Help with a rectifier

[Harold_V]

Hey guys!
I have installed strip LED lights in the wet bar in my stereo room, but they are broadcasting a serious amount of radio interference. The power supplies (made in China) are solid state (switching amps), which appears to be the source of the noise, as when I use an old magnetic transformer battery charger to power the lights, they're silent.

That being said, I'd like to use some small 12 volt transformers and rectify the output, expecting that that will eliminate the unwanted noise.
I know little about diodes, but I somewhat understand a full wave rectifier, whereby four diodes can be wired to accomplish the task. The problem I have is I don't understand enough about them to determine what I'd need. I've looked at ebay and found a wide selection of 10 amp diodes that will work, assuming they're suited to the task at hand. They are all rated @ 1,000 volts, but I don't have a clue if that makes them unsuited for use @ 12 volts.

Would you please suggest to me what diode would be best suited to my needs? Assuming I'm on the right track, would I have to filter the output, or should I expect the diodes to work silently?

Thanks in advance.

Harold

[ken572]

Hey guys!
I have installed strip LED lights in the wet bar in my stereo room, but they are broadcasting a serious amount of radio interference. The power supplies (made in China) are solid state (switching amps), which appears to be the source of the noise, as when I use an old magnetic transformer battery charger to power the lights, they're silent.

That being said, I'd like to use some small 12 volt transformers and rectify the output, expecting that that will eliminate the unwanted noise.
I know little about diodes, but I somewhat understand a full wave rectifier, whereby four diodes can be wired to accomplish the task. The problem I have is I don't understand enough about them to determine what I'd need. I've looked at ebay and found a wide selection of 10 amp diodes that will work, assuming they're suited to the task at hand. They are all rated @ 1,000 volts, but I don't have a clue if that makes them unsuited for use @ 12 volts.

Would you please suggest to me what diode would be best suited to my needs? Assuming I'm on the right track, would I have to filter the output, or should I expect the diodes to work silently?

Thanks in advance.

Harold

Good Evening Harold
Think back a few months about a power supply conversation

we had. I made the statement that for plating we always

preferred Choppy Dirty Power for better plating.

and you stated you had not heard it explained that way before.

Well now you made an example of what I was talking about.

Ideal for your situation is PURE DC. If It were me, I would

put a couple 12 volt Gel Cells in parallel with a Smart Battery

Maintainer hooked up to it and Wal La. Pure DC., without

any chokes, filter caps, or regulation, hooked up to them, and or NOISE

to worry about.

J.M.H.O.

Ken.

[Harold_V]

I'd find myself "gel cell" poor if I had to do that, and I'd really prefer getting the entire system operational without having to create places to store the added equipment. In one application, it simply wouldn't be possible. There is no room. I also don't want to find myself held captive to yet more gel cells to maintain, as I already have three of them in the alarm system.

There are several circuits involved, and not all in the same room. I tried to simplify the problem by making mention of only the bar--but it's more involved than that, but the solution to all circumstances should be identical. Solve the bar problem and the rest are solved, too.

Right now I'm hell bent on making it work with simple power supplies. If a battery charger can do it (and it does), I'm having more than just a little trouble understanding why I can't do it the way I suggested. I am leaving room for the fact that I don't understand the matter well, however.

Harold

[Steve_in_Mich]

Give it try only be sure you have the DC polarity correct in the circuit. The 1,000 volt rating is probably reference to the piv, peak inverse voltage they will be fine on the 12 volts as long as you size for the correct amperage. The full bridge rectifier will be a smoother DC and greatly reduce radio interference.

A search of "12 volt DC regulated, filtered solid state power supply" will give you many options to choose from.

Ex. a simple filtered and regulated 12V circuit

http://www.instructables.com/id/Make-a- ... er-supply/

[Harold_V]

Give it try only be sure you have the DC polarity correct in the circuit. The 1,000 volt rating is probably reference to the piv, peak inverse voltage they will be fine on the 12 volts as long as you size for the correct amperage. The full bridge rectifier will be a smoother DC and greatly reduce radio interference.

Thanks, Steve. That's what I needed to hear. I'm also considering buying premade full wave rectifiers, which are not all that expensive. Even Radio Shack shows them for about $5 each, rated @ 25 amps. The largest transformer I have is only good for about 6¼ amps (75 watts @ 12V), with the smaller ones rated @ 50 watts (4.2 amps). I have enough to build all the power supplies I'd need.

A search of "12 volt DC regulated, filtered solid state power supply" will give you many options to choose from.

I'd already done that, and understand what I have to do--I was just stumbling over the voltage reference. I figured it didn't matter, but wasn't sure. I know just enough about these things to get in trouble!

Sure is nice to have this site at our disposal when we need help.

Harold

[ken572]

Harold,

If you are going to build your own, then buy some
large computer capacitors (35 to 50 vdc.- 50,000 mf) they
look like large FAT Beer Cans. Also you will need some good
chokes, and some non inductive wire wound resisters. Full
Wave bridges are great but if you look at it alone, on a
scope you will see Pos. wave sign's next one another. You will
need to smooth them out to as close to flat line as possible,
if you want to make them noise free.
If not you will have to spend some money for a good power supply.
You might be able to buy a used Sorenen Q series lab grade supply
on ebay. they are variable voltage and current and regulated where
ever you set the dials at. Most of these are 10 thru 20 amp units.
Ken.

[stevec]

I'm no electronics expert but I would measure the output of the noisy power supplies you are trying to replace.
Once the voltage and amperage required is determined then try a "Wall Wart" of suitable output.
By "Wall Wart" I mean those little square black blocks that plug directly into a 110v receptacle.

[hammermill]

first question, do the led strips have internal current limiting?? a small resistor in line every 3 led? if so good next how much current do they require? effects size selection.
a simple transformer, diode set up and filter capicator should do the trick.. drop me a pm and I will send a supply rated 12v rated comerical supply. have quite a stack from a refit job.

I would also suggest simple shielding ie wrap the supply in foil, and ad a .001 and 1000mfd 35v eletrolitic capicator across the supply output. a second .001 from the positive to chascis/electrical grounding (green) conductor.

to answer the diode question a 10 amp bridge rated 35 volts and up will do the trick, some thing in the parts drawer at the shack.

[SteveM]

Hey guys!
I have installed strip LED lights in the wet bar in my stereo room,

I'm coming over for drinks!

Steve

[scmods]

Harold and others,

All good advice, and the PM should get you a good circuit diagram.

The diode is nothing more than a check valve. Lets the current flow one way only. The "bridge" is essentially a square configuration of four diodes connected into two parallel routes from one corner of the square to the opposite corner. Let's call them 1 and 3. The other two corners of the square occur at the junction of the two diodes making up each route. Let's call them 2 and 4. The AC is connected to 2 and 4, and the DC comes out at 1 and 3.

Since the input AC is a sinusoidal wave, the DC comes out as a "humped" or "pulsed" current. All in one orientation, but varying in strength by a half-sine wave. Much like the output of a piston water pump. Placing a capacitor in the circuit between 1 and 3, provides a storage location for the peak of the pulse that then fills up the space between adjacent pulses, much like a pressure tank or air chamber in a water system. This is my concept of "filtering" the circuit. If you have a friend who has a scope in his "shack" (yes its called a shack, from the old days), he could give you a short course in waveforms that might be interesting.

In sizing components, bear in mind that voltage and current values generally represent the maximum allowable values, and as long as the voltage and current is within these values, even way within these values, they're OK. It's kind of like towing a small utility trailer with a truck. Maybe not ideal from an economy standpoint, but if you already have the truck, it works just fine.

Along the line of chokes and non-inductive resistors. This output is DC, and if filtered, should exhibit very little of the magnetic variations that cause the inductive effect. In short, "DC don't choke". And any resistor is viewed as strictly resistive.

Shielding is always a good idea, especially where radiated energy may be an issue. In fact, placing the power supply that the lights came with inside a grounded metal enclosure may reduce or eliminate the interference present in the first place.

Just a few thoughts.

Bill Walck

[Russ Hanscom]

A grounded metal box and some small capacitors connected from the power supply outputs to ground might be all that it takes to clean up the existing power supplies, might be a good first attempt.

[steamin10]

Had a similar issue wth crane radio circuits. Once coded and triggered, the pocket type transmitters/recievers would stay open and keep the circuit active on low speeds. We had to put in aluminum shields in the case to dronw out the Noise from the power relays, and add capacitors of low value cross the coil leads to let them shut down cleanly without radio 'fuzz' acting on the curcuits.