27c512 pinouts
Gregory John Craske
I can't for the life of me (I don't write these, I just repeat them) get
hold of a databook, and don't give me no order numbers or publishers coz
I tend to ignore them (I bet I'm not alone there).
I'm talking bout the 27c512 OR 27512 64k eprom. (Burn voltage and burn time
would also be handy).
Thanks for reading up to here.
Greg Craske - s940...@arcadia.cs.rmit.edu.au
blah blah blah. .sig {some thought stimulating comment by someone no one has
heard of}.
Paul Richards
Patience!
27C512 PINOUT
A15 1 28 VCC
A12 2 27 A14
A7 3 26 A13
A6 4 25 A8
A5 5 24 A9
A4 6 23 A11
A3 7 22 /OE / VPP (ON 27C256 PIN 1 is VPP)
A2 8 21 A10
A1 9 20 /CE
A0 10 19 D7
D0 11 18 D6
D1 12 17 D5
D2 13 16 D4
GND 14 15 D3
EXAMPLE: M27C512-15XF1 SGS-THOMSON 12.5 VPP, 150NS ACCESS
50mA active, 1mA inactive.
Modern programming practice is to pulse burn repeatedly, up to, say
twenty five 1ms pulses until a verify returns the data you're writing.
Count the number of pulses. Once verified, give it a single long burn 3
times the length of the pulses so far given, limiting to, say 50ms. (Some
people say x4 or x5.) e.g. If it takes 7 times a 1ms pulse to get the
verify, follow it with a 21ms burn. 50ms is usually the upper limit.
If you are prototyping, you can program faster, but you risk the eprom
going floppy later on if you don't give it a decent burn.
Since the Vpp line is shared with /oe (on the 27c512), you will need to
switch this quickly to verify each byte. When its at VIL (0V) you can
read the device. When at VPPH (+12.5) you can program the device. In both
reading and writing, /CE goes low. After programming Vpp will normally be
at VCC=+5V or 0V when active.
The NEC data for a similar device says that VCC should be switched to +6V
during programming. Presumably the same for SGS-Thomson. They also say
program the first time with a 1ms pulse, then up to 25 x 1ms, followed by
up to 25 x 3ms pulses to burn securely. The initial program pulse is
described as a minimum of 0.95ms and a maximum of 1.05ms.
If this is your first go at programming eproms, you might just program
with a long delay and hope for the best, and try it on some 2nd hand
chips. i.e. dont verify until all bytes have been written. Perhaps with
a fixed delay of 25ms maximum per byte. Maybe it will destroy the IC, but
if it doesn't it will save you some programming effort.
Have fun.
Paul Richards
richard steven walz
In article <D4sCx...@cix.compulink.co.uk> p...@cix.compulink.co.uk ("Paul Richards") writes:
>> Just a fading hope that I'd be able to get the pinouts by some time
>this
>> year.
>> I can't for the life of me (I don't write these, I just repeat them) get
>> hold of a databook, and don't give me no order numbers or publishers coz
>> I tend to ignore them (I bet I'm not alone there).
>> I'm talking bout the 27c512 OR 27512 64k eprom. (Burn voltage and
>burn
>> time
>> would also be handy).
>
>during programming. Presumably the same for SGS-Thomson. They also say
>program the first time with a 1ms pulse, then up to 25 x 1ms, followed by
>up to 25 x 3ms pulses to burn securely. The initial program pulse is
>described as a minimum of 0.95ms and a maximum of 1.05ms.
>
>If this is your first go at programming eproms, you might just program
>with a long delay and hope for the best, and try it on some 2nd hand
>chips. i.e. dont verify until all bytes have been written. Perhaps with
>a fixed delay of 25ms maximum per byte. Maybe it will destroy the IC, but
>if it doesn't it will save you some programming effort.
------------------------------------
Sorry, Paul, that isn't good advice, this last bit. It simply doesn't work!
All the cheap 50mS programmers out there started NOT WORKING on the 27Cxxx
parts as soon as they modified their manufacturing to require the higher
Vcc during programming and the higher fractional Vpp programming voltages
as well! It is VERY important that you observe the recommendations of the
manufacturer, as these are all made from several different parameter
designs now and none of the CMOS 27xxx's will work right with that old
"pushbutton" length pulse of 50mS per byte. A lot of the JDR programmers
called "Sunshine" by MCT which didn't account for the new specs are now
only useful for the old EPROMs. I know as I have one! I had to build my own
after bothering to determine the proper programming voltages and algorithms
and Vcc enhanced voltages for all the different manufacturers and parts and
I keep it on a LONG list now which is about 20KBs!! I set the Vcc and the
Vpp for each part after I have looked it up. Then I select between five
different programming algorithms, none of them TOO terribly different now,
but they are spec'd different, so I don't risk it! Build your own
programmers, it will drive you nuts but you'll learn A LOT!!!
-Steve Walz rst...@armory.com
P.S., they DO still make the 27xxx as opposed to 27Cxxx, but they use more
juice, and they take a heck of a lamp to erase them sometimes. I don't know
HOW hard they are hitting them in final test! And sometimes they take
multiple tries to suck a cell to zero after they were really UV'd!
-Steve USE THIS ADDRESS ONLY FOR EMAIL to me: rst...@armory.com