home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!dtix!darwin.sura.net!mips!sdd.hp.com!usc!snorkelwacker.mit.edu!news.media.mit.edu!fredm
- From: fredm@media.mit.edu (Fred G Martin)
- Newsgroups: comp.robotics
- Subject: MiniBoard 2.0 serial line circuit
- Message-ID: <1992Aug17.183502.7155@news.media.mit.edu>
- Date: 17 Aug 92 18:35:02 GMT
- Sender: news@news.media.mit.edu (USENET News System)
- Organization: MIT Media Laboratory
- Lines: 61
-
- A few people have asked how the line level converter circuit works on
- the Mini Board. I thought this might be important to discuss, since
- the circuit is a bit obscure and may be incompatible with a few
- existing computers.
-
- Serial line voltages are normally +12 volts for logic low and -12
- volts for logic high. This is annoying because TTL/CMOS logic levels
- are of course +5 volts for high and 0 volts for low.
-
- In the past, people normally used a bipolar power supply to generate
- the serial line levels; nowadays most people would use a single chip
- solution like the MAX232 series of chips (which include an on-chip
- charge pump and voltage inverter to generate +/- 10 volts from a 5
- volt supply). However, in a board as small as the Mini Board, even
- one extra chip is one chip too many.
-
- So, the Mini Board uses the following hack.
-
- For its serial line logic high, the Mini Board "loops back" the logic
- high (remember, this is a -12 volts) that the computer it is connected
- to generates! This is using a 1K resistor, which shouldn't cause much
- of a voltage drop.
-
- A PNP transistor pulls this line (Mini Board serial out) to +5 volts
- to generate the serial line logic low.
-
-
- --->>--MB serial in--->>--+ +5V
- | |
- HOST / \
- 1K \ >|
- COMPUTER / |-----/\/\----<<-- Mini Board
- \ /| TTL transmit
- | /
- ---<<--MB serial out--<<--+-----
-
-
- For input, the Mini Board uses a spare CMOS inverter coupled with two
- resistors.
-
- The upshot of this solution is (1) the Mini Board can't transmit while
- the host computer is transmitting, because the incoming -12v it needs
- would be modulated by the host transmission; and (2) there might be
- some computers that don't accept +5v as serial line low.
-
- Regarding compatibility: I have personally tested the Mini Board on a
- variety of IBM and clone PCs, a variety of Macintosh models,
- DECstation 5000 computer. I know that there is at least one type of
- Sun SPARC that it works on.
-
- So it does work on most machines. Still, there is bound to be the
- model that it won't work on. A reader believes that it probably won't
- work on Sun 3 workstations. For these cases, it wouldn't be hard to
- jumper the board and wire a voltage higher than +5 to the collector of
- the transistor. This should allow the board to work with these other
- machines.
-
- - Fred
-
-
- Acknowledgment: this circuit was the brainchild of Randy Sargent.
-