AC Power Supply question

[juan_felt]

Hi!

I built a DC, regulated and filtered power supply for me, and I'd like to make an AC power supply for pedals that need such type power.

My question is if one can filter and regulate such power supplies, and if this is necessary, as I noticed that AC transformers provide a higher voltage when unloaded. Also, do commercial power supplies that provide AC power do this?

Thanks in advance!

Cheers,
J

[GibsonGM]

Hi Juan,

I am of the understanding that pretty much all (correct me if I am wrong someone!) pedals using AC power have a built-in rectifier/filter which simply turns it into DC, just as you have done with your other supply!     Semiconductors do NOT like AC...they run on DC.   

As far as your question re. 'regulation' of AC...that's a neat one.  You can level shift it with caps, resistors...I haven't heard anything about regulating anywhere, tho.  I am sure it CAN be done....but would we NEED it for what we do?    I don't think so.

Let's see where this discussion goes! 

[R.G.]

Regulation of AC power supplies can be done, but it's complex, difficult and/or expensive in some combination.

As a practical matter, none of the pedals that use AC power have any form of regulation on their incoming AC power supplies. Some of them may use some filtering, but probably not.
Pedals that use AC immediately rectify the incoming AC to DC, filter the ripple voltage with big capacitors, then use linear regulators or switching regulators to convert the DC into a different DC voltage that the pedal circuits use.
It is not necessary to filter or regulate such supplies. The pedals do their own regulating and filtering inside the pedal.

[juan_felt]

Thanks to all the answers!

[PRR]

> re. 'regulation' of AC...

Easy and routine:

https://s24.postimg.org/7vvkp0bdh/SOLA_1945.gif

Expensive, loud, nasty waveform, and loud.

[Rob Strand]

I'm with RG on this one but for interest there's actually a whole world of this typ of thing out there:

Easy and routine:

Sola invented the CV transformer.

AFIK, the Sola company is still around and make UPS's etc.

There's quite a few modern power electronics ways to do this, the simplest is:
mains AC ->  rectify filter -> boost converter regulated DC rail -> Quasi Sine Steps  ->  (Filter) ->  Regulated AC.

mains AC ->  rectify filter -> boost converter regulated DC rail -> PWM'd H-Bridge  ->  Filter ->  Regulated AC.

The boost converter can be modified to produce unity power factor.   The structure lets you convert voltage & frequency and even create 3-phase output.  If want you can lock to the incoming mains frequency.  Further mods are bidirectional power flow.  UPS's and motor controllers use these ideas.

(I did an AC regulator for my undergrad thesis,  purely in the AC domain.)

You could actually do the above on the low voltage DC side.

[anotherjim]

http://www.circuitstoday.com/cvt-constant-voltage-transformer

[PRR]

> AC transformers provide a higher voltage when unloaded.

This is buyer's-choice. Everything sags; more money buys less sag.

Same as my dog-porch. I can put in $10 of floor-beams and see it sag under the Corgi, or $50 of beams and hardly sag even under a pack of Corgis.

In transformers, sag is a lot about size. Large size gives relatively more room for fat copper, and also more economic incentive to avoid sag-waste. The 25,000 Watt transformer feeding my house sags 2% at full load. I recently consulted on a 2.5 Watt transformer which was sagging >40%! There is a "generally accepted" trend of sag/size. If you truly need better, the best buy is to over-buy (buy 50VA to get 20VA). On paper you can do a little better for low-low-sag; in practice the custom design costs make over-buy the best plan.

We normally do NOT need to "regulate the wall". Wall voltages generally run within 5% of nominal-- if they don't, customers complain of dim lamps or lamps blowing-out too quick. Yes, there are "poor" installations. While my street transformer is 2% sag, at heavy (not full) load I have 15% sag at the house. The wire is too long! But even this sag is not a real problem, except when I go to sell the house. (The lamps dim when the pump comes on and the buyer's Home Inspector wants to call that a problem.)

Yes, I have worked places with much worse wall power. This typically means a bigger problem than a guitarist can fix. (Cheap bar owners, overloaded circuits, remote locations, ugly generator power.)

Nearly ALL audio circuits will run fine on +/-10%, even +/-20%, of nominal voltage. We pick our final supply to be "bigger" than our maximum audio level. And in most speech/music work, "Max" happens less than 1% of the time. (Guitar distortion is different.) Little-bigger or more-bigger is not a big deal, as long as we don't exceed the ratings of the devices. In "small" audio, this is never a real problem: we need a few volts of DC, while 15V and 36V devices are readily available.

So "Regulation" is not too important. We would like it to be "steady", no sudden change. But the main thing is we want "NO audio crap" on the power. Because 50Hz/60Hz is some audible and 100/120Hz ripple is real annoying, we have to filter the _DC_.

Of course the _AC_ IS 50Hz/60Hz. We can't get rid of that. We usually take the AC any way we get it, convert to DC, and *then* apply our filtering techniques. (And yes, $1 regulators have become cheaper than fat chokes and capacitors.)

[PRR]

> modern power electronics ways to do this, the simplest is:

FWIW, a simpler way. I have seen this done in commercial product. It met specs, though the output "corrected" in steps.

Auto-transformer with say 4V taps around the nominal 120V point. Bunch of Triacs from taps to output. Controller engaged the appropriate Triac so-that output stayed "120V +/-5V". This was plenty good for computer network gear in bad places.

As the maximum correction was around 20V, the autotransformer only had to be sized like 20% of the total VA. To a point, the Triacs were inexpensive and reliable. No DC involved so no blocking caps or floating-bridge (broken neutral) output.

Utility companies have long used an electromechanical tap-changer for the same purpose.

As an audio "filter" it begs for "jerks" in the audio. We had such a mechanical tap-changer at work, which got confused, and was jacking 113V/118V several times a minute most days. Interesting that most audio gear showed hardly a hint of this, but I could hear it in the speakers.

[Rob Strand]

Auto-transformer with say 4V taps around the nominal 120V point. Bunch of Triacs from taps to output.
[quote/]
That would work well for areas where the power normally high or low.   Kind of an auto voltage selector.  I know the line voltage in the US varies from place to place.  The voltage here it is pretty good despite all BS with privatizing the power companies.  I guess they got the system right before it went private.

Utility companies have long used an electromechanical tap-changer for the same purpose. [quote/]
Yes, the big stuff is always simpler.  I remember my power systems lecturer saying they momentarily shorted between the two taps to prevent a voltage dip.   I always wondered about the large current that would flow shorting an effectively low voltage winding.  The transformer would survive but the contacts might take a beating.

As an audio "filter" it begs for "jerks" in the audio. We had such a mechanical tap-changer at work, which got confused, and was jacking 113V/118V several times a minute most days. Interesting that most audio gear showed hardly a hint of this, but I could hear it in the speakers.

I guess that's the down side of the method.

[R.G.]

We're confusing the beginners again, guys.