Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!orion.oac.uci.edu!nntpsrv
From: tdutton@vmsa.oac.uci.edu
Newsgroups: sci.electronics
Subject: LM628 motion control
Message-ID: <26D571E5.5494@orion.oac.uci.edu>
Date: 24 Aug 90 18:28:53 GMT
Organization: University of California, Irvine
Lines: 29

I'm trying to build a little motion control box based on National's LM628
NMOS motor control chip. I'm using an 8-bit IO port from a device based on
primarily 74HC series CMOS to interface with the chip. Another 8-bit port
on the LM628 interfaces with a D/A converter; that to an  op-amp and the
op-amp to the motor.
My problem is that I seem to be able to write both commands and data to the
LM628 and I can read the status byte from the chip, but I cannot properly
read other data from the chip.
I have a 74HCT245 transceiver for compatibility (I think) of the NMOS '628
with the 74HC controlling device.
Example of what goes wrong: I write a command asking for 4 bytes of data
and then I try to read the bytes one at a time. After I've read two bytes
and ask for a third, the LM628 gives an interrupt condition saying that
I wasn't supposed to ask for anymore data. Similarly, for a command asking
for 2 bytes, I get an error when asking for the second of the two bytes.
I believe that my commands (writes) work because commands intended to alter
the status byte of the LM628 do just that...I read the status byte and
get what I expected.
The data sheets for the chip say that a read strobe must be asserted to get
from the read pin (I haven't seen any on an oscilloscope), I am at a loss
as to what unseen forces are at work.
Anyone else experience similar problems with NMOS LSI?
anyone?
anyone?

necessary disclaimer: I'm a physics graduate student, not an EE so stupidity
                      and ignorance in these affairs may be the key.

Thanks in advance.