Voltage Divider - help with simple AC circuit

[Beside Myself]

As you'll gather quickly enough, I'm a novice, and I would appreciate
the guidance of someone experienced in electronic circuit design.

I would like to build a simple Voltage divider circuit. I want to take
the single phase 120VAC, 15A, 60Hz out of my wall and step down the
voltage/current to around 2 - 4VAC & 1A - 1.4A. My goal is to feed
this signal into my audio converters so I can see the constituent
harmonic energies and other noise in the mains AC. I would like to
read from 1Hz to 20kHz.

I'm guessing the easiest way to do this would be to use a transformer,
but as I understand it, transformers don't handle frequencies equally.
So I'm thinking a resistor-based circuit might do it. Do they make
resistors that could dissipate that much energy safely? (i.e., 1800W
less approx. 5.6W). Is it that easy...? (My suspicion is that nothing
ever is) I would also like to build in a fuse so that a transient in
the mains doesn't blow my converters!!!

There is also the possibility of a zener diode-based circuit, but I'm
unsure whether the diode would introduce unwanted changes to the
frequency spectrum, as in the case of transformers.

Your help is very much appreciated.

 

[Tomi Holger Engdahl]

As you'll gather quickly enough, I'm a novice, and I would appreciate
the guidance of someone experienced in electronic circuit design.

I would like to build a simple Voltage divider circuit. I want to take
the single phase 120VAC, 15A, 60Hz out of my wall and step down the
voltage/current to around 2 - 4VAC & 1A - 1.4A.

You are talking about pretty high power voltage divider..
1.4A from output of voltage divider would indicate that your
voltage divider would need to take at least this amount of current,
practically usually considerably more from 120V AC.

We are talking here ower 160W power losses here!

Not for any small resistors.

My goal is to feed
this signal into my audio converters so I can see the constituent
harmonic energies and other noise in the mains AC. I would like to
read from 1Hz to 20kHz.

Where do you need here that high current ?
Audio converters have typically quite high impedance inputs.

I'm guessing the easiest way to do this would be to use a transformer,

That's usually the easiest and safest. The transformer typically gives
you safety isolation from the mains voltage.. With resistor divider
you measuring circuit is directly connected to mains wires.

but as I understand it, transformers don't handle frequencies equally.

This is usually more or less the case. Some transformers do this
better, some worse.

So I'm thinking a resistor-based circuit might do it. Do they make
resistors that could dissipate that much energy safely? (i.e., 1800W
less approx. 5.6W).

There are high power resistors that can dissipate lots of power
when properly cooded. Usually the cooling is made my using
a large heatsink or make the resistor heat some liquid (water, oil).
High power resistors are used in heaters, ovens, washing machines
(to heat the water in it), tea water heaters, coffee makers...
You can easily see 1 kW resistors in those heating applications.

Is it that easy...? (My suspicion is that nothing
ever is) I would also like to build in a fuse so that a transient in
the mains doesn't blow my converters!!!

A fuse in the circuit powered from mains is practically a must
for safety reasons. It will burn if your circuit accidentally
starts to take too much power (for example short circuit).
A properly sized fuse will blow before some part of your circuit
will start burning.

Fuses do not help much to solve mains transient problems.
Fuses are too slow protection components to protect
against transients. Against transients need othet kind of
protectors.

There is also the possibility of a zener diode-based circuit, but I'm
unsure whether the diode would introduce unwanted changes to the
frequency spectrum, as in the case of transformers.
Your help is very much appreciated.

Zener diodes are used in many protection circuits. A properly
sized zener diode should not have too much changes in measurement
results. There are lots of measuring electronics that use zener
diodes and other protection circuits in their inputs to protect them.
The key in selecting those components is select them in such way
that they do not do nasty things on the normally used measuring
voltage range.

 

[Gerard Bok]

As you'll gather quickly enough, I'm a novice, and I would appreciate
the guidance of someone experienced in electronic circuit design.

I would like to build a simple Voltage divider circuit. I want to take
the single phase 120VAC, 15A, 60Hz out of my wall and step down the
voltage/current to around 2 - 4VAC & 1A - 1.4A. My goal is to feed
this signal into my audio converters so I can see the constituent
harmonic energies and other noise in the mains AC. I would like to
read from 1Hz to 20kHz.

Simple answer: this cannot be done.

If you want to see the noise and harmonics components, and also
want to live to tell it, you need some pretty sophisticated gear.

A simple devider approach won't work at all. You may think your
mains carries a steady 120 Volt, but in fact it doesn't at all.
Any load you switch on will vary the voltage on all other
outlets. And it depends on several factors (wattage of the load,
distance to then nearest power distribution point, quality of the
wiring, etc.) by how much the voltage changes.
These changes are likely to be an --or several-- order(s) of
magnitude higher than the noise and harmonics you want to see.

Also, your circuit would have to be able to cope with spikes that
occur on your mains circuit. This is also highly dependent on
your local situation. 1 KV spikes are rather common. (I don't
want to estimate what you get with pole wire feed during a
thunderstorm.)

With your approach: you will at least burn your hands. (As 1800
watts gets hot, by any standard.) Probably your PC too. And hope,
that you won't burn your house!

Bottom line: any novice should stay away from direct mains
attached stuff. If it doesn't run on batteries or a wall wart,
don't touch it!

 

[Beside Myself]

Hi Tomi

- a very sincere thanks for the trouble you've taken to help me out.

If I were to go the transformer route, is there something you could
recommend that would do the job without changing the frequency
spectrum of my AC mains?

How about something like this thing?

http://variac.com/staco_variable_transformer_100_.htm




[snip]

 

[Beside Myself]

Thanks for your advice, Gerard.

Would a variable transformer, such as the one below be useful?


http://variac.com/staco_variable_transformer_100_.htm

 

[Gerard Bok]

Would a variable transformer, such as the one below be useful?


http://variac.com/staco_variable_transformer_100_.htm

 

[petrus bitbyter]

Thanks for your advice, Gerard.

Would a variable transformer, such as the one below be useful?


http://variac.com/staco_variable_transformer_100_.htm

That variac is a nice piece of equipment but it does *not* serve your
purpose. Your questions make very clear that you are a novice indeed. I even
doubt you to understand what you want to achive. The only thing you can get
from your audio equipment is a fat 60Hz hum... if you don't blow it. If you
nevertheless want to continue, you'd better first make up your will. Add a
note to inform this group so we can nominate you for a Darwin Award.

BTW. I don't blame you for being ignorent. Consider it the strongest warning
I can think up to stay away from the mains as long as you don't even know
what you're talking about.

petrus bitbyter

 

[CWatters]

My goal is to feed
this signal into my audio converters so I can see the constituent
harmonic energies and other noise in the mains AC. I would like to
read from 1Hz to 20kHz.

Your proposal to use a transformer or zenner suggests that you might not
have the experience needed to do this safely. Best hire a line analyser.

 

[Tomi Holger Engdahl]

an audio isolating transformer (eg 600 ohm to 600 ohm) with a large series
resistance (eg 1Megaohm) on the input side

It needn't be something bulky or expensive.

The frequency response of the 600 ohms audio transformers is geenrally
the best when they are driven with 600 ohm source and connected
to 600 ohm source. Or driven with low impedance source and
connected pretty high impedance load.

I don't think 1 megaohm in series the input side would be optimal for
good frequency response.

 

[Beside Myself]

[snip]

John, thanks so much for your efforts to explain your circuit idea to
me. You've done such a thorough job that I don't have any questions
off the bat - except where your stroke of genius came from

I'm definitely going to give it a whirl. Will let you know how it
turns out. If you don't hear back from me...well, I believe one of the
responders wanted to nominate me for a Darwin award...I think I'll
make it, tho

Thanks again!

Here is a _very_ dangerous way to do it, and if you're not familiar
with working with AC mains, I suggest you _don't_ do it.

Anyway, view in Courier:




120 HOT>----+
|
[120V 100W LAMP]
|
+------> TO AUDIO CONVERTER INPUT
|
[3 OHMS]
|
120 NEUT>---+------> TO AUDIO CONVERTER GROUND


The lamp is a standard 120 volt 100 watt incandescent lamp, and when
it's on it'll be passing about

P 100W
I = --- = ------ = 0.833 amperes
E 120V

which, when passing through the series resistor, will cause a drop
across the resistor of:

E = IR = 0.833A * 3R ~ 2.5V

Then, with the lamp on, the resistor will hbe dissipating

P = IE = 0.833A * 2.5V = 2.08W

So something like a 3 ohm, 5 watt resistor would be OK to use.

Now, there are two procedures which, if you don't follow, could
result in you getting killed or your equipment getting fried, or
both.

The first thing is to make _SURE_ that you have the circuit
connected to the mains _EXACTLY_ as shown and to make _SURE_ that
you've positively identifed the hot and neutral sides of the mains.

The reason for that is that if you connect the thing backwards
you'll be connecting 120V directly to the sound card's ground, which
is not a good thing to do.

An easy way to find out is to connect an AC voltmeter from the
ground terminal (the "U" shaped one) in the receptacle to one of the
rectangular contacts. If the receptacle is wired properly, you'll
read 120V between ground and the smaller contact, and zero volts
between ground and the larger contact. That means that the small
contact is HOT and the large contact is NEUTRAL. If you find it
wired differently or if you don't know how to do the testing, then
blow off the project or get an electrician to help you.

The second thing is that you need to have the circuit connected to
the mains with the lamp burning _before_ you connect the output from
the resistor to your audio converter.

The reason for that is because when the lamp's filament is cold its
resistance is pretty low and it'll let a lot of current through
which, when it goes through the resistor could generate a high
enough voltage to maybe hurt your audio converters. I just measured
a lamp, cold, and its resistance came out to be 13 ohms, so when
first turned on, the current through the lamp and the resistor would
be:

E 120V
I = --- = ---------- = 7.5A,
R 13R + 3R

and that current will drop:

E = IR = 7.5A * 3R = 22.5V

across the resistor, inititially, and if your converter ws connected
across the resistor before the circuit was connected to the mains,
that 22.5V would appear on the input to your converter.

As the lamp filament heats up, the current will drop until it gets
to 0.83 amps, then the drop across the resistor will be 2.5V and
it'll be OK to connect your equipment across the resistor.

There is a third consideration, and that's to keep seriously high
line transients from damaging your converter, and to do that you'll
want to put a clamp of some type across the mains. Probably the
easiest way to do that would be to get a socket strip with integral
surge suppressors and a circuit breaker and use the strip to supply
the AC for the monitoring circuit. That's also a good idea since
the circuit breaker will cut out if you do anything drastically
wrong.

So that's basically it, and if you don't understand any of the
foregoing, post back with your questions and I'm sure at least one
of us will try to help you.

Understand, though, that no matter what, this is a _dangerous_
circuit and it would be a good idea if you had someone with you when
you fired it up. Just in case...