ECN Forum Forums General Topics of the Trade General Discussion Area Elementary Q... DC Xformers?

The easiest way to do some low power voltage reducing, is with zener diodes. Like Paul says, Which are fancy semiconductors that only pass the voltage that they are rated for. Resistors would work, if the current doesn't change, and you have the heat factor too. If the current changes, ie, a low battery, high charge, ohms law kicks in and your voltage drops. That doesn't happen with a zener. If you look in one of the "battery eliminators" from the "shack", the one with all of the adapter plugs, and a slide switch for which voltage your portable cd player runs on, take one of those apart, and each position of the slide has a zener diode for the corresponding voltage. The heat dissapation of the zener is the limiting factor as to how much current it can carry. These things are available in some really big packages, with heat sinks and the like. Motor T's got it right, dc and transformers aren't compatable. If you really want to get complicated, you can go with some Motorola optocoupled devices, that use light to change things around. I have some in the 32vdc to 12vdc converters that run the radio's in our locomotives(these things have a 15amp output). If you find a "transformer" in one of these battery eliminators, it's probly a "choke" coupled with a capacitor to filter out the ac noise from the dying alternator in the truck
It's been awhile since I played with this end of things. The dust on the file was pretty thick.

Trainwire

[This message has been edited by Trainwire (edited 09-25-2002).]

Perhaps I should add that the simplest arrangement from the point of view of wiring up a DC voltage reducer such as this is to use a voltage regulator I.C. (integrated circuit, or "chip").

The fixed voltage versions have just three terminals. You can no doubt guess what they are, but here goes anyway: Supply in, regulated output, and common ground. So long as you stay within the rated supply range and maximum load, they work fairly well. Some higher-power types will need to be heatsinked, and some benefit from a pair of decoupling capacitors to prevent self-oscillation (and to filter out generator and ignition hash in automotive applications, as has already been mentioned). These chips are basically the zener diode and a more sophisticated version of the series-pass arrangement described above.

Some of the cigarette-lighter adapters use these 3-terminal regulators.

On the subject of switched DC into a transformer, the old vibrator supplies that were once used to power tube radios in cars used this principle. The vibrating contacts interrupted DC power to the transformer primary winding, and the secondary output (far from a perfect sinewave!) could then be rectified back into DC and filtered.


[This message has been edited by pauluk (edited 09-25-2002).]

Virgil,

As you can tell, there are quite a few extremely valuable replies from the "Tech Gurus" here at ECN, which not only cover your original posted Q, but added more scenarios.

Great job(s) as usuall!!!

Glad we have this vast pool of knowledge here at ECN to assist when needed!!

Just to add some more info to this topic:

* Simple methods of dropping DC Voltage:

• Dropping Resistor(s),
• Voltage Dividing Network (an array of resistors),
• "Tapped" Power Resistors,
• Diode or Diodes in series with the load.

With exception of the Diode(s), these methods drop DC Voltage by Dissipating True Power and therefore produce large amounts of heat and are considered "wasting power".

The Diodes simply have a Forward Voltage Drop, which makes the output voltage lower than the input voltage (please excuse the extremely basic description!)

* More advanced and precise methods are:

• Voltage Regulators - either descrete Power Transistors with a "Fixed" Base current control, to IC type "Flying" Regulators,
• Switch Mode Power Supplies - from simple to complex types,
• Series Subtractive connections of power supplies.

For really extreme and precise stuff, this brings in DC to DC conversion - Inverting the DC into AC, then either stepping up or lowering the Voltage, then Rectifying it back to DC again.

FYI, some methods of increasing DC Voltage:

• Voltage Doublers - Tank Circuits / RC Networks (RC = Resistor / Capacitor),
• Dc to DC Converters,
• Induction coils driven by pulsed DC (DC switched on and off rapidly, such as an Automotive Ignition System),
• Series Additive connections of power supplies (dry cells, Batteries, Generators, etc.).

As to driving a Transformer with pure DC, that won't do very much after the primary becomes stable. Only while the primary is "Building Up" a charged state when the power is applied, or while the primary is "Decaying" it's charge when the power is removed, will there be a usable current flow on the Secondary side.
When the Primary is at "Steady State", there may be a very small current that can be found on the Secondary side, which can be very transient at times. This current is not very usable for a power application, so it's generally neglected as being present.

If the DC is coupled with an AC signal, then you can do a variety of things now!

On the contrary, having large levels of DC coupled with the AC output of an Audio Power Amplifier is a concern for good crossover design and loud speakers.

OK, I feel that I have gone off-topic far enough now!

Scott S.E.T.

p.s. how did everything turn out with that flame war situation regarding lethal shock levels and Transformer KVA ratings???
By the time I saw the thread, everyone at ECN had posted info to it, so I did not think I should add more [except for support!]

S.E.T.

The best way (i.e. the best for me) to lower DC volage is to use a zener diode connected to a transistor. This will allow you to use higher current devices, since a largeer transistor can dissipate more heat than a small zener diode. It's usually cheaper than buying a large zener diode rated for the load. Of course, for 5,6,8, and 12V, I just use a LM7800 series IC. It has 3 pins: In, Out, and Ground. It's usually the cheapest way to go when building your own power supply.

About the only time I will use a zener shunt regulator is with a light, rather constant, load. Otherwise, you have to pick your series resistor for minimum RL & your zener wattage for an open RL. You still need filter and bypassing caps. (El cap & ceramic disc). Rather than using a zener to drive a series pass transistor, and still having to worry about biasing requirements, I consider the three terminal regulators. PE mentioned the LM7800 series, which are the same as the LM340- series.(LM7805=LM340-5) The LM317T is a great 3 terminal regulator if you wish to have a variable supply from about 1.2 volts up to slightly less than your input supply. You just have to watch if you are in a 7800 mindset because the rascals switched from the input, ground, output, pin configuration of other positive regulators, to adjust, output, input. I love this chip because I can look in my drawer to see what value linear pots I have, decide my supply range, and have a regulator designed within 5 minutes.

Folks wanting supplies slightly higher than common fixed regulators can also cheat and think of the "gnd" pin of a fixed regulator, as the "adj" pin. Conversely, you could think of an LM317T as a fixed 1.25 volt regulator by placing the "adj" pin at ckt gnd, instead of in the middle of a voltage determining divider. The one warning I give is to place at least 1uF of capacitance on the output of LM317 supplies because small amounts of load capacitance will cause it to oscillate and not regulate.

As an example of using a fixed regulator for a higher voltage, suppose you want a 13.8 voltage regulator. (Not out of the realm of possibilities.) You happen to have an LM7812 that you measured to have a 12.00 volt output. If you now place a 12,000 ohm resistor between output and ground. This should establish exactly 1mA through the resistor. If you now place exactly 1800 ohms between the ground pin and ground, the entire 1mA,practically, will flow through the 1800 ohms, dropping 1.8 volts across it. I say practically because the ground or adjustment pins have extremely high input impedances and can be negated when considering other tolerances. So now that LM7812 can be the heart of a cheap float charger. In the real world, there's a cap across the 1800 ohms, which might really be a 2k pot, or you might have a switch to 2400 ohms for cycle charging. There would also be a couple normally reversed biased diodes in there to protect the chip from its load or cap discharge paths when things get turned off.

As always, feel free to email me with any specific questions. The datasheet for the LM317 gives alot of good applications though.
Joe


[This message has been edited by JoeTestingEngr (edited 01-02-2006).]

A note on the LM317 is, that it is limited to 1.5 Amps max. There is an LM350 rated at 3A and an LM338 that is rated to 5A. Unfortuneatly these are linear regulators so all of the overhead voltage is dissapated as heat. These units can also get kind of bulky.
If you wanted to roll your own, there are many types of DC to DC converter modules on the market. These are all different variations of switching power supplies. ie Buck, Boost, Flyback and DC-AC-DC transformer.

[This message has been edited by IanR (edited 01-02-2006).]

I agree with what IanR has stated. The reason that I gave the example that I did was that the parts are available at the local overpriced electronics store.(RS) Those higher current linears aren't as easy to find. The choppers and boost/buck converters aren't as intuitive to understand and the inductors or getting coil forms to wind them starts getting complicated. Even the diodes might have to be fast recovery types instead of what you can find in the blister packs. So I can agree with Ian while not thinking that it is a path most here would wish to follow.
Joe

Now you are talking. The 3 terminal regulator is the easy way to go. Cobbling something up is a science fair project. Digikey has a 3 terminal regulator for just aboput anything you need right off the shelf. GPS and cell phones are pretty modest loads.

The 3 terminal regulator is the best way to go, If more than 1.5 Amps is needed I put a 2N3055 powertransistor in as the power device which will easily take up the load to 5 Amps as long there is enough heatsinking, the e.g. LM 317 does the regulating, the 3055 does the hard work.

If I can find the schematic in my 'messy' workshop i will scan it in and sent it to one of the moderators to insert.

The tidiest way is to insert it in a little project box with the heatsink on outside. be sure that the heat sink is clear of any metalwork because it's at the collector level of the 2N3055.

As what JoeTestingEngr says a 1µF cap at the output is a must to avoid oscillations in the regulator. prefably a tantalum cap.

rodalco that is what they make insulators for.