Finding a Low-Cost Way to Step-Down Voltage for DIY Power Supply

[zoteman94]

Hi.
I have made a 12Vdc 3A unregulated power supply for my projects.
I want to do some electrodeposition, and the 12V are a bit too high (It starts to electrolyze water and the deposited copper breaks very easily).
I'm looking for a way to step-down the voltage without using too much money (electronic components are expensive where I live).

My power supply is a simple
Mains -> transformer -> rectifier diodes -> 4 1000 microfarad capacitors in parallel.
(As I'm not using anything sensitive on the output, I don't care about the voltage ripple).
I'm looking for advice to step-down the voltage to less than 1V without too much power loss. It doesn't have to produce a very precise voltage or be free of rippling. But should not be too inefficient and be able to manage a variable load.

Options that I have considered:
Adding another transformer to the output of my actual transformer:
This option would be pretty nice as it let's me get more current on the low voltage side. However, it's a bit expensive.
Using a step-down boost converter:
Again, seems pretty nice but somewhat expensive.
I ruled out resistor/capacitor/inductor voltage dividers because they have very big losses and I want to be able to change the output load.

I'd appreciate any other ideas. Or help selecting between what I said before.

Apologies for my bad English, and thanks for your help.

 

[tech99]

Hi.
I have made a 12Vdc 3A unregulated power supply for my projects.
I want to do some electrodeposition, and the 12V are a bit too high (It starts to electrolyze water and the deposited copper breaks very easily).
I'm looking for a way to step-down the voltage without using too much money (electronic components are expensive where I live).

My power supply is a simple
Mains -> transformer -> rectifier diodes -> 4 1000 microfarad capacitors in parallel.
(As I'm not using anything sensitive on the output, I don't care about the voltage ripple).
I'm looking for advice to step-down the voltage to less than 1V without too much power loss. It doesn't have to produce a very precise voltage or be free of rippling. But should not be too inefficient and be able to manage a variable load.

Options that I have considered:
Adding another transformer to the output of my actual transformer:
This option would be pretty nice as it let's me get more current on the low voltage side. However, it's a bit expensive.
Using a step-down boost converter:
Again, seems pretty nice but somewhat expensive.
I ruled out resistor/capacitor/inductor voltage dividers because they have very big losses and I want to be able to change the output load.

I'd appreciate any other ideas. Or help selecting between what I said before.

Apologies for my bad English, and thanks for your help.

Can you find room to wind one or two turns on to the transformer? This would give you a small voltage.

 

[zoteman94]

Can you find room to wind one or two turns on to the transformer? This would give you a small voltage.

Oh, I didn't consider that.
I will try. Thanks!

 

[cnh1995]

Oh, I didn't consider that.
I will try. Thanks!

Will a diode in parallel with the load work?? I think it will keep the voltage across the load at about 0.7-0.8 V...Or a zener diode of required rating..But for that I think you'll have to put a resistor (1k will work) in the diode-transformer loop..Diode will act as voltage regulator..Diodes are cheaper too..

 

[zoteman94]

Will a diode in parallel with the load work?? I think it will keep the voltage across the load at about 0.7-0.8 V...Or a zener diode of required rating..But for that I think you'll have to put a resistor (1k will work) in the diode-transformer loop..Diode will act as voltage regulator..Diodes are cheaper too..

Thanks for the idea, but I think that wouldn't work because I need a somewhat high current (an amp or two) and the resistor would limit it.

 

[davenn]

Thanks for the idea, but I think that wouldn't work because I need a somewhat high current (an amp or two) and the resistor would limit it.

yup

I like tech99's idea ... but ideally start with a lower voltage output transformer for a start say a 5V AC output

 

[cnh1995]

Thanks for the idea, but I think that wouldn't work because I need a somewhat high current (an amp or two) and the resistor would limit it.

I think it'll reduce the overall current..But the load current (current through the electrolyte) will be simply load voltage/load resistance.. Now since the load vtg is 0.7-0.8 V (as per your requirement), load current will depend upon that...Resistor is there to ensure proper voltage regulation..But you may need to adjust that resistance (using pot) as per the electrolyte resistance...davenn's idea will rule out all these problems from the beginning..

 

[meBigGuy]

Anything other than a buck converter, another transformer, or some synchronous input switching method will result in wasted power.

Even a buck converter is made a little more complicated because you want a voltage that is below the reference voltage of most controller IC's.

For a simple, but linear example, this linear regulator has a 0.9V reference. You would need gain between your output voltage and the feedback pin to the regulator to get down to a 0.3V output.
http://www.onsemi.com/pub_link/Collateral/NCP5663-D.PDF

Series diodes provide the simplest "almost constant voltage" way to drop the voltage.

Possibly an scr or triac based approach (like a light dimmer) on the transformer primary can lower the voltage? I think a light dimmer would have issues with a transformer (reactive load), though.

 

[jim hardy]

Options that I have considered:
Adding another transformer to the output of my actual transformer:

How about using your transformer to feed an automobile battery charger ? They're available secondhand, but be sure to get a transformer type not a newfangaled electronic one.. Transformer ones are heavy.
I use mine with a Variac...makes a great adjustable low DC source.

Another thought is step down the voltage to your supply. Yard lighting transformers aren't terribly expensive especially at yard sales.

 

[meBigGuy]

Interesting idea. Feed 12AC into a unregulated mains-powered battery charger and you get something less than 1V, probably. But only if it is a simple transformer/rectifier/filter type unit with no regulation. And the current limit, etc doesn't get messed up by the low voltages.

That's just another way to implement two transformers, yes? He basically already built a 3A battery charger (almost, sort of)

BTW, At what input voltage do you get 0.3V DC out?

 

[Baluncore]

Have you considered running 12 identical electro-deposition cells in series ?
They will all have the same current, each will drop the voltage needed to conduct.

 

[NascentOxygen]

You could try a string of 15 or so diodes in series with your cell, short out diodes to adjust the voltage in steps of about 1V. It might be a good idea to keep a minimum load on this, so try some resistance in parallel with your cell.

Note: I'm talking about a string of diodes in series, as distinct from a single diode in parallel.

 

[Svein]

Chack out http://www.linear.com/ . Choose "Power Management" and go from there.

 

[jim hardy]

But only if it is a simple transformer/rectifier/filter type unit with no regulation.

The inexpensive ones just let the transformer's inherent regulation limit current. There's a thermal overload, usually it's what fails and sends them to the trash heap.

Better ones years back employed a ferroresonant transformer that's inherently current limited and will still deliver correct voltage when used at the end of a preposterously long extension cord. You can tell them by the AC capacitor inside..

BTW, At what input voltage do you get 0.3V DC out?

I'll see if i can find my variac and answer your question ...

 

[zoteman94]

Woah, thanks for all the answers!
I just found an old adjustable voltage power supply that works with low currents.
I will take into account your suggestions when I do higher current electrodepositions.