Highish Current/Lowish Voltage Rectification

Hi all,

I have a welding transformer that produces an output of either 50V AC or 80V AC (open circuit voltage) at a range of currents of up to 200 Amps. I want to convert this to DC, preferably by full wave rectification, whilst maintaining the same range of currents. This will then form the power supply for a dual purpose TIG welder/plasma cutter I'm going to build. Any suggestions as to how to go about the rectificaiton? I've seen a lot of 50A bridge rectifiers going cheap on Ebay. Could I use four of them in some sort of combination to get the desired result? 200A versions are much harder to find and cost far more proportionally.

cheers,

cd/pb

Reply to
Cursitor Doom
Loading thread data ...

Den onsdag den 23. april 2014 00.14.45 UTC+2 skrev Cursitor Doom:

I've seen people use 50A rectifiers in parallel for diy welders run a separate wire to each and make sure the wire has enough resistance to balance the differences

-Lasse

Reply to
Lasse Langwadt Christensen

For Silicon, you'd like around 0.25 volt drop in each ballast resistor. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

You can parallel bridge rectifiers. At high currents, the diodes get ohmic and have positive TCs. Heat sink well and derate a little maybe. BRs are cheap.

In think a DC welder needs a big inductor somewhere, too.

--
John Larkin         Highland Technology, Inc 

jlarkin att highlandtechnology dott com 
 Click to see the full signature
Reply to
John Larkin

Den onsdag den 23. april 2014 03.02.27 UTC+2 skrev John Larkin:

if it is a welding transformer is is probably already there

stick and tig is constant current

-Lasse

Reply to
Lasse Langwadt Christensen

Can you explain the semiconductor physics behind a diode becoming ohmic at high currents? Or cite a paper indicating this?

Rule of thumb is diodes not on the same wafer will current hog, hence Jim's suggestion of a ballast resistor.

Note another alternative would be to go with synchronous rectification.

Reply to
miso

Even with good current sharing you should use at least 6 with ballast resistors. It is not good to run things at rated values.

?-)

Reply to
josephkk

Don't forget that a welder produces HUGE voltage transients that can zap your diodes. Don't forget reverse voltage protection.

Reply to
mike

The physics is obvious: the PN junction voltage goes as log(i) with negative tc, and it's in series with ohmic parts whose voltage goes as (i) and have a positive tc. At high currents, the ohmic stuff must win.

Just look at any good data sheet for most any diode, i/v curve over temperature.

There are small schottky diodes that have zero tc at around 10 mA. There are PN power diodes that have zero tc at 150 amps.

"Rule of thumb" = "urban legend" = "old wive's tale". For people who don't understand how things actually work.

--
John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    
 Click to see the full signature
Reply to
John Larkin

Also, havving all thaose ballast resistors in parallel will reduce 1^2R losses.

--
umop apisdn 


--- news://freenews.netfront.net/ - complaints: news@netfront.net ---
Reply to
Jasen Betts

Nope. You need a minimum voltage drop across the resistors for the stabilization to work.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

All high-current rectifiers are made with the equivalent of emitter degeneration (i.e. series resistance is built in) to prevent hot spots. Either ballasting (milliohms or microhenries) of multiple rectifiers, OR substituting an SCR (different physics, not as susceptible) instead, is the old recommendation for 100+ amp circuitry.

If I could find my trusty Sorab K. Gandhi "Power Semiconductors" book, probably that's a reference...

Yeah, there's some BIG IGBTs available. The old-school approach is a shaft/commutator/brushes run by a synchronous motor, that DOES work too.

Reply to
whit3rd

If you parallel a bunch of small bridges, the wiring helps equalize currents a little. A bigger problem is to get the heat sinking right.

--
John Larkin         Highland Technology, Inc 

jlarkin att highlandtechnology dott com 
 Click to see the full signature
Reply to
John Larkin

I haven't done a lot of high power stuff myself, so I don't know how far one can rely on the silicon and wiring resistance to equalize the currents.

I agree that there's a lot of poorly-supported dogmatism about current hogging, but in the scheme of things, it's a drop in the bucket. :(

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

Except current hogging is real. You haven't really explained much of anything here. Then again, I didn't have high hopes.

Reply to
miso

Current hogging is real. But the deal is you just buy a part rated for the current that you need, and the problem is solved. Current hogging is less of an issue on silicon than with discretes.

I've never understood the mentality of using the wrong part in a circuit. People who do so are the first to complain about the parts not working, when what isn't working is the stuff located between their ears.

Here, let me put this shock for a F150 in a Mack truck. Hey, a shock is a shock, right.

Reply to
miso

OK, here's a randomly selected power diode.

formatting link

See fig 1.

As the current goes up, the forward drop TC goes down, passes through zero, and the TC eventually becomes positive. Some diodes will reach zero TC at reasonable operating currents; high peak currents, as when rectifying AC, help push the diodes into their ohmic regions.

That effect mitigates classic thermal-driven current hogging. A little wire drop helps, too, since copper wire has a healthy positive TC.

You can parallel bridge rectifiers, especially if you do the heatsinking and wiring right.

--
John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    
 Click to see the full signature
Reply to
John Larkin

Are you sure the AC arc welding V/I load curve is optimal for TIG welding ?

In a bridge, there are two diodes conducting at the same. The voltage drop is about 1 V and at 50 A, the power dissipation is 2 x 1 x 50 =

100 W.

I just wonder how these "35 A" or even "50 A" bridges (plastic 3x3x1 or 4x4x1 cm with a hole in the middle for the mounting screw) sold on ebay or hobbyist shops are going to dissipate 70 to 100 W into the heatsink and into the ambient.

Using stud mounted individual diodes will simplify the thermal design and paralleling diodes. If you can get diodes with anode on the stud as well as cathode on the stud, you do not even have to isolate the individual diodes from the heat sink, just isolate the positive and negative side heat sink from each other and from the environment.

If you have to parallel several diodes, use separate wires from each diode isolated terminal to the transformer secondary terminal. Thus these individual wires work as current sharing resistors.

Reply to
upsidedown

On a sunny day (Thu, 24 Apr 2014 09:54:11 +0300) it happened snipped-for-privacy@downunder.com wrote in :

Those ratings are at 25 degrees C case temperature, that you will not be able to maintain, unless on a near infinite heatsink.

Right.

I remember we had a copper pipe with water flowing in it and high power germanium diodes... in the sixties.

Reply to
Jan Panteltje

Usually at 55-65C or something like that. RTFDS.

formatting link
GBPC3508, 35A DC @ 50C.

The thermal conductivity isn't very good (they're potted in black stuff, like a fullpack transistor), but good enough I guess.

I've personally cooked off only one rectifier like this. It wasn't cooled well, and it was easily over ratings: I was doing more like 30A DC on a cap-input rectifier. Only one diode actually failed -- the opposite diode apparently survived despite the ~1kA, 10ms surge that cleared the breaker. Diodes are pretty damn beefy.

Tim

--
Seven Transistor Labs 
Electrical Engineering Consultation 
 Click to see the full signature
Reply to
Tim Williams

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.