Bi-Polar Wall Wart Power Supplies

Go To Last Post
14 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A current project for myself, not for production, requires a bi-polar power supply, +/- 12 V, at about 100 mA on each supply.

I would like it to be fairly tightly regulated, and anticipated using +/- 12 V linear LDV regulators supplied by a +/- 15 V wall wart.

Having seen hundreds of various WW power supplies in catalogs I was rather suprised when I when looking,( Mouser, Jameco, Digikey), and had trouble finding a +/- 15 V WW, at ~ 100mA. I found only one, at $36 each.

I really don't wish to put a transformer in the case and suppply it with 120. A Murata mini-dip type PS is an option at about $15 each. I don't want to use two WWs, but that would be the least expensive option.

Other sources or thoughts would be appreciated.

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You can use a switcher to make regulated -12V from the +15 input.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The route I am taking nowadays is making the required voltages with SMPS (MC34063 is my favorite: easy to find, flexible etc), and then regulate down to the required voltage: cool, clean and versatile. As WW I use whatever I can find ;)
Instead of the MC34063, its bigger brother MC34163 for more juice, or LM2575 f.i.

Nard

Edited: added two more types

A GIF is worth a thousend words   She is called Sylvia (2018), lives at Mint18.3 https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Have you considered using a single AC wall wart? And then two half-wave rectifiers (two diodes, instead of the usual bridge rectifier with four diodes) to generate positive and negative unregulated DC.

To get 15 V DC you would need at least (15V + 0.7V) / 1.4 = 11.1 V AC as input. Since this is the no-load case, the AC input needs to be higher when loaded.

Using a 12 V AC wall wart might be a borderline case, depending on your load. A 15 V AC wall wart should do the trick and should be able to be rectified into +/-15V DC.

If you can go with low-drop LVOs, you might be on the safe side with a 12 V AC wall wart, too.

In total just a wall wart, two diodes, two capacitors, and then the LVOs.

Stealing Proteus doesn't make you an engineer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Wow. All great answers, and good for making me re-think my approach to this.

Thanks!

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

All of the Wall-Warts that I've seen with +/- DC voltage outputs weren't a 2.5"x2.5" cube that plugged directly into the wall receptacle, per se'.

I have one or two around with a 3"x4" footprint that have the 110VAC cord coming out of one end, and the +/- DC voltage lines coming out of the other end. In addition, the low voltage side has a connector that looks similar to the old legacy keyboard plug on the original IBM PCs.

While writing this, it came to mind that you could even use a notebook power supply. They are typically about 18VDC. Some notebook power supplies have dual voltages. In any case, you could put switching regulators on your project board that develop the necessary +/-DC voltages and simply use a single output notebook power converter as your power source.

EDIT:
DocJC,

I looked thru my 3 or 4 dozen extra Wall-Wart style power supplies. I have about 8 18VDC@1A, 2 Gateway 19VDC@3A, and several other off brand units ranging from 5VDC@300mA to 18VDC@ 1A.

A few - the notebook power supplies - are switching type power supplies.

If you can tell me exactly what you need, I'll see if I have something that will work.

It seems that I have built up quite a collection of these things over the years! I'll never use most of them, if if I have something you can use, I'll gladly get it to you...

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Will the +/-12V be used on any sort of sensitive analog circuits? Where I work they specifically avoid switching regulator power supplies because of noise issues.

(wondering out loud) Will linear "post regulators" after a switching regulator (like Nard is proposing) provide a clean enough output? Normally I would think of linear regulators having to filter out the 50/60/110/120 hertz ripple. How do they perform with the many 10's (or 100's) of kilohertz coming from a switching regulator? I know, I know, it is probably in the datasheet :-) .

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

SteveN wrote:
Will the +/-12V be used on any sort of sensitive analog circuits? Where I work they specifically avoid switching regulator power supplies because of noise issues.

(wondering out loud) Will linear "post regulators" after a switching regulator (like Nard is proposing) provide a clean enough output? Normally I would think of linear regulators having to filter out the 50/60/110/120 hertz ripple. How do they perform with the many 10's (or 100's) of kilohertz coming from a switching regulator? I know, I know, it is probably in the datasheet :-) .


In one assembly that we build at work for GE MRI data collection electronics system, they use switching regulators for all of the analog. There doesn't seem to be an issue with noise. I suppose it depends a lot on the filtering in the analog sampling circuitry...

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The dual rectifier idea with AC output transformers isn't a bad idea. Remember, however, that, ideally, the rectified voltage is sqrt(2) times the RMS voltage. So, a 12V transformer will rectify into (about) +16V and -16V.

All that is nice, but most of these transformers put out the specified voltage at the specified current. They use small wire so that they are resistance limited against short circuits. This means that lower currents will get you higher voltage. Also, the rectified voltage will be less because the filter output is like a sawtooth and the peaks will be actual peak transformer output; the average is lower. Also, the peak is lower because the current flowing during the time the rectifier diode is on is 5x to 10x the average and that is what controls the drop in the transformer resistance.

This scheme IS simple, the parts cost is low, but there are some subtleties that are not altogether obvious.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:
...most of these transformers put out the specified voltage at the specified current.
Jim

Which explains why these devices almost always measure an open terminal output voltage that is something higher then the voltage specified on the name-plate of the device. They depend on a specific current drain to produce/reduce the open terminal output voltage to the specified name-plate output voltage value.

Any current demand less than that specified on the name-plate, the output voltage with be higher then that specified on the name-plate.

Any current demand that is more then that specified on the name-plate, the output voltage will be less then that specified on the name plate.

In other words, most Wall-Warts are not regulating the output voltage in the traditional sense. Rather, they use Ohms law and a few well known design short-cuts to limit the maximum output current and loosely regulate the output voltage based on the name-plate output current.

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:
The dual rectifier idea with AC output transformers isn't a bad idea. Remember, however, that, ideally, the rectified voltage is sqrt(2) times the RMS voltage.

That is my 1.4 in the short calculation I presented. And in case someone wonders, the 0.7V in the calculation is the voltage drop over the rectifier diode.

ka7ehk wrote:
So, a 12V transformer will rectify into (about) +16V and -16V.
I Just did the calculation the other way around. What transformer would one need as a minimum if 15 V DC, no-load is desired? I am not so much concerned about a to high voltage (common LVOs can take a lot), but a to-low input voltage.

The output voltage under load would actually be the no-load voltage times (1 - sqrt(Rt / Rl)). Where Rt is the internal resistance of the transformer, and Rl the load. But since Rt isn't known, and one doesn't find it printed on cheap wall warts, I skipped that calculation, and just expressed my concern that a 12 V AC wall wart might be at the border of what could work with normal LVOs.

I think using low-drop LVOs and a 12 V AC wall wart has a very good chance to work.

Stealing Proteus doesn't make you an engineer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

@SteveN: I am (still) working on a dual lab-supply, using that principle. The single version is finished and works great.
One thing must be kept in mind: the routing of ground. Could be a movie-title ....

Nard

A GIF is worth a thousend words   She is called Sylvia (2018), lives at Mint18.3 https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Carl, Thank you for the offer! :!:

Perhaps I could just drive down and pick one up. :wink:

I don't know how much ripple I can tolerate. I've made the Leap to SMT, well, almost... I decided to build an AVR controlled, SMT fabricated signal generator.

I'm designing a simple signal generator using a Analog Devices DDS chip, the AD9833, (0-12.5MHz, Sin, Sqr, and Triangle outputs). I'm going to use a Mega168 to control it, with an LCD and lots of push buttons for the user interface, no PC required.

My design has an adjustable gain, and an adjustable DC offset, both probably more useful in the audio / sub 1 MHz range, but intrinsic to my design.

I at first thought my new learning curves would be:
1) My first SMT board, (layout, pads, soldering, etc.)
2) My first real circuit above 4 MHz, (almost "RF").
3) My first real AVR project, Butterflies and a few bread boarded projects aside.
4) My first 4 layer board, (possibly, anyways).
5) My first real circuit I can't breadboard prior to committing to a board.
6) My first playing with Variable Gain Amps at 10 MHz.
7) Several other challenges that I'm not yet insightful enough to predict.

Anyways, a "simple" bi-polar power supply was originally one of the trivial parts of the circuit, but I guess that, too, was wishful thinking.

I certainly like the simplicity of two, 1/2 wave rectifiers, with minimul parts count, but adding a switcher chip, or two, to the list of new learning opportunities is certainly doable.

The BEST PART of this entire project, however, is that I don't feel like I'm going it alone. I have the wisdom and insight from many AVRFreaks to guide me!

Thank you, everyone, for the input. Spring break with the family will present a brief delay in my experimenting, as the parts bin and a scope don't travel well...

Carl, let me work on mydesign some more, but I may well take you up on your offer down the road. Fortunately there is no time table for this project.

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:
Carl, let me work on my design some more, but I may well take you up on your offer down the road. Fortunately there is no time table for this project.

JC


Well, if I've got one Wall-wart, I've got fifty of them, and I'm adding to the stock pile all of the time.

I suppose I should round up all of Wall-warts that I can find and categorize them. Then if someone needs a Wall-wart and I have something that fits the need, I can just pass it on to that individual that needs it.

They'll be there when you need them.

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston