Does it matter if I mix up the address lines of an EPROM?

N

Nemo

Guest
Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
--
Paul Honigmann
 
Nemo wrote:

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?
Click to expand...
As long as the address (and IO bit order) is the same programmed, as it is
connected to the processor (or whatever) it's plugged into, it will work.
We've done the same on one of our boards with using static ram. It still
works (as the RAM doesn't care where you store what, as long as you know where
it is when you want it again). We had to do it, as routing traces according to
'standard' address naming conventions would have proved difficult at the time.

A somewhat related mixup is used as a simple form of copy protection. Prior
to programming, you mix up the address and/or IO lines (via software on the
eprom binary, or via an IC header), and do the 'reverse' on the PCB.
Removing the eprom and attempting to dissasemble will prove fruitless unless
they 'unjumble' the lines again. This is easy to do, as they just have to
trace the PCB tracks to see where they go. It'll only really buys you time, as
those who are determined will do it anyway. It does however stop Idiots Inc.

--
Linux Registered User # 302622 <http://counter.li.org>
 
Probably. Test it first.

"Nemo" <Paul@nospam.nospam.nospam.nospam.co.uk> wrote in message
news:HXHBEbD6BUqAFwLA@furfur.demon.co.uk...
Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
--
Paul Honigmann
Click to expand...
 
I have done that with byte wide RAMs with no problem. If A0 is really one of
the swapped pins, and this is other than an 8 bit RAM it probably won't
work. If this is a FLASH ROM,don't even think about it..

Tam
"Nemo" <Paul@nospam.nospam.nospam.nospam.co.uk> wrote in message
news:HXHBEbD6BUqAFwLA@furfur.demon.co.uk...
Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
--
Paul Honigmann
Click to expand...
 
Only if all used address locations are palindromes :)

Thanks,
Kais


Nemo <Paul@nospam.nospam.nospam.nospam.co.uk> wrote in message news:<HXHBEbD6BUqAFwLA@furfur.demon.co.uk>...
Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
Click to expand...
 
Nemo <Paul@nospam.nospam.nospam.nospam.co.uk> wrote in message news:<HXHBEbD6BUqAFwLA@furfur.demon.co.uk>...
Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
Click to expand...
I had this question 18 yrs ago at ABC when a Motorola part had
different pin names from a 27xx eprom. We tried it and it worked. Are
there any setup/hold time differences between the alternate and
original? THAT work be a problem. As far as the address names go, the
data doesn't know the difference. Just be sure to program it to the
original part addressing so its compatible with the rest of the
system.
GG
 
On Mon, 17 May 2004 23:45:14 +0100, Nemo wrote:

Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
Click to expand...
The only time I've had a problem doing this sort of thing is with dual port
or multiplexed address busses where two or more devices selectively acces
memory. In that case the address busses must be in identical order.
Otherwise no problemo.

Bob
--
"Just machines that make big decisions
programmed by fellas with compassion and vision."
-D. Fagen
(remove yomama)
 
On a sunny day (Mon, 17 May 2004 23:45:14 +0100) it happened Nemo
<Paul@nospam.nospam.nospam.nospam.co.uk> wrote in
<HXHBEbD6BUqAFwLA@furfur.demon.co.uk>:

Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
Perhaps a drawing error in the data sheet.
Click to expand...
In that case it should make no difference of cause.
But in general crossing address lines makes a difference for EPROM, IF
you program these in an external programmer.
If on-board programming, it makes no difference.
(Like in some old 8052 setup I have).
The point is the MC should read the right code from the right address.
For example if you swapped A3 A4 in the programmer, and also on the board
from the MC, then it makes no difference.
Still think it is a drawing error, but why not request a sample, and try!
This is the only way to be 100% sure.
JP
 
On Tue, 18 May 2004 15:57:00 GMT, the renowned Jan Panteltje
<pNaonStpealmtje@yahoo.com> wrote:

On a sunny day (Mon, 17 May 2004 23:45:14 +0100) it happened Nemo
Paul@nospam.nospam.nospam.nospam.co.uk> wrote in
HXHBEbD6BUqAFwLA@furfur.demon.co.uk>:

Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
Perhaps a drawing error in the data sheet.
In that case it should make no difference of cause.
But in general crossing address lines makes a difference for EPROM, IF
you program these in an external programmer.
If on-board programming, it makes no difference.
(Like in some old 8052 setup I have).
The point is the MC should read the right code from the right address.
For example if you swapped A3 A4 in the programmer, and also on the board
from the MC, then it makes no difference.
Still think it is a drawing error, but why not request a sample, and try!
This is the only way to be 100% sure.
JP
Click to expand...
The numbers the chip manufacturer puts on the address pins are
irrelevant in any case, so long as the external programmer agrees with
the target board in terms of address line numbering.

Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
 
it's like pins 1-20 are all address pins,
but where one chip is A0 to A19, the other is A19 to A0.
Can I drop it in as a second source anyhow?
Nemo

Only if all used address locations are palindromes :)
Kais Badami
Click to expand...
:cool:
A peg at lovely Tsar, a style voltage, pa
 
"Spehro Pefhany" <speffSNIP@interlogDOTyou.knowwhat> schreef in bericht
news:1jkka0d9re67uef6bqf4oqafdq8ngn73jh@4ax.com...
On Tue, 18 May 2004 15:57:00 GMT, the renowned Jan Panteltje
pNaonStpealmtje@yahoo.com> wrote:

On a sunny day (Mon, 17 May 2004 23:45:14 +0100) it happened Nemo
Paul@nospam.nospam.nospam.nospam.co.uk> wrote in
HXHBEbD6BUqAFwLA@furfur.demon.co.uk>:

Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
Perhaps a drawing error in the data sheet.
In that case it should make no difference of cause.
But in general crossing address lines makes a difference for EPROM, IF
you program these in an external programmer.
If on-board programming, it makes no difference.
(Like in some old 8052 setup I have).
The point is the MC should read the right code from the right address.
For example if you swapped A3 A4 in the programmer, and also on the board
from the MC, then it makes no difference.
Still think it is a drawing error, but why not request a sample, and try!
This is the only way to be 100% sure.
JP

The numbers the chip manufacturer puts on the address pins are
irrelevant in any case, so long as the external programmer agrees with
the target board in terms of address line numbering.
Click to expand...
Strictly speaking you should use a driver for your programmer, made for
that particular eprom, following the manufacturers' datasheet, not some
generic driver that follows the industry standard ;)
 
On Wed, 19 May 2004 09:42:38 +0200, the renowned "Frank Bemelman"
<f.bemelmanx@planet.invalid.nl> wrote:

"Spehro Pefhany" <speffSNIP@interlogDOTyou.knowwhat> schreef in bericht
news:1jkka0d9re67uef6bqf4oqafdq8ngn73jh@4ax.com...
On Tue, 18 May 2004 15:57:00 GMT, the renowned Jan Panteltje
pNaonStpealmtje@yahoo.com> wrote:

On a sunny day (Mon, 17 May 2004 23:45:14 +0100) it happened Nemo
Paul@nospam.nospam.nospam.nospam.co.uk> wrote in
HXHBEbD6BUqAFwLA@furfur.demon.co.uk>:

Hi,

I have been asked by our Purchasing department if they can use an
alternative memory part. Looks the same except the address lines are
numbered differently - to simplify, it's like pins 1-20 are all address
pins, but where one chip is A0 to A19, the other is A19 to A0. Can I
drop it in as a second source anyhow?

Thank you,
Perhaps a drawing error in the data sheet.
In that case it should make no difference of cause.
But in general crossing address lines makes a difference for EPROM, IF
you program these in an external programmer.
If on-board programming, it makes no difference.
(Like in some old 8052 setup I have).
The point is the MC should read the right code from the right address.
For example if you swapped A3 A4 in the programmer, and also on the board
from the MC, then it makes no difference.
Still think it is a drawing error, but why not request a sample, and try!
This is the only way to be 100% sure.
JP

The numbers the chip manufacturer puts on the address pins are
irrelevant in any case, so long as the external programmer agrees with
the target board in terms of address line numbering.

Strictly speaking you should use a driver for your programmer, made for
that particular eprom, following the manufacturers' datasheet, not some
generic driver that follows the industry standard ;)
Click to expand...
Assuming the author of the driver would actually follow that
particular silly data sheet to the letter. ;-)

Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top