[net.analog] Any 555 mavens or PUT weenies out there?

zben@umd5.UUCP (04/09/85)

The situation is that I want to build a bitbanger serial interface for my
kid brother's Apple ][+ that plugs into the game port, which has several
TTL inputs and outputs.  The problem is that the only power supply available
on the game paddle connector is +5 - and I'd rather not kluge in the Apple's
two 12 volt supplies on the unused pins or run more wires.  I need a nominal
amount of + and - 12 for the TTL-RS232 converters (MC1488 or MC1489) and for
some strapping for DTR type stuff.  Rather than build a separate plug into 
the wall supply or run on (ick) batteries, I thought I could design a small
DC-DC converter to run off the available +5 and produce these voltages.

I had thought about something like this circuit:

       R1
   +-/\/\/\-+---+---o +5
   |        |   |
   |    +---+---+---+            5-1        o +5
   +----+7  8   4   |        TRANSFORMER    |
   |    |          3+---------~~~~~~~~~-----+
   /    |           |         =========               +----+--/\/\/\--+--o +12
   \    |    555    |      +--~~~~+~~~~--+            |    |          |
R2 /    |           |      |      |      |   +-----+  | + ---       \---\
   \    |           |      |     ===     |   |     |  |   ---        / \ ZENER
   |    |           |      |      =      +---+ ~ + +--+    |         ---
   +----+2,6    1   |      |                 |     |       +----+-----+
   |    +-------+---+      +-----------------+ ~ - +--+    |   ===    |
  ---           |                            |     |  | + ---   =   \---\
  --- C1       ===                           +-----+  |   ---        / \ ZENER
   |            =                            BRIDGE   |    |         ---
  ===                                      RECTIFIER  +----+--/\/\/\--+--o -12
   =


with the 555 set up to run somewhere in the audio range, hopefully in the
efficient range of the transformer, which is an audio transformer.  My
questions are:

Can a 555 drive a transformer like this?  Is it better to run it to +5 like
in this diagram, or to ground?  Would it be better to use a tapped primary
and/or a drive to "halfway" provided by a voltage divider?  Would it be
better to set the 555 at 60 cycles and use a power transformer?  Would it be
better to use a Programmable Unijunction Transistor (PUT) in this application?
Would it be better for me to wind a torroidal transformer, and if so, about
how many turns do I need on the primary (I can compute secondary given this)?

Ad(thanks)vance for your thoughts on this matter.

-- 
Ben Cranston  ...{seismo!umcp-cs,ihnp4!rlgvax}!cvl!umd5!zben  zben@umd2.ARPA

karsh@geowhiz.UUCP (Bruce Karsh) (04/12/85)

> some strapping for DTR type stuff.  Rather than build a separate plug into 
> the wall supply or run on (ick) batteries, I thought I could design a small
> DC-DC converter to run off the available +5 and produce these voltages.

  How about using some C-MOS inverters and making a charge pump.  This does
not require the transformer, will be a lot smaller, and will probably be 
less expensive and have a smaller parts count.  

-- 
Bruce Karsh                           |
U. Wisc. Dept. Geology and Geophysics |
1215 W Dayton, Madison, WI 53706      | This space for rent.
(608) 262-1697                        |
{ihnp4,seismo}!uwvax!geowhiz!karsh    |

al@psivax.UUCP (Al Schwartz) (04/13/85)

In article <469@umd5.UUCP> zben@umd5.UUCP (Ben Cranston) writes:
>The situation is that I want to build a bitbanger serial interface for my
>kid brother's Apple ][+ that plugs into the game port, which has several
>TTL inputs and outputs.  The problem is that the only power supply available
>on the game paddle connector is +5 - and I'd rather not kluge in the Apple's
>two 12 volt supplies on the unused pins or run more wires.  I need a nominal
>amount of + and - 12 for the TTL-RS232 converters (MC1488 or MC1489) and for
>some strapping for DTR type stuff.  Rather than build a separate plug into 
>the wall supply or run on (ick) batteries, I thought I could design a small
>DC-DC converter to run off the available +5 and produce these voltages.
>Ben Cranston  ...{seismo!umcp-cs,ihnp4!rlgvax}!cvl!umd5!zben  zben@umd2.ARPA

You can run an RS232 with the voltages +5 and -5.  You already have the +5.
In order to make the -5 you need to get an Intersil 7660 (negative voltage
generator).  I have heard that this part is available at such places as
Radio Shack.  This chip and two capacitors is all that is required to generate
the -5 voltage.
-- 
                                      Al Schwartz
                                      Pacesetter Systems Inc., Sylmar, CA 
{trwrb|allegra|burdvax|cbosgd|hplabs|ihnp4|sdcsvax|aero|uscvax|ucla-cs|
 bmcg|sdccsu3|csun|orstcs|akgua|randvax}!sdcrdcf!psivax!al
or {seismo|citcsv|engvax|wlbr|zeus}!scgvaxd!psivax!al

jeffw@tekecs.UUCP (Jeff Winslow) (04/14/85)

I tried to mail this, and it got stopped somewhere, so I'll go this way...

Assuming you only need 10 mA or so, your circuit ought to work, except that
you will need a coupling capacitor between the 555 output and the transformer,
otherwise the DC component in the driving waveform will saturate the
transformer core. And why don't you try driving the xformer at 20kHz. It's
true it's probably not very efficient there (core losses will be higher and
leakage inductance will reduce output voltage somewhat), but at these power
levels, that won't matter much, and I would think it would be worth it not
to have to listen to the whine.

				Jeff Winslow

zben@umd5.UUCP (04/15/85)

Many thanks to all who responded, especially to those who pointed me to ICL
chips, and those who pointed out that the 1488 will work fine from +/-5 V.
I got the entire hardware and software for the project working this weekend
using the Apple +5 supply and my bench -5 supply.  watmath!jsgray suggested
the ICL7660CPA chip which eats +5 and puts out -5 and is available at Radio
Shack.  I am going there later tonight to try to snarf up one - and that
should just about complete the project...

-- 
Ben Cranston  ...{seismo!umcp-cs,ihnp4!rlgvax}!cvl!umd5!zben  zben@umd2.ARPA

cem@intelca.UUCP (Chuck McManis) (04/17/85)

> In article <469@umd5.UUCP> zben@umd5.UUCP (Ben Cranston) writes:
> >The situation is that I want to build a bitbanger serial interface for my
    ... Message about problem goes here ...
Followd by answer : 
> 
> You can run an RS232 with the voltages +5 and -5.  You already have the +5.
> In order to make the -5 you need to get an Intersil 7660 (negative voltage
> generator).  I have heard that this part is available at such places as
> Radio Shack.  This chip and two capacitors is all that is required to generate
> the -5 voltage.
> -- 

I tried this method once and found that the Intersil part couldn't supply
enough current to run the data line, much less the handshake lines over 
any length of cable. The voltage went from -4.8V to -2V and out of the
allowable +/- 3V specs for RS-232C. However, if you first use your handy
555 voltage doubling circuit and feed the 7660 with 9.9V which is inverted
to -9.7 volts which only degrades to -5.5 volts when running all of the 
handshake lines too. It works quite nicely.

--Chuck
-- 
                                            - - - D I S C L A I M E R - - - 
{ihnp4,fortune}!dual\                     All opinions expressed herein are my
        {qantel,idi}-> !intelca!cem       own and not those of my employer, my
 {ucbvax,hao}!hplabs/                     friends, or my avocado plant. :-}

ericksen@unc.UUCP (James P. Ericksen) (04/18/85)

>  The problem is that the only power supply available
>on the game paddle connector is +5 - and I'd rather not kluge in the Apple's
>two 12 volt supplies on the unused pins or run more wires.  I need a nominal
>amount of + and - 12 for the TTL-RS232 converters (MC1488 or MC1489) and for
>some strapping for DTR type stuff.

>> You can run an RS232 with the voltages +5 and -5.  You already have the +5.
>> In order to make the -5 you need to get an Intersil 7660 (negative voltage
>> generator).  I have heard that this part is available at such places as
>> Radio Shack.  This chip and two capacitors is all that is required to 
>> generate the -5 voltage.


Good idea.  One word of warning though ... the MC1488 RS232 driver will NOT work
with a +5 volt supply (it requires at least +9 volts, an undocumented fact i
learned the hard way).  The easiest way to drive an RS232 line with +/- 5 volt
supplies is to use a PNP transistor:


                         +5v
                          |
                          \
                          / 3.3k            +5v
                          \                  |
                          /                  |
                          |                  /
                          |      3.3k      |L
                TTL  -----------/\/\/------|
                                           |\
                                             \
                                             |
                                             |------------ RS232
                                             |
                                             /
                                             \  330 ohms
                                             /
                                             \
                                             |
                                            -5v

Jim Ericksen
Dept of Comp Sci
UNC Chapel Hill

henry@utzoo.UUCP (Henry Spencer) (04/19/85)

It seems that a remarkable number of people try to use the 7660 without
reading the datasheet, or at least without paying attention to its
rather limited ability to source/sink current.
-- 
				Henry Spencer @ U of Toronto Zoology
				{allegra,ihnp4,linus,decvax}!utzoo!henry

henry@utzoo.UUCP (Henry Spencer) (04/19/85)

As somebody has pointed out, you can't reliably run a 1488 on +-5.
This is a major reason for spending some time and effort on getting a
higher voltage, rather than trying to run RS232 on +-5 (even though
it's within the RS232 spec).  The combination of a 1488 and a 1489 is
amazingly tolerant of all kinds of out-of-RS232-spec conditions; those
chips are the major reason why you can get away with running RS232 cables
much farther than the spec says.  It is probably better to use these
decidedly-superior off-the-shelf drivers and receivers, even though it
does mean some hassle to get the 1488 the $#%$#@ supply voltages it needs.
-- 
				Henry Spencer @ U of Toronto Zoology
				{allegra,ihnp4,linus,decvax}!utzoo!henry