| View previous topic :: View next topic |
| Author |
Message |
gg
Guest
|
| Switching over to 16F628A and 16F877A ( is it OK) |
 Posted: Mon Jun 27, 2005 8:29 am |
 |
|
Hi,
1 am thinking of switching over to 16F628A from 16F628 , are there any problems in doing this,
I know I should have done this long time ago...
I am also thinking of changing to 16F877A from 16F877.
Are there any differences between these mcus ( the A and the one without A). Also do the pins match.
thanks
gg |
|
 |
valemike
Guest
|
|
| Posted: Mon Jun 27, 2005 9:22 am |
|
|
As long as they are pin-to-pin compatible, yes they are migrateable. And as long as the next micro up has the same peripherals on the same pins that you would need, you'll be okay.
Only one way to find out if it works is to change the
#include <18XXXX.h> CCS header file, compile, and try it.
If all else fails, then read the App notes that deal with migrating to higher micros.
-Mike |
|
|
treitmey
Joined: 23 Jan 2004
Posts: 1094
Location: Appleton,WI USA
|
|
| Posted: Mon Jun 27, 2005 10:47 am |
|
|
YES, there are differences in 16F877 and 16F877A. Thats why they changed the number. Not big differences. It may not be a drop in substitution.
Can you change firmware? or are you stuck reusing old firmware?
Also take a look at the migration doc.
http://ww1.microchip.com/downloads/en/DeviceDoc/39591a.pdf
This should answer some of your questions. |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Mon Jun 27, 2005 11:53 am |
|
|
Always check the errata sheets. Here's the one for the 16F628A.
http://ww1.microchip.com/downloads/en/DeviceDoc/80151L.pdf
Notice on the 2nd page at the top left, it says:
| Quote: | Unexpected program execution may occur during data eeprom
write cycles. |
| Quote: | Note: This problem is corrected in PIC16F648A
Rev. A5 and PIC16F627A/628A in Rev. A8. |
Ummm.... Rev. A8
They have gone through 8 revisions of the silicon so far,
on the 16F628A.
They use a different silicon process for their "A" chips. It's much
harder for their chip designers to make the chips work properly.
It's clear that Microchip management insists that the company use
the "A" silicon process as much as possible. The "A" parts are much
cheaper. Here are the prices from http://www/digikey.com
| Quote: |
PIC16F628A-I/SO $1.65 (US) in 100 quantity.
PIC16F628-20I/SO $2.64 (US) in 100 quantity. |
But do I as a designer really want to use a chip that is now in
its 8th silicon rev ? For me, the answer is no. |
|
|
Guest
|
|
| Posted: Mon Feb 06, 2006 2:49 am |
|
|
| valemike wrote: | As long as they are pin-to-pin compatible, yes they are migrateable. And as long as the next micro up has the same peripherals on the same pins that you would need, you'll be okay.
Only one way to find out if it works is to change the
#include <18XXXX.h> CCS header file, compile, and try it.
If all else fails, then read the App notes that deal with migrating to higher micros.
-Mike |
|
|
|
Eugeneo
Joined: 30 Aug 2005
Posts: 155
Location: Calgary, AB
|
|
| Posted: Mon Feb 06, 2006 4:02 am |
|
|
I've changed one of my designs from 877 to 877A
A couple of things I had to change to keep the reliability. But remember I did not do extensive research on these fixes and should be used with discretion.
Make sure you have a couple 0.1uf caps close to the chip
The pins on port B and D seem to have a slightly less capacitance so noise became an issue where it was never a problem before.
If you have a bootloader you must use the new A part.
The comparator should be disabled even though the data sheet says otherwise.
But on the plus side, they are cheaper and program faster.
Best of luck |
|
|
rnielsen
Joined: 23 Sep 2003
Posts: 852
Location: Utah
|
|
| Posted: Mon Feb 06, 2006 9:37 am |
|
|
| The original post of this thread is several months old. It's just been hit by the phantom guest poster. |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
