Z4648252@SFAUSTIN.BITNET (Z4648252) (07/17/89)
I'll be posting a formal article later, but the following is an
overview of how to give your Mega ST automatic power up sequencing when
a hard drive is a part of the system.
As you should know, the ST, in a hard drive environment, is incapable
of automatically auto booting from a power up condition. This is a bad
situation if the ST is used in a lab environment or if used as a BBS
system.
One solution is to build an external power delay box that the ST can
plug into. When power is applied to the system, the hard drive spins up
and a few seconds later, power automatically applies to the computer. The
only problem that I've seen with this is that there is only one box for
sale which offers the sequencing. It is the sequencer that I've been
building since 1986. I hate building it, it takes five hours to build a
single unit and loath a request for one. Really, I mean it!
Another way is to install a 555 timer circuit inside of the ST. It
will delay power to the motherboard and I've seen the circuit description
floating around for about a year. It is a bit involved, though and features
a little overkill.
A simpler way is to use that which the ST, specifically the Mega,
already possesses. I say Mega because I don't have access to a 520 or
1040 but assume that these platforms can also profit from the project that
I'm proposing. The following project is quick, easy, and WORKS! There are
lots of things that you can do with this which will be the subject of an
article later. If anyone can observe design risks with this, please alert
us. And as usual, if you fry your machine, well.... Of course you know
this voids your warranty!!!
---------
Inside of the Mega ST2, revision #5, a very remarkable chip has been
wisely used by Atari Computer Division. The 555 timer chip is, according
to Forrest M. Mims III, "one of the most...versatile integrated circuits
ever produced. It includes 23 transistors, 2 diodes and 16 resistors..."
I rejoice in Atari's use of this chip, especially since it governs
the RESET circuit. And, THAT, is very important for our project.
As a side note, hardware hackers love the ST. Easy to modify, tough
as nails, and quite versatile, it is a joy to play around with. All that
is missing are slots!
What we want is to increase the delay time for the ST to come on-line
when power is applied. The 555 timer does this but, due to values of
resistors and capacitors chosen by Atari, the timer gives a very abbreviated
delay. What we want to do is to increase the delay in two experimental
phases.
Keeping everything simple is the goal, hence, the phases. The first
phase is the easiest, one merely installs a capacitor on top of another.
If you are lucky, this is all that you have to do. The second phase is
a little bit more involved and necessary only if you can't get a long
enough delay. However, it only involves the addition of a resistor or
potentiometer.
How do you know if the delay is long enough? Easy, apply power to
both the hard drive and ST at the same time. If the ST ignores the
hard drive, then the delay was not long enough.
All of the above is the background information. Now for the meat.
Next to the reset switch is the 555 timer and its support group. These
components determine the delay. The first component that we want to
experiment with is a 10mfd capacitor labeled C3. Pull out of your parts
box capacitors of 75mfd on up and tack solder one on to the leads of C3.
Prepare the ST for power up sequencing.
You should note a delay. Keep experimenting and increasing the values
of the capacitor until the delay is maxed out.
For the most part, the delay is "almost" long enough. I noticed that
just this simple capacitor addition to C3 results in the system auto-
sequencing properly 9 times out of 10. However, that is not good enough.
We need more delay.
Ok, here comes the second phase. Note the following diagram:
8 7 6 5
U1 -- 555 timer
-----------------
-----------------
1 2 3 4
You will find that the underside of the motherboard has a bridge
between pins 7 and 8. You've got to cut that bridge, trace would be the
more appropriate term. Now, the ST won't boot up at all!!!!! Not to
worry....
What we want to do is to install a potentiometer between 7 and 8.
To get all of this to fit, you'll find convenient traces running on the
motherboard which you can use, going to the two pins. The power supply
fits on top of the 555 timer area hurting access to the pot. Just find
a convenient spot, finding appropriate traces. I found an area next to
where the disk drive and power supply plugs are.
Adjust the pot about 1/3rd up. Oh, pot value? To be honest, I used
one from the parts box and it isn't labeled. I've no idea as to what the
value is. Just grab one. I am using a service mount pot and it works
very nicely. Ok, turn the power off, wait half a minute or so and apply
power to the hard drive and ST at the same time, just as if you were
recovering from a power outtage.
If you've applied too much resistance, then the ST won't boot. Lower
the pot's adjusted value. Try again. Keep on doing this and eventually,
you will find the correct value.
That's it! All that we've done is increase the delay in two stages:
C3 -- capacitor. Originally 10mfd. Tack at least a 75mfd on it.
555 pins 7&8. Break the connecting trace. Connect a pot in between.
There are all sorts of modifications that you can do without garbaging
the computer with gadget junk. One thing you can do is to install a
master rotary switch on the side of the Mega. By switching the rotary on,
power is applied to both ST and hard drive. That's neat. Another thing
that can be done is to create a bypass of your delay modifications so
that the ST acts "stock".
I hope that this is of help to those who need automatic powerup
sequencing. Again, if anyone sees design flaws that can be a risk to the
Mega as a result of the modification, please let me know. Both phases
have been my Mega for three months and I've had no problems.
Larry Rymal in East Texas <Z4648252@SFAUSTIN.BITNET>mark@ccvr1.uucp (Mark Boyd) (07/17/89)
I took apart a 1040 ST to check the reset timer. It is a 556 ( a dual 555 ) instead of the Mega's 555. I traced the wiring from the reset switch and found that ( on an older 1040 ) the timing resistor is 22K and the capacitor is 2.2 uF. This would give a delay of less than .1 second. If you want a reasonable delay for HD startup, a 10 Meg potentiometer should work with the existing capacitor. Put it in series with the resistor ( marked R8 on the system I checked ). You'll get more reliable timing by going to a larger capacitor - just solder it in parallel with the 2.2 uF. ( marked C4 on the old 1040ST ). I'd start with a 22 or 33 uF capacitor ( these are polarized, so make sure the + is connected to the + of the 2.2 ) and a 1 M potentiometer. All of the above is purely hypothetical for me. My hard disk is too noisy to use all of the time, so I'm not going to make this mod myself. I have done lots of hardware stuff and I've used lots of 555's, so I'm offering an little 'informed opinion'. This does sound like a simple, reliable, way to get a proper power on boot from the hardisk. Mark Boyd Computer Science mark@ccvr1.ncsu.edu mbunca@ecsvax.uncecs.edu