Hi Amperage Power Supply.....

Hello all,and Happy Easter...I am in the process of building a High amperage variable voltage regulated power supply.The regulated voltage should be adjustable from 14 to 20 volts DC,at 300 amps.The input voltage will be 24 Volts DC,able to handle the load.What I have built so far is a bank of about 15-20 2n3055 Transistors and a large Heat Sink and fan setup.This bank of Transistors is driven by 2 2n3055's,and that is driven by a regulator circuit.I built an LM317 regulated circuit the drives a TIP31,which in turn drives the two 2n3055,then on to the bank.When the whole circuit is not on full load it will regulate a load tester within +or- 3 Volts from input voltage,then the voltage will dramatically drop at around 200 amps to below 12volts.I have tried changing components,and even tried another type of regulator circuit with some success,however the voltage is no where's near where I want it to be.Any help and or design ideas would be much appreciated.......

hi,

Assuming that you have 20 * 2N3055's in final bank and they are carrying a current of 300 amps, I make that 15 amp/transistor.

What gain have you calculated for the 2N3055 at this current also what gain for the 2*2N3055 drivers.?

What provision have you made to balance the current thru the individual 2N3055's.?

Im assumming that you have it all on a 'big' fan cooled heatsink assembly.

Regards

A 2N3055 doesn't have spec's for current gain above only 10A. Its minimum gain at 10A is only 5. So the max base current for 15A might be its max of 7A. Then 20 of them would have a max base current of 140A! The two driver transistors would catch on fire if they tried to pass 70A each! The TIP31 transistor and the LM317 regulator would also catch on fire.

The 2N3055 transistor saturates very poorly. At 10A and a whopping 3.3A base current, its max saturation voltage is 3V.
The 2N3055 is a poor choice for this job.

This power supply doesn't have any negative feedback so its output voltage is not regulated.

I agree with Audioguru, you need to change to different series pass transistor,
perhaps a power mosfet such as an IRF1404. This would require a higher drive voltage to the gate, but the on resistance is pretty low.
I have built a lot of linear power supplies and use the 3000uF per amp of current rule of thumb for filtering.
Another thing that you don't see published much is the current rating of the transformer. That is to say if the load is continous at lets say 200A the transformer better be rated at 200/.6 or 333A. This is because of the charging current on brute force filters is much higher than the load current. The extra rating on the transformer is necessary to prevent overheating of the transformer.

Has anyone considered where the incoming supply is coming from?, an output of 20V at 300A is 6000W, so the incoming requirement will probably be 8000W or so, almost three times more than you can get from a UK 13A 230V socket, and even more than you can get from a 110V socket in America.

As for 2N3055's, it's 'doable', but I would use 60 in the final stages, and lots more to drive them - plus seriously massive fan cooled heatsinking.

As usual in these sorts of questions, no mention is given for the application - which may be VERY crucial to how to do it. Going on past questions electro-plating might be a possibility?.

I think you misread his post Nigel, he said the input voltage is 24 volts, so what he wants is a linear voltage regulator between 14-20 volts at 300 amps. That's a LOT of wasted energy... Why on earth do you need to regulate those kinds of currents?

You need a welding inverter.

Thanks for all the replys,the knowledge on this site is truly amazing,I only wish I was half as knowledgeable as some of you guys.So yes,as you can see I am building a crazy power supply,and sceadwian your post cracked me up cause your right,LOTS of power.....Anyway,a friend of mine contacted me to to some R&D on a power supply that would provide the above said specs. for very powerful CB transmitting equipment.Basically these crazy things put in excess of 10,000 Watts out,so you see the amperage aspect.The power source is two very LARGE 12volt batteries in series.At this point the amplifiers need the 13 through 20 volts at 300 amps depending on how much power these crazy guys want to put out.Now the thing is,there is already equipment out there that does this,hence why I have a bunch of 3055 lying around,but these homeade devices are crude and don't operate that well.I wanted to design and build something that would be more stable.I have looked at these(Link at bottom of thread) By adding these together I should be able to get the amperage I need,but not sure how to keep things linear and stable.Let me know what you think,and thanks again...

**broken link removed**

Rescue1 said:

Thanks for all the replys,the knowledge on this site is truly amazing,I only wish I was half as knowledgeable as some of you guys.So yes,as you can see I am building a crazy power supply,and sceadwian your post cracked me up cause your right,LOTS of power.....Anyway,a friend of mine contacted me to to some R&D on a power supply that would provide the above said specs. for very powerful CB transmitting equipment.Basically these crazy things put in excess of 10,000 Watts out,so you see the amperage aspect.The power source is two very LARGE 12volt batteries in series.At this point the amplifiers need the 13 through 20 volts at 300 amps depending on how much power these crazy guys want to put out.Now the thing is,there is already equipment out there that does this,hence why I have a bunch of 3055 lying around,but these homeade devices are crude and don't operate that well.I wanted to design and build something that would be more stable.

Click to expand...

Well apart from the highly illegal aspect of it, it seems pretty pointless as well? - if you want to transmit long distances get an amateur licence and use a more suitable band?.

And if they really want to do such a silly thing, why not just use a 24V transmitter and use the batteries direct?.

Lead acid batteries are also 2V per cell, if you could access individual cells you could select any voltage between 2V and 24V with 2V resolution - old batteries, particularly LARGE ones, used to have the cells joined together where you could access them.

I have looked at these(Link at bottom of thread) By adding these together I should be able to get the amperage I need,but not sure how to keep things linear and stable.Let me know what you think,and thanks again...

**broken link removed**

Click to expand...

I wouldn't have thought they were any use, I imagine they are designed for switching, NOT linear operation.

First I wanted to apoloigize if I offended anyone with my choice of projects.You asked what the power supply was for and I answered.Now like I said,a friend of mine approached me knowing of my mild knowledge in electronics and asked if I could come up with a better design than what was available.Now is it silly,well I would agree with that,but its not any more silly than some of the projects that float around here,or on the internet for that matter.It is to me kinda cool being able to regulate that kind of power,which is why I agreed to help.I really don't care if he hooks an industrial size sex toy to it,thats his business(Although disturbing..).I am used to small logic circuits,and wanted to build this for fun.I also wanted to learn something,which is why I came here,and why we are all here,to learn.So again,I apologize if some don't agree,but last I checked a power supply isn't illegal.....Now,thanks again to all who replyed,I picked up a few more ideas because of it.If anyone has any other ideas,please let me know,it is much appreciated.....

Why even regulate it? DC from a three phase rectifier is pretty smooth already, just get a heavy duty capacitor array to help smooth it out a bit more and you should be fine. Blowing possibly 3000 watts in a linear regulator is just plain absurd. Building something like that for fun is not really possible, you have to spend huge amounts of money on components that can handle that kind of current, let alone burn it off.