Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!zaphod.mps.ohio-state.edu!unix.cis.pitt.edu!fmgst
From: fmgst@unix.cis.pitt.edu (Filip M Gieszczykiewicz)
Newsgroups: sci.electronics
Subject: CMOS cpu info needed
Message-ID: <33469@unix.cis.pitt.edu>
Date: 30 Aug 90 04:58:59 GMT
Reply-To: fmgst@unix.cis.pitt.edu (Filip Gieszczykiewicz)
Followup-To: sci.electronics
Distribution: usa
Organization: The Last Jedi
Lines: 66


	Greetings. I am in the process of planning a micro-based
	project. The requiraments are:

	1) Runs for min. 8 days on a 9V battery (maybe 2 in parallel)
	2) Samples 8 analog channels every 1<time<10 minutes
	   and stores in RAM.
	3) includes a serial port for downloading to a host.
	4) inexpen$ive...

	I have several approaches:

	1) timer, address gen., latch, USART, ADC, RAM 
	ie. no CPU. It's cheap but hard to get working.
	The amount of discrete components (total power
	consumption) is way too high.

	2) 4bit micro. based. I am thinking of using a
	NEC PD7533 4bit CMOS CPU. It has:
	o 67 instuctions
	o 8-bit ADC
	o 8 high current lines (LED connection)
	o 2 4-bit input ports
	o 1 2-bit output port
	o 1 4-bit output port
	o 8-bit serial interface
	o STOP and HALT modes (real low power consumption)
	o 2 internal and 1 external interrupts.
	o operating feq. to 500kHz (wow!)
	It comes in a SM package and can access 65K RAM
	(uses something like 500 uA @ 500kHz!)

	3) 16-bit micro. base. Another setup I considered would
	use the NEC PD7810 (ROMless). It has:
	o 44 I/O lines
	o 2 8-bit timers
	o 8085A bus compatible (60K address space)
	o 8 channel, 8-bit ADC
	o Full duplex USART
	o 153 instructions
	o prioritized interrupts (3 external, 8 internal)
	o STANDBY mode	
	o on-chip clock generator
	(A bit of an overkill, if you ask me, but it will
	handle everything I need.) Uses 30mA of power!

	4) Do you have any other ideas? Using other chips?

	I'm still in the process of choosing the best approach.
	The discrete one (no CPU) fails unless I can get SM
	8 channel ADC and associated logic (in CMOS)

	(SM = surface mount)

	Since most of the above micros handle only 60K address
	spaces, I am forced to look into "paged" RAM. Does anyone
	know of a chip (or approach) that will handle this?

	Take care.

	P.S. I will summarize and make available anything I get. 
-- 
_______________________________________________________________________________
"The Force will be with you, always." It _is_ with me and has been for 10 years
Filip Gieszczykiewicz    "A man without a dream is like a fish without water."
FMGST@PITTVMS  or  fmgst@unix.cis.pitt.edu "My ideas. ALL MINE!!"