Theory behing 110 V AC RC driver Circuit ???

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Post Reply
jalbers
Posts: 18
Joined: Tue Dec 18, 2001 1:01 am
Contact:

Theory behing 110 V AC RC driver Circuit ???

Post by jalbers » Mon Jun 23, 2008 10:03 am

A while back I saw a post about powering an LED using 110V AC. The circuit consisted of wiring the LED in series with a 1K resister and .1 uF capacitor. A reversed diode was also wired in parallel with the LED to protect the LED from reverse bias.

I have a few questions about the circuit:

1. How were the values for the R & C determined? There are lots of values for R and C that would make |Z| around 2.6K . What if anything makes the values specified in the circuit special?

2. Why not just use C and not RC. |Z| for C alone comes to about 2654 ohms. |Z| for RC comes to about 2836 ohms. Z for RC comes to [email protected] deg. I have a feeling that the phase shift value is important. What phase shift would give the hightest brightness?

3. How to model the circuit using complex numbers. I am interested in "instantaneous" voltage drops across, and "instantaneous" currents through R, C, and LED?

For example, if V source is 45 deg into its cycle, [email protected] deg, what would be the instantaneous current through R, C and LED, and what would be the instantaneous voltages across R, C, and LED?

4. Would the ciurcuit work if the diode was wired in series with the LED instead of parallel to it? Capacitors are able to pass fluctuating DC correct?

Any help would be greatly appreciated. Thanks

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Post by MrAl » Mon Jun 23, 2008 10:50 am

Hi there,


I think you triple posted maybe?

The line driven LED circuit with R and C to limit current can be explained
using complex numbers as i think you already figured out :smile:
There is a catch however, and that is that the AC analysis of the
circuit is not enough by itself...we must also look at the transient
response. Doing this leads to the explanation of what the R is
doing there.

Normally when the line is running full speed, the C limits the current
and R does very little but waste energy (not much however if the
circuit is designed right). The complex number AC circuit analysis
will take us this far. We can select the C value by doing a few
calculations and computing the average current through the
LED for the most part.

Next comes the transient analysis. When the circuit is completely off,
there is no voltage across C or the LED, and when the circuit is
suddenly switched on (perhaps the unit is plugged into the wall
for the first time in several hours or even days) if the line is
at 0.00v (crossing zero at zero degrees or 180 degrees) then
we have no problem. The C charges up as expected and everything
goes well. It's when the unit happens to be plugged in when the
line is at its peak (170v peak) that the problem comes in. With
170v applied to the circuit input, for the first few instants the cap
looks like a short circuit (transient response of the cap) and this
means mucho current flows into the input, which in turn usually
makes its way through the LED too. With no R in series, this can
be a very, very high current! Adding a series R however, even only
100 ohms, limits this short but potentially damaging inrush of
current to around 1.7 amps. This is a bit high too, so maybe 200
ohms would be better, as this would limit the current pulse to
0.85 amps max which the LED can handle for a short time
period. The down side to this is that the resistor dissipates
energy, whereas the cap doesnt as much. We have to then go
ahead and analyze the circuit for power dissipation in the R
using complex AC circuit analysis to make sure we size the
resistor R at a wattage that will make sure it doesnt burn up
during normal operation.

A summary would then be:
1. C limits current during normal operation
2. R limits inrush current to protect the LED during turn on, plug in,
and abnormal line transients. This resistor should be made a flameproof
type to prevent a fire.
3. A low value fuse should also be used in the input circuit.


Modeling the circuit could be as simple as this:

Include R, C, and make the LED a voltage source of around +3.5v
when the current is positive, but -3.5v when the current goes negative.

Alternately, calculate a resistor whos value is equal to the voltage
divided by the expected current. Dont worry about polarity, and do
an AC analysis. Use this resistor for the LED.

To get into the transient part, use v=i*dt/C for the approximation
to the capacitor, take small time steps, and calculate v and i for
each small time step. You can use a non linear LED model for
the LED doing it this way, which more closely resembles the real
life device.
LEDs vs Bulbs, LEDs are winning.

jalbers
Posts: 18
Joined: Tue Dec 18, 2001 1:01 am
Contact:

Post by jalbers » Mon Jun 23, 2008 10:56 am

Sorry about the multiple postings. After I hit submit, I waited about 10 miniutes for something to happen. I figured something was wrong so I hit the back button and tried again.

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Post by MrAl » Mon Jun 23, 2008 11:04 am

Hi again,

Oh yes, that happens to me sometimes too.
I hope the above post helped a little anyway.
LEDs vs Bulbs, LEDs are winning.

magnetar
Posts: 2
Joined: Fri Jul 25, 2008 2:14 am
Contact:

Post by magnetar » Fri Jul 25, 2008 2:28 am

Hi,
recently I try this circuit for a light switch indicator.
(using a capacitor, diode, resistor and led)
If the lamp is off then the led on, and if the lamp on then the led off.

However, I've told by some people, you can only just use a resistor and a led and plug it to 220VAC. It's common on rice cooker's standby indicator led. So i give it a try and it works. I use a 120K resistor.

I wonder, is this ok?
What about the more complex circuit using cap and resistor?

dyarker
Posts: 1752
Joined: Fri Aug 22, 2003 1:01 am
Location: Izmir, Turkiye; from Rochester, NY
Contact:

Post by dyarker » Fri Jul 25, 2008 5:58 am

You can, but using a capacitor is less power wasted. The capacitor shifts the phase of current relative to the voltage. You get current limiting with less heat.
Dale Y

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Post by MrAl » Fri Jul 25, 2008 10:44 am

magnetar wrote:Hi,
recently I try this circuit for a light switch indicator.
(using a capacitor, diode, resistor and led)
If the lamp is off then the led on, and if the lamp on then the led off.

However, I've told by some people, you can only just use a resistor and a led and plug it to 220VAC. It's common on rice cooker's standby indicator led. So i give it a try and it works. I use a 120K resistor.

I wonder, is this ok?
What about the more complex circuit using cap and resistor?
Hi there,

Yes, a 120k resistor (and two diodes) will work but the average current
through the LED is only going to be less than 1 milliamp, which isnt much.
It will work as an indicator lamp with the high brightness LEDs, but to
get some real light out of the LED you need to pump more current
through the LED.
Normally you would probably want to go with 15 or 20ma, but even
for 10ma (average current) you would have to use a 10k resistor
and that resistor would have to be rated for about 5 watts because it
will have to dissipate about 2.5 watts of heat. This might not sound
like much but it will get quite hot and so will not be contained within
a small enclosure very well at all.
Using a resistor and cap (and a bridge rectifier) you get even more
current (20ma average) with almost no heat at all (less that 0.5 watts).
This is the main reason for using a cap with the resistor to drive the
LED.

Just to note, if you tried running the LED at 20ma average with just
a resistor and two diodes, you'd have to use a 5k resistor and it would
have to be rated for 10 watts as it would be dissipating 5 watts and that
would get pretty hot.
LEDs vs Bulbs, LEDs are winning.

magnetar
Posts: 2
Joined: Fri Jul 25, 2008 2:14 am
Contact:

Post by magnetar » Sun Jul 27, 2008 8:33 pm

Thanks. I really appreciate this.
You've enlightened my mind :grin:

User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Post by Bob Scott » Mon Jul 28, 2008 12:12 pm

jalbers wrote:Sorry about the multiple postings. After I hit submit, I waited about 10 miniutes for something to happen. I figured something was wrong so I hit the back button and tried again.
In IE, when you hit the back button, the screen loads the old screen information that you saw before. Hit F5 to refresh and load and see the updated screen.

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Post by MrAl » Mon Jul 28, 2008 3:58 pm

Hi again,

Oh yeah, this happens to me once in a while. I think the message goes
through to the host web site (here) but for some reason it takes a while
to get the response that confirms you've actually sent something.
With this site it doesnt happen as much to me as with other sites, but
when it does i just delete the clone.

If anyone else has this problem feel free to PM me and tell me where
it happened and i'll delete the clone asap.

BTW, magnetar, if you have any other questions like this feel free to
PM me.
LEDs vs Bulbs, LEDs are winning.

Post Reply

Who is online

Users browsing this forum: No registered users and 35 guests