Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
02/20/08 08:38
Read: times


 
#151193 - lots of i/o part 2
Responding to: ???'s previous message
Goodmorning forum!

And thanx for the replies. I have a few comments:

16 bit expanders from nxp or texas might be the way to go. Proven, simple, cheap, and with some ESD protection on chip

Jan absolutly has a point: The µc solution IS more sexy (or fancy) And it might end up as a white elephant; it is more complicated than a handfuld of i2c port expanders.

Erik has a point about choosing a processor with enough pins. But even SiLabs stops at 64 pins: That will give me my 32 inputs and 32 outputs, but I will need at least 2 pins for Tx and Rx, and some additional pins for a lc display.
BTW Erik - the '535 is an Siemens/Infeneon device, obsolete and with external memory. But an okay device from years ago.

This could of course be the aplication for picking a non 8051 - an ARM7 from nxp. They come with almost the pincount you want.
There is however an learningcurve............
Ups - non 8051 µc on 8052.com. I'll go and wash my mouth (med brun sæbe)!

As I said in my original post, I dont want to borther with parallel busses; to much trouble, to much pcb real estate. And if i shall use programming time, i'll prefer the µc solution for the CPLD.

The MAX 7300/7301 look pretty nice, but lack the ESD protection the nxp/ti devices have.

I'll guess i'll have to do some more headscratching.
I'll post my decision, when I've made it.
In the mean time any comments still apreachiated!

Ha' en (fortsat) go' dag

Per

List of 16 messages in thread
TopicAuthorDate
lots of i/o            01/01/70 00:00      
   There are 16-bit Expanders            01/01/70 00:00      
      The + and - of the slave processor            01/01/70 00:00      
         the logical solution is to ...            01/01/70 00:00      
   Roll 'yer own, and get exactly what you want            01/01/70 00:00      
      why bother            01/01/70 00:00      
         you're right, if "conventional" I/O ports are OK            01/01/70 00:00      
   Max 7301            01/01/70 00:00      
   lots of i/o part 2            01/01/70 00:00      
      Theres a 40way NXP I/O expander            01/01/70 00:00      
      Simple Distributed Concept            01/01/70 00:00      
         very valid point            01/01/70 00:00      
            For sure.....the multi-port job....            01/01/70 00:00      
               lots of i/o part 3            01/01/70 00:00      
                  I love slave processors            01/01/70 00:00      
                  MCP23008            01/01/70 00:00      

Back to Subject List