ole00
commented
4 years ago Hi,
reading the fuse map seems to be the same that is in the jed file only the last 6 lines of every single one is kinda diferent,
That should be OK. The last number in the original JED file is an 'end of transmission' character (0x03) followed by transmission checksum, which afterburner does not calculate. But that should not affect functionality. Some software require transmission checksum to be present in the JDEC file, in such case using value 0000 will do the job (it means ignore transmission checksum).
As far as I can tell, if you are able to read back all fuses from the IC and the contents match then the device is correctly programmed. My understanding is that pull down resistors ensure that the pins don't toggle during programming (either randomly or as a consequence of the previous logic programmed in the device) for devices that are sensitive to such changes. It looks like Atmel devices do not require pull downs (see Bruce Abbott's notes linked on the front page), so using 10k or 1k pull-downs should probably be OK in your case (using Atmel IC).
also I have to config WinCupl in g16v8ma mode if not the chip almost does nothing
In Wincupl sepcify:
Device G16V8;
in your PLD design file.
And in Menu->Option->Devices specify DIP | Atmel | ATF16V8B. That's how it works for me OK.
I'm not quite sure what you mean by 'the chip almost does nothing'. Can you exercise AND | OR | XOR | NOT combinations on several inputs and feed the output to several output pins? Does that work for you? I think that's pretty much all you can do with these 16V8 devices. You can also synchronize the output via register based on input clock pin 1 (only on that one) like this:
Device G16V8;
/** Inputs **/
PIN 1 = clock; /* Register Clock */
PIN [2,3,4] = [D0..2]; /* Data Input */
/** Outputs **/
PIN [14, 15, 16] = [Q0..2]; /* Register Output */
/** Logic Equations **/
[Q0..2].d = [D0..2]; /* note .d is used */ Also after some research I found that it was something kinda common for the Atmel PLDs, only GALs from lattice were working (with universal programmers as the TL866).
I don't know about that (might be true though for certain batches or revisions of the chip). Could you link some sources that suggest that?
Just a silly question: do you test your programmed Atmel chip outside of the afterburner programmer, right? So, do you take it out of afterburner and place it in a bread board, or place it to your testing protoboard? Do you have decoupling capacitors installed on VCC when testing on your test board/pcb? I'm sure you do that, I'm just ensuring there is no misunderstanding what afterburner can do and can not do.
If you have a .PLD script you would like me to try, feel free to post it here and I can test it with ATF and Lattice chip. Just make sure the script is compatible with G16V8 devices.
Sorry if my english is bad.
Your English is very good !
admo64
mecparts
Don't know why but the chip just doesnt work as supposed, only simple equations works (for example when only one gate type is used for each output), 3.6V outputs instead of 5V, only the low signals are really 0V, also I have to config WinCupl in g16v8ma mode if not the chip almost does nothing, even if everything in the process of programing this ICs seems to be ok, reading the fuse map seems to be the same that is in the jed file only the last 6 lines of every single one is kinda diferent, example:
JEDEC file: L02048 01000000001100000011000000100000 L02112 00000000111111111111111111111111 L02144 11111111111111111111111111111111 L02176 111111111111111111 C14B5 9091
Read from the chip: L2048 01000000 N UES 0C 0C 04 00 00 00 00 00 L2056 0011000000110000001000000000000000000000000000000000000000000000 L2120 11111111111111111111111111111111111111111111111111111111111111111111111111 N PES 31 42 38 56 36 31 46 31* C14B5
I follow the instructions to program these, 10V, erase first, also tried the conditioning jed file from ATFblast for the ATF16V8B but in this case afterburner returns an error saying that the fuse map wasn't uploaded. Must say that in the moment I ran out of 4.7K resistor, so I used 10k and also tried 1K to see if there was any diference but it was the same, tried the x64 exe in windows 10, the linux executable in debian running under virtual box, even compiling with WinCupl in a windows 98 VM, all were the same results.
Also after some research I found that it was something kinda common for the Atmel PLDs, only GALs from lattice were working (with universal programmers as the TL866).
Did you have the same issue?
Sorry if my english is bad. Thanks in advance