Date   : Sat, 29 Oct 2005 14:31:59 +0100
From   : Jules Richardson <julesrichardsonuk@...>
Subject: Re: (BBC-Micro) Possible EPROM???

A. J. Davis wrote:
> Peter Craven wrote:
> 
>> Hello
>>
>> Can a the following EPROM be used in a Master
>> Computer:
>>
>> The 27C512-90 is a 64k x 8 bits 90ns CMOS EPROM with a
>> transparent glass window allowing the device to be
>> erased and written over multiple of times by the
>> programmer.
>> Features:    
>>
>>    * 5V +/-10% Supply Voltage in Read Operation
>>    * 90ns fast access time
>>    * Low power consumption: Active Current 30mA;
>> Standby Current 100=B5A
>>    * Programming voltage: 12.75V +/-0.25V
>>    * Programming time: 6 seconds (Presto IIB
>> Algorithm)
>>    * Electronic signature
>>    * Ideally suited for microprocessor systems
>>    * Organisation: 65,536 x 8 bits
>>    * UV Erasable
>>
>> Cat No.     Z9359
>> Category:     ICs - EPROMS
>> Support:     27C512-90 64k x 8 90nS EPROM (Z9359)
>>
>>
>> Dr Peter Craven (MBBS Uni. of Qld)
>>
>>  
>>
> I have some AMD 27C512-120 which work ok.

I certainly expect it's electrically compatible...

> The only thing to remember is to put your ROM file at the 'end' of the 
> EPROM.  So, if you have a 16k ROM image the first 48k must be blank.  
> This is what I found when putting a 16K Wordwise image onto a 32k chip 
> anyway.

Hmm, I guess that makes sense - the Master *probably* happens to tie the 
  top few address lines high

Differences between the typical 272128 (16K x 8) and the 27512:

pin  27128  27512
  1    vpp    A15
22    -oe    -oe / vpp
27    -pgm   A14

... so for normal operation -oe (pin 22) will be the same between both 
devices (it just has a dual function on the 27512 for programming 
conditions).

pin 27 - A14 on the 27512 - will always be held high by the Master 
because it expects that to be the pin which is active low when 
programming and high during normal operation.

pin 1 is the strange one, beause I'm surprised the Master doesn't tie 
vpp to 0V for the expected 27128 device (and so for a 27512 A15 would be 
low and Master would end up addressing the chip in the second quarter of 
the chip's address space, not the last quarter). Maybe the EPROM specs 
say that vpp should be set to the programming voltage during programming 
and 5V otherwise, then things would work. Or maybe it's a jumper change 
on the Master board?

cheers

Jules