Re: TI-H: Re: UPSI: Need PCF8574 Source...
[Prev][Next][Index][Thread]
Re: TI-H: Re: UPSI: Need PCF8574 Source...
Richard Piotter wrote:
>>Hmm, what's this? A couple of PCF8574's in a cartridge?
>>That could be used as a Centronics printer port, with a driver...
>
> Where do I get them???
I don't know. I live in Finland.
> Can four be connected at the same time???
Yes. You select the I2C address of a PCF8574 with 3 pins, so you can have
up to eight PCF8574's on the bus. With PCF8574A that is otherwise identical
but has a different base address, this can be increased to sixteen.
> How fast would it take to update the data or recieve the data in all
> four chips???
It depends entirely on the software routines involved. With my current
MultiMaster I2C routines, you could update one chip with about 2.2
KBytes/sec (it's going to increase still) if (when) you only need to send
the address byte once. I haven't tested them with multiple chips, but
currently it might be something like 2*50 bytes/sec, because I've extended
the START condition with two HALT's so receiving TI-85's can keep up with
the speed. Those HALT's can safely be commented out if you only intend to
connect I2C chips. In the near future I will add #define's to customize the
routines to whatever purpose they're going to be used for in a program before
compiling. When that is done, I think you can get speeds of about
2*1Kbyte/sec with two chips, 4*500 bytes/sec with four chips etc.
These are, however, just very rough estimates of the actual speeds.
-Ozone
--
*** Osma Suominen *** ozone@clinet.fi *** http://www.clinet.fi/~ozone ***
References: