My power supply vs the average wall wort. / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

My power supply vs the average wall wort.

Move Thread

LAN_403

MikeT1982

Sat Aug 11 2007, 06:51AM

Registered Member #621 Joined: Sun Apr 01 2007, 12:37AM
Location:
Posts: 119

Ok if I put in enough smoothing capacitance to yeild a 1% ripple at full draw, then put on a voltage regulator (which you guys taught me will then decrease the ripple by about 1000 to 10,000 times, that ought to give me very little ripple.

My question is this. The things being ran by the power supply will be 4 tiny 5milliwatt self contained with driver circuit included, diode laser assemblies. These are to help me aim my Ruby laser. I have seen the same diode lasers runnning fine off a wall wort transformer first hand.

So how would my design of 1% ripple and a v-reg stack up to your average radio shack wall wort? I really want to get ahold of one and tear it open to see how much capacitance they have inside to smooth.....I am curious if wall worts are built with bare minimum needed orif they are generous.

Of course I could just integrate a wall wort into my laser, but no I am too stubborn and have to make my own power supply!

...

Sat Aug 11 2007, 07:23AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

normal wall warts are just a diode bridge and a few thoursand uf (if you are lucky) of caps. You supply is way superior...

If all you needed to do was get a 5mw laser goin you just needed to get it to like 10% ripple and you would have been fine... Just make sure that you either run all of the diode modules in series, or give each one its own current limiting resistor (assuming it doesn't have a driver already), as if you parallel all of the diodes they will fight over the power and end up dying one by one.

MikeT1982

Sat Aug 11 2007, 08:00AM

Registered Member #621 Joined: Sun Apr 01 2007, 12:37AM
Location:
Posts: 119

Wow thanks. Do you mean 10% ripple before the v-reg or even without the v-reg 10% ripple would be ok? I'll still overdo it to be safe though.

Simon

Sat Aug 11 2007, 10:15AM

Registered Member #32 Joined: Sat Feb 04 2006, 08:58AM
Location: Australia
Posts: 549

It's not hard to improve on the wall wart. They're made on minimum cost.

If you're running a diode laser, you'll be having current regulating circuitry on the diode itself so extreme regulation of the power supply is superfluous.

Since you seem interested in regulation, another tip is to have separate power supplies for separate parts of the circuit. A lot of ripple comes from loads inside the circuitry.

In a typical audio circuit, I'd have a separate regulator for any higher power amps. Otherwise the power amps act as an oscillating load, which generates oscillations in the power supply rails, that feed into the signal amps, which boost the oscillations to get fed in a loop into the power amps. You'll hear this in the output as a buzz and the simplest solution is to just give the power amps their own regulators.

Steve Conner

Sat Aug 11 2007, 10:44AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

2200 uF per amp of current draw is a rule of thumb I often hear quoted when sizing smoothing caps for low voltage power supplies.

Too much capacitance can be as bad as too little, because it decreases diode conduction angles and leads to poor power factor on the line, and high losses in the transformer and diodes.

In your low powered application, this will hardly matter, the only drawback will be the extra space taken up by a big pile of caps

But I would say that 1% ripple implies way too much capacitance, I think these power supplies usually have 5-10% and the regulator just mops up the rest.

In audio amps, the main power supply is usually unregulated, and the negative feedback loop in the power amp removes the ripple from the audio output, just like a voltage regulator does. The power amp I built has +/-50V rails with 10,000uF smoothing capacitance per rail. At full output, there is about 5V of line frequency ripple on the rails, but a spectrum analysis of the output shows the line frequency components are about 80dB down from the signal.

MikeT1982

Sat Aug 11 2007, 05:05PM

Registered Member #621 Joined: Sun Apr 01 2007, 12:37AM
Location:
Posts: 119

Awesome, awesome. I never put 2 and 2 together (which I tend to not do) about unregulated audio amps! I often ran into unregulated amps in the car audio world and regulated ones. Thats impressive how they use their internals to do the same mopping up of the ripple as a V-reg would.

Simon, just as you say with the separate power supplies I was throwing that idea around. In my other post I mentioned a 3-4 amp draw, and that was because I was going to run my centrifigul blowers for forced air cooling, audible alert buzzers (trying to make this a fun laser LOL) and relays AND the aiming lasers all off the same transformer and smoothing cap bank. Now with all you guy's input I decided I'd be best to run one step down transformer (which I will center tap for 6.3 volts) to one bridge and smoothing bank for the diodes, tap for it's full 12.6 volts and do the same for the relays buzzers etc. And a separate second transformer/bridge/smoothing bank/v-reg for the blowers that are the current hogs...1.2 amps each. Just like you said Steve I'll have to suck up the space waste for the pile of caps, and maybe I will reduce theri number a bit to between a 5 percent to 10 perscent ripple before the v-reg. I'll tell ya, without this place my project would be a no go. Thanks!!!

GreySoul

Sat Aug 11 2007, 06:16PM

Registered Member #546 Joined: Fri Feb 23 2007, 11:43PM
Location: Albuquerque, NM
Posts: 239

Certainly not all transformers are equal in design...

I have about 10,000 9v wall warts. Most are 500ma with a full wave bridge and a 25v 470uF cap. They out out good power with very little ripple. they're higher quality than the warts from RadioCrap.

The others are 200ma, made with an ugly transformer, a single diode and no caps. They measure 8-11V and have the expected half wave ripple. They've killed almost everything I've hooked them up to. If anyone wants them they're yours for $100 + S&H cost - have to take them all, about 120 pounds worth. There's about 12g of copper and a 1N4002 diode in each one, along with 4" of wire - no plugs.

the 500ma's I'm keeping... gonna make a giant 9v 3000A power mountain with them :P

-Doug

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob