i need about 100amps at 12 volts

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
i am trying to hook up a hifonics 1,000watt RMS amplifier to a subwoofer in the house.

i am trying to come up with a cost effective approach at finding the amperage needed.
i've thought about buying a car charger with a 200amp jump start feature.
i've thought about combining computer power supplys together to equal the amperage needed.

maybe someone can give me a parts list to build my own power supply with a soldering gun and PCB board.

i know that combining computer power supplys is an escape goat.. but i think with the amperage that i need, this option isnt the best financial choice (probably the easiest choice though).

so i started looking into building my own AC to DC convertor that would supply 12volts at 150amps.

i learned that the start of the process comes from a step down transformer that will reduce the 110volts down to 12volts.
although i cant find any that will pass along the amperage that i need for this massive amplifier.

the amp is said to be 800watts at 4ohm bridged
900watts at 3ohm bridged
1,000watts at 2ohm bridged

i intend to use the amp at 3ohms.. so:
900watts @ 12volts = 75amps

if i use computer power supplys.. i would connect the 12volt wire with the 3.3volt wire to make 15.3volts (something closer to the car, and ideal for the amplifier)
900watts @ 15volts = 60amps

now that would be the amount of amperage needed if the amplifier was 100% efficient.. so i need extra, which is why i said a home-built 150amp 110v AC to 12v DC convertor would be good for the amplifier's thirst

do i hunt down a 110v to 12volt transformer.. get some rectifiers and use a kinetik powercell as a stabilizer?
(and if so, where do i find the parts and how do the legos go together?)

the output should be massive.. but i will be using an equalizer to give myself about 80-90dB @ 40-50hz
and then a substantially solid boost at 10-30hz (which is where the power is going to go me thinks)

i'm not an electrical engineer, but i do own a soldering iron and some solder.. i would be willing to follow instructions on how to build my own for a fraction of the price for something already pre-built / intended for something else.

besides, i think the 200amp automotive jump starter/battery charger on wheels go dead after only 3 tries on the key.. so therefore, the unit would drain too fast despite being plugged into an AC outlet.

anyone with some suggestions?
 
i intend to use the amp at 3ohms.. so:
900watts @ 12volts = 75amps

It will only draw 75 amps intermittently if operating on a music signal. The continuous average is probably 20% of that, or less.

if i use computer power supplys.. i would connect the 12volt wire with the 3.3volt wire to make 15.3volts

I don't think you can do that - all outputs of a PC supply share the same ground, can't series wire them. And the 12V supply has a WAY lower current than the 3.3v output.

Such a power supply is not easy to build, nor cheap. Would have to be a switched mode supply, a 100A 60Hz transformer is going to be humongous and heavy and expensive. (think of the size of old fashioned welding transformers)

An easy solution is to use an automobile battery or two, and float charge it.
That way, you use it up for a couple of hours, and recharge it the rest of the day, with a low current charger (10 amps say).
The continuous current draw of the amp is much smaller than the maximum, so the battery will not drain at the rate it would supplying 100 amps.

BETTER solution, cheaper, easier: buy a 1kW sound reinforcement / PA amplifier designed to run off 110V ac.

In the old days before switched mode power supplies, this would be done with a rotary converter: get a big old old car/truck alternator, and drive run it from a 110V AC induction motor.
 
i had no idea that a transformer to fit my needs would be so large/expensive.

i have two power supplies.. one is 460watt and the other is 375watt.
the combined amperage on the 12volt rails equals 62amps.

that 1kW sound amplifier would probably cost at least $600-$800

i dont see why i cant wire the 12volt rail to the 3.3volt rail in parallel.. they are both active and functioning at the same time when the computer is turned on and used normally.

there is an idea to use an AC motor belted up to some alternators on a piece of wood.
 
It can be done but it would be more prudent to buy the correct amp. What you will be doing is starting with 110v, lowering it with a trafo to 9v, rectifying it to 12v DC, and filtering it to a lower ripple percentage, feeding it into the amp, where the 12vdc is converted back into a higher dc voltage to run the amp internally. Off the top of my head the amp would actually run on around 60v to be capable of the rated power. You're talking about a massive power supply with a 1.5KVA 110/9v trafo and industrial rectifier diodes, and huge capacitors. $500 easy.
 
i have two power supplies.. one is 460watt and the other is 375watt.
the combined amperage on the 12volt rails equals 62amps.

Ok, if you have the figures. The ones I've encountered all give big amps on the +5 and +3.3 V, but about 1/3 of that on the 12V rails.
You have to be careful though, parallelling supplies, since they will each put out slightly different voltages, and then fight each other trying to force the output to what each one wants. That will result in current circulating only through the power supplies, wasting a lot of energy.

i dont see why i cant wire the 12volt rail to the 3.3volt rail in parallel..

Because the (-) of the 3.3V rail and the (-) of the +12V rail are internally the same thing; you can't wire the (-) of the 12V to the (+) of the 3.3V
You could do it with the -12v and 3.3V supplies, but I think -12V rails are even lower current capability than the +12V.

A possible solution - but it's not simple if you don't have a good knowledge of electronics - is to modify a computer power supplies feedback sensing, and reconfiguring its 'donut' transformer. Doing that can get the 5V rail to go up to +12V (or more) and still deliver almost the same current that the +5V was originally specified at.

A pair of microwave oven transformers used 'backwards' might work; backwards they'd only give about 6V out I think - hence the need for 2. But the smoothing caps still have to be BIG.
 
Last edited:
i'm tempted to re-wire a transformer and check the voltage output with a multimeter.. but having no money and therefore no room for any accidents.. i am passing on it.

i've just started reading about building my own AC to DC convertor and therefore i doing want the trouble of any errors i might encounter.

i will eventually sit down with the brain power and financial resources to play around.

meanwhile.. this thing looks like one bad mama-jama.
i'd hook up a kinetik powercell to that and then let the amp suck amperage off the two of 'em.

but hey.. if they can do it in a relatively small package (and under $500)
i'm sure i could too.

i read about first winding and secondary winding.. the ratio and such to make it a step up or a step down transformer.
i just dont trust my multimeter and i dont have money for a new one.
nor do i have money for little things like electrical tape.

so i'm just waiting and chatting about it.
 
Because the (-) of the 3.3V rail and the (-) of the +12V rail are internally the same thing; you can't wire the (-) of the 12V to the (+) of the 3.3V
You could do it with the -12v and 3.3V supplies, but I think -12V rails are even lower current capability than the +12V.

A possible solution - but it's not simple if you don't have a good knowledge of electronics - is to modify a computer power supplies feedback sensing, and reconfiguring its 'donut' transformer. Doing that can get the 5V rail to go up to +12V (or more) and still deliver almost the same current that the +5V was originally specified at.

A pair of microwave oven transformers used 'backwards' might work; backwards they'd only give about 6V out I think - hence the need for 2. But the smoothing caps still have to be BIG.

that is a cute little mod to reconfigure the donut transformer to make the 5v rail go up to 12volts and deliver almost the same current that the 5v was specified at.

see.. i figured if i do one +12 and one ground combined with one +3.3 and one ground, there would be a simple addition taking place.
i understand the added current from the +3.3 would only be a fraction (also known as 'only an extra 100watts that the powersupply is helping my amp produce').

your saying if i did the one +12 and one ground combined with one +3.3 and one ground.. WITH ONE POWER SUPPLY that the two rails are going to float/bleed/fight and then shorten the life of something?
(or are you saying that the combining of two power supplys will do that?)

i thought that there would simply be more stress on the filter caps and more stress on the step down transformer.. but nothing cross-eyed unless i started to mess around with the -12 or -3.3 or -5 rails (which i have ZERO intention of doing)
 
see.. i figured if i do one +12 and one ground combined with one +3.3 and one ground, there would be a simple addition taking place.
i understand the added current from the +3.3 would only be a fraction (also known as 'only an extra 100watts that the powersupply is helping my amp produce').

Noo.. If you series-wire two power supplies to get the sum of their voltages, the available current is the smaller of the two individual supplies (i.e., the current of the !2 v supply). Currents do not add in a series circuit.

your saying if i did the one +12 and one ground combined with one +3.3 and one ground.. WITH ONE POWER SUPPLY that the two rails are going to float/bleed/fight and then shorten the life of something?
(or are you saying that the combining of two power supplys will do that?)

You physically cannot do it with a single power supply. The (-) connections of both the 12v and the 3V are the same point electrically.
To do what you want, you'd need to connect the (-) of the 3V section to the (+) of the 12v section. Only possible with 2 supplies.

What i was saying also. if you take two 12V supplies, and parallell wire them, to add the current (amps), still only giving 12V total, they will fight over what the voltage should be. Neither will be putting out exactly 12.000000V.

i thought that there would simply be more stress on the filter caps and more stress on the step down transformer.. but nothing cross-eyed unless

Depends on the magnitude of the voltage difference. At best, the supplies will just run hot. At worst they wil burn out, or less catastrophic, just shut down due to overcurrent protection.
 
That Kinetic thing is just an expensive car battery. And at $600, you'd still save money by selling you exsting amp and buying a mains powered 1000W amp.

the 'car battery' lets out high amperage at almost a dead short they say.
even the cheapest one they offer is 'stronger than 100 one farad capacitors'

besides.. it is sealed up tight so i could re-charge it in the house with no ventilation.
they go on ebay for starting $105
 
I'd forget the computer PSUs myself, you'll need to do a heck of a lot of messing about as has already been suggested :eek: Getting them to current share is a nightmare with special feedback circuitry needed so they both contribute equally, that is unless you stick a dirty great big low resistance resistor on the positive output of each supply to force current sharing (but it's going to be very lossy).

Your best bet if you are determined to do this is either to purchase something already designed for the job - read loadsamoney :( Or find 3 surplus high current 5V supplies & connect the outputs in series. You'll need a diode (100A plus) accross the output of each supply (reverse connected) just in case one of the supplies goes down so they don't want to destroy themselves, but it can be done.

I happen to have 3 x 750W 5V (150Amp) supplies here to test car amps so i know it can be done, you'll need one hell of a decent mains supply though unless they are power factor corrected.
 
You physically cannot do it with a single power supply. The (-) connections of both the 12v and the 3V are the same point electrically.
To do what you want, you'd need to connect the (-) of the 3V section to the (+) of the 12v section. Only possible with 2 supplies.

What i was saying also. if you take two 12V supplies, and parallell wire them, to add the current (amps), still only giving 12V total, they will fight over what the voltage should be. Neither will be putting out exactly 12.000000V.

you have said something here that has struck my interest.
so the power comes in from the transformer and are the +12 and +3.3 rails connected to the same transformer?
because if they had the same electrical connection the +12 rail could suck up all the juice leaving the +3.3 to dwindle down and feed from a (if any) capacitor that is not being re-charged because the +12 rail is being a hog.
yes..? no..?

does the +3.3 rail have its own transformer? (i'm thinking it would be smarter)
because if not.. then the +3.3 would have to have voltage regulation, but using the same current bank.

see i was under the impression that there are more than one current bank to use.
but if the +12 and the +3.3 have the same 'bank account'
it would be wise for me to simply use the +12 rail ONLY and suck the blessid thing with the amplifier.

i guess that question can only be answered by looking inside for more than one transformer though eh?
am i right or wrong about that.
 
does the +3.3 rail have its own transformer? (i'm thinking it would be smarter)
because if not.. then the +3.3 would have to have voltage regulation, but using the same current bank.
No, they are all run off of the same transformer. Generally the output with the highest current is the regulated output, the others including +12 tend to get regulated by the amount of turns on the transformer & the output inductor. They'll be linked on the ground line so can't really be split.

If you increase the output voltage on the main regulated output then all the voltage outputs will generally follow suit, that is unless they are followed by a post regulator such as a power mosfet controlled by other circuitry.
 
As said before, the car amp as an SMPS that takes the 12v input and convert it to 60-0-60v or more to feed the amplifier, so why don't you just open the amplifier, disconnect the smps from the amplifier and use a centertap transformer rated for the current and voltage supplied by the smps originally?
Would be easier and a lot cheaper than your initial idea.
 
I'd forget the computer PSUs myself, you'll need to do a heck of a lot of messing about as has already been suggested :eek: Getting them to current share is a nightmare with special feedback circuitry needed so they both contribute equally, that is unless you stick a dirty great big low resistance resistor on the positive output of each supply to force current sharing (but it's going to be very lossy).

Your best bet if you are determined to do this is either to purchase something already designed for the job - read loadsamoney :( Or find 3 surplus high current 5V supplies & connect the outputs in series. You'll need a diode (100A plus) accross the output of each supply (reverse connected) just in case one of the supplies goes down so they don't want to destroy themselves, but it can be done.

I happen to have 3 x 750W 5V (150Amp) supplies here to test car amps so i know it can be done, you'll need one hell of a decent mains supply though unless they are power factor corrected.

so you are saying literally daisy chaining power supplys in series rather than parallel (just like series and parallel with speakers) is beneficial, but only if the supplys are of the same rating?

i am understanding that there is circuitry with IC chips that could or could not help the situation.

what about the power supplys being ran off the same 60hz wave.
i mean i think that would set the timing of the transformers in sync (especially if they are the same)
and then the rest of the current is direct/constant.. so i am failing to see where/how the circulation would be taking place.

the filter capacitors would heave-ho eachother?
the difference in transfomers would heave-ho eachother?
the (if any) feedback circuitry would cause a heave-ho between the two supplys if wired in parallel?

because this is where i'm confused at.. i dont see the electricity as an ocean of water yet.
 
No, they are all run off of the same transformer. Generally the output with the highest current is the regulated output, the others including +12 tend to get regulated by the amount of turns on the transformer & the output inductor. They'll be linked on the ground line so can't really be split.

If you increase the output voltage on the main regulated output then all the voltage outputs will generally follow suit, that is unless they are followed by a post regulator such as a power mosfet controlled by other circuitry.

this makes me think that the step down transformer drops the 110v down to 5v or 3.3v and then uses another step up transformer to provide the +12v

am i speculating different design options here?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.