Convert 120 VAC to 5 pulsating DC.

[m2mike]

What would I need to convert 120 Volts AC to 5 Volts of pulsating DC? Wouldn't I just need a transformer to convert to 5 volts or close to it? Wouldn't I then need a rectifier? I don't want to filter it as I want the pulses to be easily detectable. If someone can suggest an implementation, I would be grateful.

<small>[ February 16, 2006, 08:33 PM: Message edited by: m2mike ]</small>

[Chris Smith]

That will only give you AC 60 Hz at 5 volts.

To convert it you need to completely rectify it to a DC voltage and current, store it in a cap or battery, and then use a oscillator to dish out DC pulses at the 5 Hz rate and the 5 volts you want.

However, a regulator will also most likely be involved in order to maintain a 5VDC level with the pulses from the storage method.

[Robert Reed]

Transform 120vac to 5 vac, rectify it and you have "5 vdc" pulses. Half wave rectification will give you 60 pulses per second. Full wave or full wave bridge will give you 120 pulses per second. Just remember that we usually call out ac as rms and dc as peak voltage, so your 5 vac (rms) will give you ~7 vdc (peak) or more exactly Rms x 1.414. I feel like I may be missing something from your post thogh, as this is quite simple and obvious to answer. Also they won't be pulses per-se, but half sinewaves.

[m2mike]

Thank you. I should have specified if I was trying to get half wave rectification or full wave rectification. By the time I re-checked my first post, you guys had already answered. Thank you for filling in the blanks. I would like to convert from 120 Volts AC at 60 Hz to 5 Volts DC at 60 Hz. So I know I need a transformer and diode bridge, right? Can someone recommend some actual components for both half wave and full wave implementations?

[dacflyer]

full wave 4 dioads.or a one piece bridge unit.

1/2 wave can be 1 dioad for a single out put supply..or 2 dioads for a center tapped supply.

what other components are you refering to?
to get what you want..you basically just need the transformer and dioads.. unless your wanting a case or what ever..

give us more info..

[jimandy]

I want the pulses to be easily detectable.

Your quote suggests you may use these pulses elsewhere in a digital circuit. Best to "condition" them, i.e. make the pulses have sharp edges, often done with a Schmitt trigger. Google that in and you'll find lots of info.

[MrAl]

Hi there m2mike,

What you didnt mention was how much "power" you
need from the output. If you need enough power
then yes, you should use a transformer, but
if you dont need much power (ie sync signal) then
you can simply use a resistive voltage divider
with a ratio of 5/120.
The idea is to half wave rectify FIRST (one diode)
and then use a resistive voltage divider to
divide 120v down to 5v.

If it really is a sync signal you're after and you
really need 5v peak then you can also use a
single resistor in series with a zener diode
(5v) in parallel with a reverse connected diode.
The output signal will be much squarer than the
output from a rectifed sine wave and the dv/dt
near zero degrees will be much steeper providing
for a very clean sync signal. You only have
to choose your resistor so it doesnt heat up
nor overheat the diodes.
I actually had to use this technique once when
designing a piece of commercial equipment in the
past that needed to be sync'd to the line.


Take care,
Al

[jimandy]

Al's suggestion is simple and economical and is one I would use in a pinch. But generally, in electronic design, any circuit that is connected galvanically to the hot side of the power line is to be avoided.

[MrAl]

Hi jimandy,

Hee hee... in our application we had to switch
three phases of 120v lines. Thus, the equipment
HAD to be connected to the main power line

Take care,
Al

[jimandy]

Al, I figgered you had good reasons for your approach. Just wanted M2mike to be aware since I have no idea of what he is trying to do with the circuit.

[Robert Reed]

MrAls suggestion is simple, However the succeeding circuit would have to share its ground with power line common. Something to keep in mind as the common lead can have a certain amount of garbage on it when referenced to true ground (could be a problem for separate grounds). The transformer will give you complete isolation,but your choice depends a lot on how critical your counting needs to be.

[ecerfoglio]

If you already use a transformer, rectifier, filter (and regulator?) to power your circuit, the "pulse" supply can be just an aditional diode (or two for 120 Hz).

It would be better if the main suply is either half wave or has a center tapped transformer, as in that case you have one side of the transformer at ground potential. But it will work even with a bridge full wave suply.

Let's see if I can draw it in ASCII:

Code: Select all

 
 
 
                                     + V to filter 
                        +-------+---------------- 
                        |       |
                        |       |        Pulsed DC
                        |       |        +-------+----
                        |       |        |       |
                       ___     ___      ___     ___     
                        ^       ^        ^       ^     
                       / \     / \      / \     / \     
                       ---     ---      ---     ---    
                        |       |        |       |
             °----------+----------------+       |
 Transformer                    |                |
 (center tap)°--------+         |                | 
 .                    |         |                |
             °------------------+----------------+
 .                    |
 .                    |
                    Ground
 
 Note:
 
      |                 |
 -----------------------+----
      |                 |
 No Connection     Connection
 
  

[Gorgon]

Hi m2mike,
If it is only the signal you want, you could also use an optocoupler to transfer it galvanicly isolated from the power grid. If you use a normal optocoupler, the output would be halfwave rectified to 60Hz. You would of course need to protect the led in the optocoupler from the reverse voltage with a normal diode. Use an optocoupler with high efficiency and a current of 1 to 2 mA. Drop the voltage over 3 or 4 resistors to reduce the voltage/power over each.

TOK

[m2mike]

The idea here is to create a DC 5V 50% duty cycle at 60 Hz. I should have mentioned that as well. Is it wiser to just convert to 5 Volts DC and then use a 555 timer? I think I can figure it out. Thank you very much.

<small>[ February 17, 2006, 04:13 PM: Message edited by: m2mike ]</small>

[MrAl]

Hi jimandy and Robert,


Yes, good points...it's good to remember the
possible dangers that can arise from connecting
something directly to the power line. I would
have to recommend that the signal ground be
isolated from the chassis ground.
In the app i was talking about we enclosed the
device in an aluminum chassis and ran the line
'ground' directly to the chassis (as required
by law for that kind of device) but the signal
ground was isolated from the chassis.

I've also done a triac controller where i had
to sync the triac firing angle to the line
waveform. Since the triac already had to connect
to the line part of the circuit had to be
connected to the line anyway. The trick was
(again) to keep the chassis ground separate from
the signal ground.

I like using a transformer too though, and even
better an AC wall wart which has to have gotten
UL approval already.
I built a special kind of power supply using an
AC wall wart once and dont have to worry about
UL approval (which can cost) because the wall
wart has it already.


Take care,
Al