power supply help unregulated

[windtech]

OK here is the question i want to build a power supply 12-20 V 250 amp. i am trying to figure the capacitance needed i am coming up with 1.2 F am i close? or is my math off and how do i come up with what capacitance needed and can i use something else other than 25 v cap's. can i use 30 or 50 volt cap's? and what is the difference going to be using a larger voltage cap. help i am no electronic student but i am trying to self learn what i need to work on stuff. schooling would be nice but i cant afford it and I'm getting to old to switch field's. but a better understanding of electronics would help my current field of work. thank you for helping out on an old fool's quest for knowledge.

 

[ericgibbs]

hi,
From one old fool to another...

It depends upon the degree of ripple that the power supply is designed for.

Is it a full wave rectified supply to the cap and what is the supply frequency, ie: 50/60hZ.?

 

[cowana]

1) What ripple voltage is acceptable on the output?
2) What frequency is mains voltage?
3) Is it a switched mode power supply, or conventional transformer?
4) If the latter, what sort of transformer?
5) How did you work out that capacitance??

Andrew

Edit - clashed with Eric...

 

[windtech]

the transformer is out of a welder 230v 60 Hz ac to 24 to 28 volt "dont know yet i still haven't plugged it in to find out yet". an i am going to use a full wave out of skotty diodes. i am trying to run a hf amp 2sc2789 transistors about 12 to 16 of them. so i am trying to get close to no hum as i can

c= 5*I/Vp*f
I is current
f for frequency
Vp peak voltage of capacitance
C is capacitance

 

[windtech]

sorry I'm trying not to go over 18 volts on load and 20 stand by. and whats ripple voltage?

 

[Adjuster]

Even if the secondary voltage is just 24V, you would expect considerably more than 20V on standby. With a simple peak-charging circuit, expect to get over 30V, unless you use some sort of regulating system.

Obtaining the sort of voltage and current you require from this sort of transformer would be a challenging task even for a suitably experienced engineer. You are proposing to build a 5kVA power unit: quite frankly I do not think that this is a good project for a beginner. Apart from anything else, there are significant safety issues.

Ripple voltage is the fluctuation on the DC supply which results from using rectified AC. Filtering is used to reduce this to an acceptable value. It may be what you are referring to when you say that you want a low level of hum.

 

[unclejed613]

i was going to ask what kind of rectifier you planned on using... i guess you meant to say Schottky? you'll have some very large charging currents as well as charging current spikes when the supply is under load. so you'll need a bunch of parallel diodes


what frequency range are you amplifying? the 2SC2789 only has a 20Mhz Ft. it's a switching transistor, not an RF amp.

i agree with Adjuster.... this is not a trivial amount of power, and you won't be getting that kind of power from a wall outlet either. that's about 25A on a 220V line.... you will want a safe way of bleeding the charge from the caps if you need to work on the inside of the supply, as this kind of current and charge capacity, even at 24V is DANGEROUS... as in vaporizing a screwdriver shaft if you happen to short the caps with it, or worse, killing you instantly if YOU get across it' some good places to get safety tips as well as construction details for large capacitor banks and high current supplies like this are builders of rail guns. but be really careful.... these are very large amounts of power for an experimenter to be messing with.... please don't earn a Wile E Coyote award....

 

[Jaguarjoe]

Aren't welding transformers duty cycle limited? A Lincoln buzz box I had years ago was only 10 or 20% on.

 

[windtech]

boy oh boy this is small compared to the voltages that i work with. i work on wind turbines that can produce 1500 volts dc at 1000 amps. ive worked on 2.5 megawatt and 6.8 megawatt turbines. i just dont know how to make a small supply the theory is the same but i dont know about the small stuff. we have engineers for that i all do is change parts. and yes i am high voltage trained. but this is not 220 or 690 or 34500 volts three faze. sorry i blew up.

OK the welder is a Lincoln 250 amp ac arc welder this isn't a small one. this came out of a welding shop that it had been used in for the last 20 years 5 days a week. i know that regulating it down might have to be but how do i do that. i will be checking the voltages to be sure once i have it apart. i dont have the unit yet, i have to get it home and it doesn't fit in the car. ive asked the guys at work and they wont give me the time because I'm a grunt. "i dont have an alphabet behind my name". any help i can get from you guys in here is a help. i really dont know where to start. ive got an idea but thats it an idea. i want to make this for me. help dont help but please dont give me the wrong information or my wife might be filing a wrongful death suit on my behalf.

and the 2sc2789a "pill" is used in the construction of an RF amp i can send you info on it if you want to build one just let me know. ive built 1 to 24 pill amps there not hard at all. i just haven't built one for my self to use. right now i test them with a battery charger and a bank of battery's and that is getting old.

 

[windtech]

OK i did some checking the welder is rated 230 ac at 50amps in 250 amps at 25 volts out at 20% duty cycle. so i should have no problem getting the 100 amps from this. i will have to build it for 250 amps but I'm not going to need more than 100 amps so i should be fine. now how do i regulate down to 12 volts but make it variable to say 18 volts? i realize that i could buy a transformer to do this but this is what i have on hand. it will all so get me closer to knowing a little more about power supply's and how they work. sure beats that 10 amp power supply that you have on the bench. lol

 

[unclejed613]

k, did another search for 2SC2789.....

2SC2789

Si NPN Power BJT
Shindengen Electric Mfg. Co., Ltd.
npn transistors,transistors
V(BR)CEO (V)=400
V(BR)CBO (V)=500
I(C) Abs.(A) Collector Current=8.0
Absolute Max. Power Diss. (W)=80
I(CBO) Max. (A)=100u
h(FE) Min. Static Current Gain=15
h(FE) Max. Current gain.=20
@I(C) (A) (Test Condition)=4.0
@V(CE) (V) (Test Condition)=2.0
t(f) Max. (s) Fall time.=700n
Status=Discontinued
Package=N/A
Military=N

2SC2789

Silicon NPN
N/A
npn transistors,transistors
Category: NPN Transistor, Transistor
MHz: <1 MHz
Amps: 10A
Volts: 500V

japanese type numbers don't get re-used when discontinued

so i did one for a 2SC2879.... that's probably what you're talking about....
got a schematic for the amp?

 

[windtech]

here is a one pill amp

 

[windtech]

here is one all done

 

[tcmtech]

What brand and model of welder power supply was it? I used to work for Praxair as a service tech and have loads of files on welders and plasma cutter power supplies. Odds are would have a compatible voltage regulator schematic that would give you automatic voltage stabilization and regulation even at the high power levels.

 

[windtech]

tcmtech
it is a model Lincoln ac-225-s it is a 225 amp welder 25v wile welding and 79v open voltage. i am going to have to regulate this down to 13.8 if i can but i would like to make it variable to 18 volts for testing some of the larger amps i build. i just got it about an hour ago it was droped off for me.

 

[tcmtech]

Thats the wrong type of welder to be trying to make any sort of stable constant voltage power supply out of. Being constant current and not constant voltage makes it useless for building a stable constant voltage power source out of.

To make a good high current constant voltage DC power supply the power supply's that are in MIG (wire feed) welders are the best choice to start with. They are inherently stable voltage wise and also have substantial current overload capacity.

 

[windtech]

OK i got it open and i see what your talking about its no good for anything that i want to do with it. so drop back and punt i will have to figure out something else sucks that i cant use it but if everything went good i wouldn't have anything to ***** about and my wife would be happy.

well thank you for the help guys sorry for me blowing up. if any one wants plans for building an hf amp thats only good for 10 11 12 meter bands ive got the info right down to the parts lists

 

[tcmtech]

If its built like the old Lincoln 225 where there is a big primary and a big multi tapped secondary winding in line with each other with one or two smaller tertiary windings sitting 90 degrees to them on smaller iron cross sections it may be possible to cut those sideways tertiary windings out and then knock those core sections out that they where wound on which would then turn the constant current transformer into a constant voltage transformer.

I don't know what you would get for an open circuit voltage on the secondary if you did that but I suspect that the voltages across some of the taps may come out closer to what you a are needing.

Its just a thought.

 

[windtech]

If its built like the old Lincoln 225 where there is a big primary and a big multi tapped secondary winding in line with each other with one or two smaller tertiary windings sitting 90 degrees to them on smaller iron cross sections it may be possible to cut those sideways tertiary windings out and then knock those core sections out that they where wound on which would then turn the constant current transformer into a constant voltage transformer.

I don't know what you would get for an open circuit voltage on the secondary if you did that but I suspect that the voltages across some of the taps may come out closer to what you a are needing.

Its just a thought.

ill take some pictures of the transformer tomorrow and ill post them. it does have a set of windings that are 90 degrees.