What's faster to get done, is building a simple logic analyzer circuit on an FPGA prototype board.

If you have an FPGA prototype board available, one on which you can add a bit of circuitry (you won't need much) and a connector or two, then you're on your way. Unfortunately, the ones on which I've previously addressed this need are no longer readily available. The "cool" thing about FPGA boards with prototype area is that they allow you to program your FPGA, which, generally speaking, means some internal RAM, which you cah use for a buffer, or in which you can store trigger equations, or both (believe me, the triggering is always the most complicated portion of a logic analyzer.)

If you were to use an oscilloscope, it would be much easier to display your sampled signals, but if appears, since you don't own one, that you should use the PC, which you do own, instead. This means writing your own display and data manipulation routines. Of course, you can do this under DOS, using the old Borland 'C' compiler, which supports a GOTOXY function, and which allows for line drawing.

The way in which you'd produce the oscilloscope-compatible display would be to route the signals to a single BNC connector, through a primitive D-to-A converter consisting of a resistor ladder, or using an external DAC. The four upper bits would be for identifying (steering) the trace on the display, and the lower four would be for ensuring separation between the traces. That's where you'd attach your 'scope.

If you want to use a PC, you simply route the sample data to the PC through the parallel port, and sort it out in the PC.

If you're a real glutton for punishment, you include a simple CRTC in your FPGA, and drive a monitor directly. That's an "exercise for the student," however.

The problem with the FPGA board is finding and getting one in time.

What you probably want is a XILINX XC2S300E (1.8-volt logic, but 3-volt tolerant with series resistors) in BGA256 or PQFP208, if it's available. That one has 64Kbits of internal block RAM available. The ones I've used for this sort of thing are boards with XC2S200 FPGA (5-volt-tolerant 2.5-volt logic) with 56Kbits of block RAM. I paid about $100 each for those boards.


We can explore this further if this hasn't scared you off.

You can design and implement whatever you like on this type of board, and, if you can build your own board in a reasonable time frame, you can use it for many things.

If you have the funds available, however, and if you plan to do any sort of hardware development in the immediate future, I'd recommend you consider expending a few dollars on eBay for what's claimed to be a fully functional oscilloscope, without which you'll be pretty helpless anyway. A pretty decent Tektronix 465 or 475 should cost between 100 and 250 dollars U.S. That's a lot, but, properly cared for, it will last a long time. My old '465, purchased used in 1979 lasted me through 25 years of hard use, until I gave it away to a friend. That's a very useable 100 MHz instrument. Now, you can do without one, but not very easily.

RE