A viable solution[Was]Re: MM5837?

[Magnus Danielson]

From: patchell <patchell at silcom.com>
Subject: A viable solution[Was]Re: MM5837?
Date: Sun, 03 Sep 2000 13:31:59 -0700

>     Here I am listening to my chugga chugga (I found it, didn't even have to do so by
> inspection, I just listened to it, it is a whole lot worse than I remember), and a good
> solution to this problem is staring me right in the face.
> 
>     If you would all please turn to page 147 in the good book (Digikey-July-Sept 2000),
> you will notice that the XC9536-15PC44C is only $3.30 in unit quantities.  It would be
> entirely posible to put a 32 bit shift reg and XOR feedback into one of these little
> suckers.  I just have to dig up the article I have stashed somewhere that shows which
> cells to tap.  About a half hour at Xilinx Foundation, and I should have a hex file.
> (Although, I will not be back at work until this Wedensday).  I would post the hex file
> and schematic (and encourage others to post it on their site).
> 
>     Anybody interested?

Good idea. You should be able to get a 36 tap easy, just tap at 25 and you are
home free. The XC9536 contains 36 macro-blocks and each of those contains an
DFF.

The good point about a loooooong sequence is that you can afford a higher
clockrate without the risc of frequent looping. Also, as it has been pointed
out before, you get a denser frequency spectrum so single frequencies does not
stick out in narrow filters.

Cheers,
Magnus