed with a scope - no pull-ups anymore.

The downside is that it looks like all input cells are measuring 1's. Output capability remains and can be seen working fine.

I have used EPM570 before with earlier versions of QII and I don't recall anything like this - it has been working fine if my memory serves right.

Anyone seen this before? All help appreciated.

altera_forum · Answer

Answering to my own post.

It appears that even in this situation one has to enable "Always enable input buffers" as described in the document

IEEE 1149.1 (JTAG) Boundary-Scan Testing for MAX II Devices

It is said that this is necessary only if there are dedicated outputs, but apparently this is needed in this situation as well.

Procedure in the a.m. document:

1. On the Assignments menu, click Settings.

2. Under Category, select Assembler.

3. Turn on Always Enable Input Buffers.

Forum Discussion