DC Rectification: Capacitor Banks & Inductors

admin

What does all these capacitors do, exactly? How's many farads is enough? How does that stored energy release in volts and current? Does input voltage equal output voltage from a bank? Is inrush current the sole limiting factor?Inductors? Im not even sure how these work really, other than it seems like peak voltage and low voltages induce a common voltage in the output? Can one build an inductor from welding leads? How does iron core size effect the functionality? Please don't ask why Im doing this, Im just looking to purify DC current for phun & curiosity.
Reply:It would take way too long to answer all those questions definitively in one post. i would refer you to some sites that offer basic electronics theory so you can read about capacitors and inductors. How many farads depends on the frequency you are trying to filter or let pass. Inductors work on the basic premise that any wire with electricity flowing through it creates a magnetic field around it. When the voltage is removed, the magnetic field collapses and induces a reverse voltage into the inductor. They too are frequency selective.After an AC waveform is rectified, it becomes a pulsating DC waveform. The capacitor across this voltage will charge as the voltage rises, then release its stored charge as the voltage decreases, thus reducing the 'ripple' of the pulsating DC to a smoother, more consistent voltage level.Again, this is the basics, you can find a lot more info by searching the web.PapaLincoln Idealarc 250 (circa 1962)Lincoln Weldpak 155 w/Mig KitLincoln Squarewave TIG 175
Reply:Yeah, I get all that. I've built a few rectifiers, my self. Im looking for more technical explanation regarding welders.I've read several books and many websights on the subject, now Im looking to understand how this all practically applies to welding machines.So assuming you have 60hz rectified to 120hz pulse at ~40v DC, at which point should adding more farads to the bank clash with the law of diminishing returns? How is this calculated? I have seen little in regards to inductor construction for welders? I assume a iron or ferrite core would be ideal, but Im not sure how eddy currents would play into this application.?My goal is a nice pretty line on an occilisiscope.
Reply:Always use capacitors that are rated for double your voltageNever use a dc capacitor on an ac line Never hook up a dc capacitor reverse polarityAny over the above will result if catastrophic failure of the capacitor (an explosion by any other name is still an explosion)Winding inductor's ,chokes and coils is a PITA    My knowledge is limited to ham radio and low current power supplies 5 amp and less and those were done from plansBacked my CATMA over your CARMA oops clusmy me  What would SATAN do ?? Miller Trailblazer 302 AirPakMiller Digital Elite  Optrel Welding HatArcair K4000Suitcase 12RC / 12 VSHypertherm PM-45Rage 3 sawRusty old Truck
Reply:There are pages and pages on the web, of how to construct a linear power supply.I calculated the amount of capacitance to use in the 151 cap mod, by figuring the distributed resistance of the electrical parts and the arc channel.A formula that calculates it as percentages works,and a 37-63 ratio seems to most resemble what many of the welder designers use, is.Time Constant t= R*C.For instance 120A@20v=.166ohms of resistanceAt .1666ohms and a a time constant of .00833 seconds or 120hz you would need 49890mfd or about 50,000mfd per 120A.So it would end up about 40,000uf per 100A.That seems to be the way the manufacturers figure it.Getting zero ripple on a power supply while welding,with any normal power supply isn't really feasible.
Reply:Maybe that's part of the reason why the big DC generators (SA 200's, etc.) are so smooth to weld with.City of L.A. Structural; Manual & Semi-Automatic;"Surely there is a mine for silver, and a place where gold is refined. Iron is taken from the earth, and copper is smelted from ore."Job 28:1,2Lincoln, Miller, Victor & ISV BibleDanny
Reply:Using 3 phases is a big deal, as far as ripple goes.The power never goes to zero,so the arc channel is more stable, and any capacitance added works on 3 times the ripple frequency.That and many of the large mobile and stationary welders are usually running on the low side of their output capacity.
Reply:Originally Posted by BlueweldersThere are pages and pages on the web, of how to construct a linear power supply.I calculated the amount of capacitance to use in the 151 cap mod, by figuring the distributed resistance of the electrical parts and the arc channel.A formula that calculates it as percentages works,and a 37-63 ratio seems to most resemble what many of the welder designers use, is.Time Constant t= R*C.For instance 120A@20v=.166ohms of resistanceAt .1666ohms and a a time constant of .00833 seconds or 120hz you would need 49890mfd or about 50,000mfd per 120A.So it would end up about 40,000uf per 100A.That seems to be the way the manufacturers figure it.Getting zero ripple on a power supply while welding,with any normal power supply isn't really feasible.
Reply:120 amps is about 20 volts while welding.To figure the Resistance you divide volts by amps R=E/I.The conductivity of the arc channel changes with the arc current density.From around an ohm of resistance at 17 amps,to about 0.1 ohms at 200amps.The lower the resistance of the arc channel,the more stable the arc.
Reply:Bluewelders - great info.  I always wondered how to apply those formulas to reality!  Following along on this, here's my view...It's all about the time constant T. T is the time required to charge a capacitor or an inductor through the resistor, to 63 percent of full charge voltage; or to discharge it to 37 percent of its initial voltage.Now correct me if I am wrong... If a 60 cycle AC powered welder has 25 DC arc volts, the voltage output from the full-wave rectifier is peaking at 35 volts and falling to 0 volts 120 times per second.  While the average output voltage is 25 volts, this is pretty lumpy voltage & current to weld with!Let's look at the capacitor RC CircuitIn an resistance-capacitance (RC) circuit composed of a single resistor and capacitor, the time constant T (in seconds) is T = R x C where R is the resistance (in ohms) and C is the capacitance (in farads).e.g. With a fixed resistance, the time to charge/discharge a capacitor is proportional to the size of the capacitor.A large capacitor acts to smooth the ripples in the voltage - preventing the voltage from dropping to zero.Analogy:  Imagine an old style water-wheel 'pump' fitted with buckets to pick up the water.  As the wheel turns, it delivers buckets of water... water / no-water, water / no-water.  Hook a rigid pipe to the pump and the far outlet of the pipe will deliver pulses of water.  A capacitor acts like a reservoir tank inserted into the head of the pipe.  The tank receiving the buckets of water holds them, the water level rising as a bucket of water drops in and falling as the water flows away down the rigid pipe.  The result is a more constant flow of water at the pipe outlet.Ohms Law: Voltage V = Current I x Resistance RLet's consider a welder with an output of say 130 amps at 25 arc volts.  e.g. Typical DC MIG or DC Stick Welder output.Resistance R = V/I = 25 volt / 130 amps = 0.1923 ohmsTime Constant T = 1/f when f is the frequency; f = 120 Hz for full-wave recitified 60 cycle AC current.  T = 1/120 Hz = 0.00833 secondsFrom T = RC we have, C = T/R = 0.00833/0.1923 = 0.0433 farads or 43,300 micro-farads... much as Bluewelders showed in his posts.However, in welding machines, after rectification by the diodes, a capacitor is almost never used alone.  Why?  The initial arc strike with stick is a momentary short circuit.  MIG is often done with in 'short-circuit mode' with repeated short-circuits followed by an arc.  The resistance in those brief short-circuits is very low.  If there is a fully charged capacitor in the circuit, the intial arc strike will cause a huge current to flow from the capacitor and blow the end off the electrode... making starting the arc a real challenge!  Similar problems occur with MIG... as you are sticking the arc all the time in short-circuit mode!"My goal is a nice pretty line on an occilisiscope."  Maybe not...I've welded with big DC batteries; they had a flat line voltage on an oscilloscope - but welded like crap.  I had to add a big inductor in series with those constant voltage batteries to get a decent arc.NOTE: All transformer-based DC welders (stick, MIG and TIG) use an inductor (coil, choke) as an arc stabilzer.Let's look at the inductor RL CircuitIn an resistance-inductor (RL) circuit composed of a single resistor and inductor, the time constant T (in seconds) is T = L/R where R is the resistance (in ohms) and L is the inductance (in henries).e.g. With a fixed resistance, the time to charge/discharge an inductor is proportional to the size of the inductor.A large inductor acts to smooth the ripples in the current - preventing large current pulses.Analogy:  Imagine that previous 'pump' delivering pulses of water.  An inductor acts like balloon inserted into the pipe just before the outlet.  The balloon, receiving the pulses of water, will expand with each pulse and contact in between pulses.  The balloon acts to smooth out the pulses.  The result is a more constant flow of water at the pipe outlet.Let's consider the same welder as before with an output of say 130 amps at 25 arc volts.As before, Resistance R = 0.1923 ohms and Time Constant T = 0.00833 secondsFrom T = L/R we have, L = T x R = 0.00833 x 0.1923 = 0.0016 henries or 1.6 milli-Henries.In transformer-based DC stick-welding machines, after rectification by the diodes, an inductor is almost always used.  This produces a constant current output.  Why?  The inductor is a stuborn animal; it opposes any change in current flow.  A sudden demand for big current (e.g. short-circuit arc strike) and the inductor drops the output voltage - trying to maintain the same current as before.  A sudden drop in current flow (e.g. momentary long arc) and the inductor raises the output voltage - trying to maintain the same current as before.No capacitor is used on the output of a stick welder.In transformer-based DC MIG welding machines, after rectification by the diodes, an inductor is almost always used.  On the cheaper welders, this may be all that is used... not ideal for MIG.  On the better welders, a capacitor is connected across the output of the diodes and the inductor is inserted after the capacitor in the positive lead.  This produces a constant voltage machine - ideal for MIG welding.  The capacitor acts to smooth the pulsing DC voltage and the following inductor acts to smooth the pulsing DC amperage.That's my understanding of it... Rick V 1 Airco Heliwelder 3A/DDR3 CTC 70/90 amp Stick/Tig Inverters in Parallel1 Lincoln MIG PAK 151 Oxy-Acet
Reply:True, though sometimes I wonder if running the choke on the third harmonic might work better.
Reply:Originally Posted by BlueweldersUsing 3 phases is a big deal, as far as ripple goes.The power never goes to zero,so the arc channel is more stable, and any capacitance added works on 3 times the ripple frequency.That and many of the large mobile and stationary welders are usually running on the low side of their output capacity.
Reply:Originally Posted by WarpspeedI agree, less dc ripple, no capacitors required, and no huge inrush or stored energy to worry about.  And the mains voltage regulation is going to be much better pulling a third of the power from each phase, instead of all from one phase.For a constant voltage dc welding process like MIG, a three phase welder is the simplest and very best way to do it.
Reply:A battery is NOT a CV power source. The voltage can vary a lot based on current draw and internal resistance of the  battery. You have gotten some very good responses, but if you think you can get a full understanding of this subject with a few threads on any forum, you need to rethink that. The questions you are asking takes years of study to fully understand, just like no one is going to learn to weld in a few easy lessons. If you are seriously wanting to understand these principles of electricity, start with a few good books. If it is just couriousity, study the responses you already have and try to understand them.
Reply:I wouldn't think a choke behind a battery would do much.With a CV power supply the inductor slows the amperage rise so the small wires don't just bust 1/2" of wire off at a time. The current rises to the pinch point and the wire separates nicely with a small gap if things are right.As mentioned above 3 phase machines have caps on the main transformer. Newer SCR controled machines have large ones and older tech (tapped reactor, resistor primary) controled machines usually had a smaller AC one (usually looked like a 1/2 pint whiskey flask).I've never thought about those caps? I wonder if the deal is to help with delta power (240/277/240)? For most purposes 3 phase incoming is considered to have 1% or less ripple coming out of the rectifiers.MattLast edited by Matt_Maguire; 01-30-2012 at 07:44 PM.
Reply:Originally Posted by Davidbrown...Concerning Rick V, you say you have had trouble with battery welders. Would a battery not be a CV power source? How would an inductor help at all in such a situation? How do you size an inductor? Can you have too much choke? What differentiates the good welders from the cheap concerning choke? What issues occur from running a choke in series with another choke?
Reply:Based on this thread I have High currant Rectifiers for sale; you won't find anything cheaper; also I have Caps and muffin fans.  I have used some of these items to upgrade my cheap Mini Mig; works great; also built DC box for AC Stick Welder, good for 300 Amps.  So let me know.