aliman wrote:
thx, this is exactly what i needed^^
but is it possible to get a picture of your schematic that is more clearly because it is hard to read some numbers and parts.
and is it possible to use 62256 without modification of the program or the schematic?
That's all I got... Take a look at the original schematic of the EEprog I linked to. I'll strongly caution again, this is no where close to polished... I'd like it to be as I want to make my own programmer with it for all the memories I'd like to use, but it's a low priority. Things will NOT work as I've posted, you'd have to work on some things for yourself really.
As for components the D- (white) pull up resistor should really be 2.2Kohm(even though it says 1.5K), the series usb resistors are 68ohm, and the zeners are 3.6v 500mA or less diodes. Any large power cap should do. I used a 16Mhz xtal but you could also go with 12Mhz. There should also be ~20pF caps on the two xtal pins going to ground. The flipflop is a 54HC374 8x flipflop. The design in the schematic supports up to 16KB, but I modded the schematic from what you see here and put A5 on the flipflop. That's how the code is written, as is, it ONLY works with 8KB since that's all I cared about for this proof of concept.
If you reaproriated the LED pin as A6 and shifted all the Flipflop addresses by one you could do 32KB for your chip. I only had one pin for /OE and /CE in the schematic, just hardwire is as you like was my thought. You could even ditch the /WE pin and just hardwire the SRAM's pin to gnd or put a switch in there. If you did all that you could do 64KB. To go beyond that without disabling the reset on the AVR you'd have to add a flip flop or just program one page at a time or something. Just adding another set of FFs would add support for 32Mb (4MB).
Keep in mind you not only have to fix the schematic and mcu code but also the host application because the pages sizes aren't setup for this stuff.