Problems with Winbond W29C040

[Asmusr]

Is there any reason a Winbond W29C040 shouldn't work with a red/512K board? It's programmed and verified OK, and on the TI the cart appears in the menu, but when I select an entry the TI crashes. The same board works fine with the same image on an AT27C040.

[Stuart]

Comparing the AT27C040 and W29C040, the pinout is slightly different (look at A18, /WE, VPP).

[+Ksarul]

Switch JP8 from 1-3 to 3-4 and see what happens. You won't have the WE signal connected, but you will have A18 on the correct pin at that point, so it won't be interfering with WE anymore. . .note this is not one of the normal board settings, but the use of NVSRAM chips requires the same move of A18, so it is already available to us in the design. When J8 is on 1-3, A18 is on Pin31. When you move it to 3-4, A18 is on Pin 1.

 

To force WE high all the time (probably a very good idea): if you look right next to Pin 3 of JP8 there is a large via with a short trace leading to Pin 16 of the 74LS378. If you solder a flying wire to that via and put a jumper socket onto it, you can connect that to Pin 3 of JP8 and you've now forced WE permanently high.

 

Your W29C040 should work perfectly at that point.

Edited July 17, 2016 by Ksarul

[Asmusr]

Switch JP8 from 1-3 to 3-4 and see what happens. You won't have the WE signal connected, but you will have A18 on the correct pin at that point, so it won't be interfering with WE anymore. . .note this is not one of the normal board settings, but the use of NVSRAM chips requires the same move of A18, so it is already available to us in the design. When J8 is on 1-3, A18 is on Pin31. When you move it to 3-4, A18 is on Pin 1.

 

To force WE high all the time (probably a very good idea): if you look right next to Pin 3 of JP8 there is a large via with a short trace leading to Pin 16 of the 74LS378. If you solder a flying wire to that via and put a jumper socket onto it, you can connect that to Pin 3 of JP8 and you've now forced WE permanently high.

 

Your W29C040 should work perfectly at that point.

 

Thanks, it works fine after only moving the jumper.

[+Ksarul]

The timing on this question is actually perfect, as I was getting ready to do another run of these boards. The necessary changes to use the W29C040 (the change was to force Pin 31 to +5V) in a mode that actively blocks writing to it are now integrated into the board layout. I'll send you a pair of them for your entertainment once I get the boards in, Rasmus. The jumper to prevent accidental writes to the chip isn't strictly necessary, but it does ensure that it won't happen accidentally.

[+OLD CS1]

THANK YOU for this thread. Rasmus, I took your advice and picked up ten of these Winbonds. So nice, indeed. I had t he same problems with the red board and the solution was right here.

 

Am I to understand, then, that this Winbond W29C040 is not a perfect drop-in replacement for 512k ROMs? I was hoping to use these also for some updated Amiga Kickstart ROMs.