I want to built a basic power supply with 6 7[89]?? regulators. I have
a line transformer capable of supplying 31VDC (15.5 + center-tap,
scrap transformer).
You probably mean VAC. Transformers on their own don't work well with
DC. Where did you find the term 31VDC written?
By the way, I love the idea of learning some electronics by trying a
hand at building a power supply. So this is a great project to work
on. You can start by ignoring some details that would probably be
more confusing that helpful and later get better at it by bringing in
those details later on when you are ready for them.
Ah.. yes it is VAC, not VDC. sorry.
Hehe. I had worried a little.
And yes, I do have some bridge
rectifiers (left over from a project using relays, where the relays
only allowed switching in 1 direction, but I needed to be able to use
2).
Okay. I'm still not clear on the amps available. Do you have a way
of finding out? Or, alternately, can you say what you want to be able
to support?
15.5VAC is written on the transformer, but it's center-tapped and
the markings seem to indicate that is between two adjacent taps (and
not the end taps).
It might help to measure from the ends, as well. In other words, work
out the details. I gather you have a meter, so use it. Doesn't take
that much to hook it up to a snipped off AC cord (I have those laying
around from stuff I throw away.) Alligator clip it, if you have to.
And this isn't really something that I'm doing for
one of my first projects - I'm building an 8-bit TTL-based calculator
right now. I just posted it in basics because it's a basic power
supply - nothing fancy. So gimme the details, if you want.
Well, I'll do what I can. I learn as I go, too, being a hobbyist. The
"big boys" will jump in when they feel I screw up badly, I'm sure.
I want variable voltage from the regulators, which I understand can be
done by either putting a pot on the output (relatively obvious)
Yes, but really, really terrible, too. In the sense of -- "it just
ain't done" except in VERY low current situations (your usual
'divider' situation.) Certainly, not for a power supply.
to
lower it, or floating the regulator above ground with a resistor
between ground and what the reg "thinks" is ground. I know that
electrolytic caps have a polarity (though I believe I have blown at
least one by hooking up a not-so-well-marked car radio up backwards),
so that's not an issue.
Yeah. I've hooked them up to AC power outlets back as a kid just to
watch them blow up. Bad gases, I heard later. But the smell isn't
the kind of thing you stick your nose into, anyway, so I mostly
avoided getting too much into my system back then. And I didn't blow
up more than a small number, even then. Gets boring.
I found two pairs lying around which are 30uF and 1uf. Will that be
good enough?
For some of the IC regulators, perhaps. What bothers me is that you
may need something larger than 30uF to keep the ripple down leading to
your regulator if the current is much of anything. The dV is easily
seen from the I=C dV/dt equation, as dV = (I/C) dt. Your dt is pretty
much determined by your AC Hz and the choice you make for the diode
topology. With 60Hz and a bridge, you are talking about 120Hz bumps
coming out. That's 8.33ms per. And the capacitor may need to supply
current for a substantial part of that time. Assume for a moment that
it is 72% of the time or about 6ms. Then you have a ripple dV of some
200*I. If you can accept a 2V ripple, that means your load current
needs to be 10mA or less. Basically, a heavy load suggests bigger
caps to keep the ripple down to something manageable. And if you
accept a lot of input ripple, just to keep the caps smaller, then your
power supply regulator needs to be better at ignoring the ripple. If
it is a linear one besides, you still need to make sure that the
lowest point in the ripple is still large enough that it provides the
output voltage required plus whatever overhead is required by the
regulation system.
So it's important to know what you plan for the regulator (switcher vs
linear), what you hope to get out as a maximum load current, and over
what range of output voltages, plus some idea of what you are willing
to accept as ripple on your regulated output, too. Among other things
I'm sure I will be told about.
I have plenty of resources - true-RMS multimeter with transistor
checker and (!) capacitance measurement, 100MHz oscilloscope (it cost
about 1/4 of the online price at a place called Stewart Smith's (I
live in Syracuse, New York)), home-built parallel port logic analyzer
(might even be powered off of this supply!), and of course the
mandatory connection to the net.
Well, that sounds great. You have some decent equipment and hopefully
have reasonable understanding about applying them (I learn myself over
time to use what I have better.)
I have an old HP 524C counter which I have gotten numerous golden-
years-of-electronics-no-clue-what-semiconductors-are-age components,
including the above caps. And there are two electronics stores in
town, so no problem getting parts.
Okay.
Are you interested in working through the design thinking? Or just
want a circuit and don't care to understand why?
And in the meantime, can you measure that transformer and let us know
what range of output voltages are okay with you (both the low end as
well as the high end -- does it really need to be able to go down to
0.2V or to 0.00V, for example?) Also, what compliance you want for
the current? (Which gets back to maybe telling us just how heavy that
transformer is or if it has a rating on it, somewhere, or part
number.)
Jon