|
|
Yep, this is the Successful Sequel to the original posting, Welding With Car Batteries - http://www.weldingweb.com/vbb/showthread...ight=batteries.This sequel deals with quality welding with just Two 12-volt lead acid batteries BUT using an ARC STABILIZER!There are two main problems when trying to weld with just two 12-volt batteries:1 - The arc is very difficult to start.Striking a new rod creates a brief short circuit, but with a battery power supply this flows a lot more current than a typical arc welder whose constant current design tends to limit the current to a set value. In comparison, batteries deliver high current that tends to blow the end off the rod! leaving the metal of the rod tip recessed inside a non-conducting sleeve (with 6013, 7014 and 7018 rods). The gap between the recessed rod tip and the work piece cannot be jumped with an open circuit voltage of only 24 volts from a battery power supply unlike a typical welder that may have an open circuit voltage from 40 to 80 volts and be able to jump a sizeable gap maintaining an arc. Once the arc is out, one has to break off the surrounding sleeve, usually by scrapping (dragging) the rod tip along a concrete floor.2 The arc is difficult to maintain. e.g. With 3/32 inch diameter Lincoln 7018AC, one has to weld by repeatedly moving the rod IN/OUT. IN - for an instant short circuit and then OUT to prevent sticking the rod to the work piece do this in/out (push/pull) as you drag the rod along the work piece. You kind of feel the rod tip mush into the work (arc is very short) and pull back slightly (arc is longer but in danger of going out) to prevent fusing the rod solid to the work piece then repeat. It is a dance between sticking the rod and the arc going out. Arc stoppages are frequent - in my experience, about 8 times on a 2.5-inch-long weld! Each stoppage requires breaking/scrapping the sleeve off the rod tip. It is a real pain in the butt trying to weld like this!The solution to both problems is an ARC STABILZER!An arc stabilizer is an inductor (a coil of wire) inserted in series with one of the welding cables.An inductor opposes any rapid change in current through itself.3 Starting the ArcWhen you strike an arc, there is a short circuit between the rod tip and the work piece. With a battery power supply, a large current rushes through the welding cables usually blowing the tip of the rod. However, with the proper size inductor, that large rush of current is prevented and goes into creating a large magnetic field within the inductor. The arc strikes more normally (with less current) and a reasonable arc forms. Note: As soon as the arc is first established, the stored magnetic energy in the coil from the arc strike is released as an increase in voltage and current through the rod thus maintaining the arc.4 Maintaining the ArcAs you weld, the rod burns down and the distance between the rod tip and the work piece increases and you have to move the rod in closer to the work. Since no welder is in perfect control, the length of the arc varies and this changes the electrical resistance of the arc a longer arc having greater resistance and requiring a larger voltage to maintain a constant welding current. Arc welders are designed for this, being constant current machines they adjust the voltage to maintain a constant current across the arc. Also, a typical arc welder has a 40 to 80 volt open circuit voltage that is able to jump a substantial arc gap.With a battery power supply, you have a constant voltage machine of only 24 volts, so welding current varies greatly with the length of the arc. However, with the proper size inductor, the large variations in current with variations in arc length are prevented; the inductor storing magnetic energy on short arcs delivering less voltage and less welding current; the inductor releasing stored magnetic energy on long arcs delivering more voltage and more welding current. In short, the inductor acts to stabilize the arc and acts to maintain the arc.BUILDING THE ARC STABILZERI figured that with only 24 volts, I did not want to drop any more than 0.5 volt (at 100 amps) across an inductor/coil. I had good experience with an arc stabilizer using 50 feet of 3/8 inch steel cable that worked well with three 12-volt batteries, so I figured that something like 40 feet of #1 gauge copper wire would do. I modelled various coils using a formula for multi-layer wound coils and it seemed that a 3 inch long form, 6 inches in diameter with a core of 2.6 inch diameter with about 5 layers of 6 turns each would do it produce a coil with an inductance of ~70 microHenries in air. Stuffing the core with steel, I expected this to increase by 400 times!I didnt want to buy 40 feet of #1 gauge wire. Rummaging in the basement, I found a couple of chunky coils of #15 gauge enamel coated wire that I wound 25 years ago! [Keep your Junk!]The total wire available was about 1,000 feet. So, I used that I paralleled 26 stands, 40 feet long, of #15 gauge wire. This gave me the equivalent resistance of 40 feet of #1 gauge wire R=0.005 ohms (Measured as 0.00492). With 100 amps of expected welding current through the coil, the power loss would only be I^2R = 10,000 x 0.005 = 50 watts and the voltage drop V = IR = 100 x 0.005 = 0.5 volt.However, when I built the coil using an H form of steel, I didnt get anywhere near the expected 400 times increase by stuffing the core with steel. Instead I got only a factor of 2!I made a bad design! My air core (theory) of 70 micro Henries measured as only 138 micro Henries.Note: Playing later with other small coils, I could obtain a factor of 5 or 6X using steel to fill the core and a factor over 12X using ferrite in the core far better than my measly 2X! Ok, since I used 26 parallel enamel coated #15 wires, I modified my wire groupings making two bundles, each of 13 wires. In parallel, the two bundles measured inductance was 138 micro Henries and resistance was 0.005 ohms. In series, the two bundles (twice as many turns) measured inductance was four times as high at 490 micro Henries - but the resistance was also four times as high at 0.02 ohms. At 100 amps welding current, the higher resistance would cause of power loss in the coil of I^2R = 10,000 x 0.02 = 200 watts and the voltage drop V = IR = 100 x 0.02 = 2.0 volt. That 2 volt drop looked significant meaning my welding voltage would drop from 24 volts to only 22
and that looked really marginal.DID THE ARC STABILZER WORK?***** You bet! *****Note: Batteries were series connected with a 5-foot length of #4 gauge wire. Current from the negative terminal of the batteries to the work piece was through a 20-foot length of #1 gauge wire. Current from the positive terminal of the batteries to arc stabilizer was through a 20-foot length of #1 gauge wire. Current from the arc stabilizer to the rod holder was through a 5-foot length of #4 gauge wire. In all three cases below, the weld work pieces were 2.5-inch-long T configuration (for filet weld) of 1/4 inch-thick mild-steel plates.A)Two 12-volt batteries - 24 voltsAs described, this was a real pain starting the arc and trying to maintain it was a frustrating experience. After several (~8) arc stoppages, I managed to complete a 2.5-inch-long filet weld. Weld quality was so-so.B)Two 12-volt batteries and a 138 micro Henry coil dropping ~0.4 volts at 80 amps.This was pretty good. Starting the arc was much easier. Maintaining the arc was not bad but I had to restart the arc twice during an identical weld. Weld quality was not bad at all.C)Two 12-volt batteries and a 490 micro Henry coil dropping ~1.6 volts at 80 amps.This was surprisingly good even with the ~2 volt drop in welding voltage. Starting the arc was easy. Maintaining the arc was simple; I did not have to restart the arc at all during an identical weld. Weld quality was very good. This test was surprising in that it showed that upside of having more inductance was of greater importance than the downside of having more resistance. Where to go from here?It is obvious that increasing inductance is the road forward.I need to re-design and remake my poor H-core design. The magnetic field is being lost in the sides of the H originally made of steel to hold the wire turns in place; this need to be a non-conductor and non-magnetic. If I can find and use ferrite instead of steel for the core, my existing coils should be capable of 5-fold increase in inductance. That would mean that my parallel wound inductor would go from 138 to 690 micro Henries and my series wound configuration should go from 490 to 2,450 micro Henries. Recall, welding was good with ~500 micro Henries.Pretty neat stuff! I had a lot of fun!Rick V
Reply:Very interesting, Rick V. Thanks for the detailed analysis and discussion of your experiments. A couple of questions come to mind.Where in the world did you come up with an "H" core in steel? Is this made up of steel transformer core laminations? (Just curious why you chose that design. I'm not tempted to try it as you already told us it was not an efficient configuration.)While we know that the conventional wisdom mandates use of insulated (varnished) transformer steel laminations for a low-frequency inductor core, for a quasi-DC application like this is there any reason one cannot use plain steel bars to form an inductor core? My understanding is that the primary reason for using laminated (silicon) steel as the core is to avoid eddy currents in a solid steel core induced by the fluctuating magnetic flux. If the only AC components in the magnetic flux are due to current transients during the initiation and termination of the arc, and the fluctuations incidental to operator technique, would eddy currents in an uninsulated core really be a significant energy loss, heat problem or "Q" killer? Or is part of the problem in using ordinary hot rolled steel as a core that mild steel just doesn't have the properties to do a good job as a magnetic core?awright
Reply:Hey awright - nice to hear from you...as always.Where in the world did you come up with an "H" core in steel?The 'H' core...I had to make a spool to hold the coil wire. I didn't have anything suitable so I went 'a-welding'. I welded four 1.5 x 3 x 3/32 pieces of steel together to form a square core/hub 1.5 x 1.5 inches square and 3 inches long. Then I welded up 4 pieces of the same material to form the square sides ~6 x 6 inches (left side of the letter H, and right side of the letter H) with a hole matching the core in the middle. I welded the sides to the core - forming a flat sided wheel - looks like an 'H' - a square wire spool viewed on the side view, wires are wrapped around the horizontal bar of the letter 'H'. Anyway, I did it to make a rigid spool. My wires didn't like bending 90 degrees around the square hub - so I built it up with layers of tape and strips of steel on the four sides to approximate a round core.Later, playing with small coils, I discovered that the magnetic field formed in the ends of the core flows into the H sides and this is not so good for getting high inductance values. e.g. I should have made the form out of plastic or wood - then loaded the hollow core with steel/iron. Mine - you could load the hollow core and it made zip difference.Is this made up of steel transformer core laminations?Nope - mild steel... as described above.is there any reason one cannot use plain steel bars to form an inductor core? I didn't have a suitable junk transformer around; I used mild steel bar stock.Eddy currents - I agree with you... since we are not running 60 cycles here, I don't expect much eddy current heating, as in "the only AC components in the magnetic flux are due to current transients during the initiation and termination of the arc".Yep, I think " mild steel just doesn't have the properties to do a good job as a magnetic core" for this application. Playing with several smaller air wound coils, stuffing the core with steel would get me 5 to 7 times the inductance with the core empty (air). A couple of 6 inch long 1/4 inch diameter ferrite antenna rods (Old am antenna rods radio's) would push the inductance up 10X - and the rods were filling only say 20- 30% of the core.Anyway, I am checking into core materials now - Internet is great for looking for tech info! This application is a mite different and I'm not certain what will work best in reality. The ferrite looks good on the BK inductance meter but what happens when one wacks 100 amps around it? If a ferrite core is going to saturate, then it kicks back into being like air - and one looses the high inductance seen at the very low test currents used by the inductance meter.I want to run this down; this welder runs real sweet and could be sweeter yet! Hey every time two fellows (or gals) get together, there are likely two car batteries just waiting to be used! I am quite amazed that the arc stabilizer has such a positive effect of the arc quality from the batteries - turning the battery welding from next to useless into quite a viable approach.This this thing has real potential! It just need some optimization.Awright, if you got some pointers or suggestions, I'm listening.Rick V
Reply:Maybe a suitable core material a ferrite toriod core ( donut shaped ). I have used them in high powered RF amplifier projects in my Ham radio days. see the link. http://hyperphysics.phy-astr.gsu.edu...ic/indtor.htmlNot a beginner, not a pro !
Reply:Well the mystery is solved.. Just dont do it near me.. Just for sh!+s and giggles..Who is gonna bust this myth... and show results..Not me.. I've done my duty Lets go...your turn.. ...zap!I am not completely insane..Some parts are missing Professional Driver on a closed course....Do not attempt.Just because I'm a dumbass don't mean that you can be too.So DON'T try any of this **** l do at home.
Reply:Very interesting read, thanks for sharing. Can we see some pics of the accomplished work?John - fabricator extraordinaire, car nut!- bleeding Miller blue! http://www.weldfabzone.com
Reply:Hey Folks,Ok before I destroy and rebuild the existing arc stabilizer, I thought I should take some photos. Pic1 shows the arc stabilizer in use, the coil-wire bundles configured in parallel to acts as ~31.5 turns of 39 feet #1 cable with a measured resistance of 0.005 ohms and an inductance of 138 micro Henries. Pic1Pic2 shows the arc stabilizer with the coil-wire bundles configured in series to acts as ~63 turns of 78 feet #4 cable with a measured resistance of 0.020 ohms and an inductance of 490 micro Henries. Below the arc stabilizer are the first three weld specimens: left = coil in series, middle = coil in parallel, right = no coil. So, heres some closer up pictures of the welds.Pic3 - Weld with coil-wire bundle in seriesPic4 - Weld with coil-wire bundle in parallel Pic5 - Weld with no coil ... continued on next post (photo limit 5 pics per post)Rick V Attached Images
Reply:Contunied from previous post...Before destroying/rebuilding the existing arc stabilizer, I decided to repeat the whole exercise because the first welds looked pretty sad in the photos. Also, I wanted to know what sort of welding current I had - without and with the arc stabilizer from two 12-volt batteries in series.Measuring the current:Im using #1 gauge booster cables. According to the charts, the resistance R of #1 gauge wire is 0.1264 ohms per 1000 feet. That means that 9.91 feet of wire has a resistance of 0.00100 ohms.Recall Voltage V = Current I x Resistance R.If we pass 1 amp through that resistance, we will create a voltage drop of 0.001 volt (1 millivolt).More importantly, a welding a current of 100 amps in the booster cable will create 100 millivolts on a voltmeter. So, we can read millivolts just as if it was amps of current. So, I connected the leads from a digital voltmeter into the #1 gauge ground cable at two points 9.91 feet (~9 ft 11 inches) apart just pushed the needle-like tips into the booster cable (see Pic6). Folks, I found that I could not weld and read the voltmeter at the same time
so my wife bundled in her winter coat, suffered the cold temperature in my frosty-breath garage, and called out the voltage readings while I welded! Thanks Honey!Ok, so welding:Pic7 - No CoilAs before, it was stab, mush the rod and pull back repeated ad infinitum with several breaks to scrape the rod tip on the floor to remove the non-conducting sleeve. Amperage (millivolt) readings were all over the place but the average majority were in the range of 90 to 135 amps. Pic8 - Coil in parallel ~31.5 turns of 39 feet #1 cable, resistance of 0.005 ohms, inductance of 138 micro Henries.The arc was much easier to maintain, one could actual weld as per normal but with a very short arc; I had to hard drag the rod along the metal. The arc went out a few times and I had to scrape the rod on the floor to get ride of the sleeve. I reached the end of the weld and still had 1/3 of the rod left, so I cleaned the slag off the weld and laid in a light second pass using up the rod. Amperage readings were more stable than with no arc stabilizer, the average majority in the range of 105 to 120 amps. Pic9 - Coil in series~63 turns of 78 feet #4 cable, resistance of 0.020 ohms, inductance of 490 micro Henries.The arc was relatively easy to maintain easier that with #2 coil configuration. The system seemed to weld best by lightly dragging the rod along the metal - the arc was short. The arc went out a couple times and I did the scrape-the-floor thing till I discovered that I could just push the rod into metal and the arc would restart! I reached the end of the weld and still had 1/2 the rod left. I cleaned the slag off the weld and noted that the weld was rather thin and proud like I was a mite low on amperage. I decided to lay down a second pass over the first and I used up the rod. Amperage readings were more stable that previously, the average majority in the range of 95 to 115 amps. I apologize for the quality of the photos first time with digital camera (Ive got a lot to learn.)Rick V Attached Images
Reply:Pics are fine, Rick V. Remind us of what your plan is for rebuilding the stabilizer.I don't have any particular information or knowledge to pass on, but one thing you should be aware of is the possibility of saturation of the steel core due to high amp-turns product. In your case no damage will accrue because you are using the coil to stabilize the arc, rather than to control welding current. If you did go into saturation, you would just loose the benefit of the iron core in contributing to the inductance of the stabilizer and be back effectively to an air-core coil. Such things can be easily predicted with a little basic math if you know or can measure the magnetic properties of the steel, but I don't remember the calculations offhand. I think the experimental approach is more fun and probably just as educational at this stage. Doing the math first would allow you to minimize the error part of trial-and-error by predicting the inductance you will get and the current at which the core will saturate.Saturation will occur at lower amp-turns product for a smaller core cross-section and for a completely closed magnetic path around the core, although a closed core will give the highest inductance for a given core size and number of turns. That's why an air-gap is normally provided in higher current inductors - usually by putting insulating, non-magnetic shims between the E-stack and the I-stack of laminations (for a steel core). So part of the design effort is to come up with a core and winding configuration giving the desired inductance but which does not go into saturation over the range of currents you will be using. Notice that cores consisting of straight bundles of steel wire or straight bars of steel laminations have an enormous air gap from end to end and are thus saturation resistant (but provide low inductance for the amount of steel present).Such things as molded, powdered iron cores and ferrite cores have a certain amount of inherent protection against saturation because the magnetic particles are dispersed in a non-magnetic binder that provides the effect of a distributed air gap. These properties are very specific to the particular core material and dimensions, so you can't assume that any random ferrite or powdered iron core you find in your junk box will have the appropriate properties for your project without checking the manufacturer's specs on the core or experimenting. I suppose one good clue would be if the core came with windings on it of about the size you would be using.I don't know how you would determine that saturation has occurred other than performance of the welder or with incremental inductance measurements as DC current was increased. At some current you would see a fall-off of inductance, indicating some degree of saturation. Whether the saturation occurs suddenly or gradually is a function of core material and air gap. While many ferrite materials are specifically designed for abrupt saturation ("Square-Loop Cores"), you would want a softer, gradual saturation that is more common of ordinary line-powered transformer steel laminations. A junk yard might be a good source (although around here, the junque yards have gotten quite snotty about scrounging).That exhausts my rudimentary knowledge (?) of inductor design. Hope it is of interest. By the way, DVMs that hold max readings are fairly cheap these days and might be good marriage insurance.awrightLast edited by awright; 02-01-2007 at 02:32 PM.
Reply:Hey awright,RE: possibility of saturation of the steel core due to high amp-turns product. In my case, I don't think the combo of ~30 turns and 100 amps (3,000 amp-turns) is sufficient to cause magnetic saturation of the core - which is a pretty fair chunk of iron/steel.RE: a closed core will give the highest inductance for a given core size and number of turns. That's why an air-gap is normally provided in higher current inductors - In my H-coil design, I have a huge 3 inch air gap between the sides of the 'H' form - so I don't expect magnetic saturation.As to a closed form, the measured inductance of original coil, comprised of ~30.5 turns of 39 feet of #1 gauge wire, was L = 138 mirco Henries. I did try adding steel plates to close in the 'H' but I noted that the inductance only went up to 143 mirco Henries - hardly worth the effort.I think most power transformers/inductors use some form of closed core, not to increase inductance, but to confine the magnetic lines of force to the form - otherwise the magnetic field would affect things around it... disturbing electronics, attracting magnetic dust, etc. In my application, since I am using the arc stabilizer outside of a welder box, I don't think it matters.RE: So part of the design effort is to come up with a core and winding configuration giving the desired inductance but which does not go into saturationYes, I am going to try two approaches:1 - A simple steel core of 2.5 inch diameter x 6 inches long cold rolled steel.2 - A cardboard/plastic tube of ~2.5 inch diameter x 6 inches long filled with powdered iron oxide. (I have a source iron oxide powder/dust here used for magnetic-particle inspection.)I will wind a few turns of wire around each of these cores and use the inductance meter see which core has the higher inductance - then I'll use that core to make my Mark II arc stabilizer - same 30 turns of #1 wire wound over the center 3 inches of the core - leaving 1.5 inches of core sticking out either end.RE: By the way, DVMs that hold max readings are fairly cheap these days and might be good marriage insurance.Holding the max reading is not what I want here.e.g. Knowing I have say a max of 200 amps when I strike the arc is not as important as knowing that when actually welding the average current is 90-100 amps.I think I'll just place a small capacitor across the leads at the entry into the meter - that should slow up the change in voltage readings and let me (ok, let my wife) see the average readings. Anyway, the weekend awaits and I have all the new core material. Also, on the digital camera, I now know how to turn of the flash, use macro close up and manual focus! So pictures should be much better... it bothered me that the pictures of the coil welds looked worse that no coil weld - that was not the case. It's not so easy to capture the contour of a weld either.Rick V
Reply:A welding experimenter with an inductance meter is a pretty rare beast.I didn't want to add yet more ruminations to my too lengthy last post, but had thought about addressing just your comment on peak-hold meters. Some meters with reading hold function have an "Average" reading built in along with the min and max. Don't know about all meters, but my 15 year-old Fluke 87 would be easy to use for your measurement. With a long enough sample of welding current, the pauses at the ends of the welding period as you reach for the meter "HOLD" button and any high voltage starting transient could be very well cancelled out by a lengthy sample time.awright
Reply:Hey awright...RE: A welding experimenter with an inductance meter is a pretty rare beast.Yeah, I signed it out from work - a benefit of my work place (steel, alum, magnesium casting and rolling research, pipeline welding research and non-destructive testing practical examinaton centre).RE: Some meters with reading hold function have an "Average" reading built in along with the min and max.True... but I try to make do with what I have but I could sign another meter out from work - long as I don't burn holes in it!OK - LATEST RESULTS - NEW CORE MATERIALSSorry, I'm not having much luck with new cores
for the arc stabilizer coil.I had two new cores to try:a)2.5 inch diameter x 6 inch long round stock, cold-rolled steelb)2.5 inch diameter cardboard mailing tube filled with iron-oxide powderBefore going through the lengthy process of constructing a new stabilizer coil, I decided a quick comparison test would save time/effort. I took a spool of 25 feet of #16 gauge insulated wire and wrapped in around my prospective cores. I measured the inductance (L) on the cores and in air. To get a measure of the improvement, I calculated the Ratio of inductance = L(core) / L(air). Note: According to the books, maximum inductance occurs when the coil length = coil radius.Coil length usually means coil width, the width across the parallel windings.Coil radius means the distance from the centre of the coil to the middle of the windings.e.g. If the coil inner diameter = 2.5 inch (start of windings) and the outer diameter = 6 inch (last winding), then inner radius = 1.25, outer radius = 3, coil radius = 1.25 + (3 1.25)/2 = 2.125 inch.Anyway, all this really means is that maximum mutual inductance between wire occurs when the wire in the very centre of all the windings looks around and sees a maximum number of wires surrounding it in all directions! Core Results#1 Coil: 2.5 inch inner diameter, 31 turns (7 turns per layer x ~4.5 layers), ~1 inch longInductance (air) = 85 micro HenriesInductance (a - centre of 6 inch long steel round stock) = 134 micro Henries, Ratio = 1.57XInductance (b centre of 6 inch long iron-oxide tube) = 141 micro Henries, Ratio = 1.66XPretty Pathetic! My existing arc stabilizer coil already has a Ratio = ~2X.(I dont have an air measurement for my stabilizer coil, theory in air predicted 70 micro Henries and measured on steel was 138 micro Henries; hence a Ratio = ~2X.)I have to do way better than this!#2 Coil: I wound the 25 feet of #16 wire in a single layer (37 turns) around core a), the 2.5 inch diameter steel round stock. The measured inductance was L = 85 micro Henries. Crap! This the same as the 1st coil in air!I should have expected this
recall that max mutual inductance occurs when the central wire is equally surrounded by mutual friends (wires). Strung out in a single layer along the length of the core, each wire turn has only two neighbours one on either side and nobody living above or below
pretty lonely - limited mutual interaction.#3 Coil: 2 5/8 inch inner diameter, ~29 turns (7 turns per layer x ~4 layers), ~1 inch longInductance (air) = 79 micro HenriesInductance (a - centre of 6 inch long steel round stock) = 121 micro Henries, Ratio = 1.53XInductance (b centre of 6 inch long iron-oxide tube) = 128 micro Henries, Ratio = 1.62XThis is pretty much the same as #2 coil; the coil was only slightly larger to more easily slide over the two cores.Just for kicks, I placed two ferrite rods (old ARMACO FBC Ferrite Antenna Rod) in this coil. The two rods did not come even close to filling the core opening in the coil as each rod was only 3/8 inch diameter x 7 inch long. Anyway,Inductance = 166 micro Henries, Ratio = 2.1XFerrite wins!Coil #4:OK so I grabbed the two ferrite rods and tightly wrapped the 25 feet of wire around the rods - looked like a small OO-shaped form. Yeah, I kept the winding short and bunched up for higher mutual inductance.Inductance = 637 micro Henries!I carefully slipped the wire coil off the ferrite rods and measured the inductance in air.Inductance = 80 micro Henries.Hey, this gives a ratio of 637 / 80 = 8X !!Folks, I need FERRITE!Anything else for a core, like steel bars, steel round stock, laminated iron strips, iron-oxide powder, etc. is a waste of time give me that FERRITE !!!Now I can't wrap #1 gauge wire around those two small ferrite rods - not enough material to avoid saturation when flowing 100 amps about the rods. All I need is bundle of those ARMACO FBC Ferrite Antenna Rods - another 40 of those rods will do. Hmmm
Unfortunately, a quick search of the Internet indicates that ARMACO FBC Ferrite Antenna Rod distributed by R. MAC & CO. LTD. back in the mid 1970s is not to be found.So, any suggestions as to where I can find an inexpensive source of suitable ferrite rods or suitable core materials?Rick V
Reply:Ferrite?I tried inserting ferrite magnets into a coil of wire - and there was hardly any change in inductance.Important Lesson #1: magnets use 'hard' ferrite - useless for transformer/inductor applications.I need 'soft' magnetic material - able to be easily magnetized.I looked into Core Materials a little further.Here is a brief table of likely candidates.Material----------------Permeability------Saturation flux density B gaussCold rolled steel------ 180 - 2,000 ------- 21,000Iron------------------- 200 - 5,000 ------- 21,500Purified iron 99.95% ---5,000 - 180,000 --- 21,5004% Silicon-iron---------500 - 7,000 ------- 19,70045 Permalloy----------- 2,500 - 25,000 ---- 16,00045 Permalloy2---------- 4,000 - 50,000 ---- 16,00078 Permalloy----------- 8,000 -100,000 ---- 10,7004-79 Permalloy--------- 20,000-100,000 ---- 8,700Mu metal--------------- 20,000-100,000 ---- 6,500Supermalloy------------ 100,000-800,000 --- 8,000What we want firstly is high permeability and then high saturation flux.You can see that the permeability of cold rolled steel is not so great.Anyway, of all those materials I probably have quickest access to high purity iron from the casting department at work - likely in the form of shot (small spheres).Given the variety of materials I could try/use, I'm thinking to rebuild my coil using a hollow plastic tube as the center core around which I will wind the wire. That way, I can dump powder iron (I have than), high purity iron shot (likey have that soon), or whatever down the tube.Rick V
Reply:Recall that I was looking for a material with high magnetic permeability to use as the core of a new coil - hoping to increase the inductance of the coil/arc-stabilizer so that I would have a smoother and more stable arc when stick welding at ~100 amps using two 12-volt car batteries.I visited the casting department at work - armed with a test coil and the inductance meter. There were lots of bins full of materials such as electrolytic iron [Fe](99.99% pure), nickel [Ni], cobalt [Co], chromium [Cr], and mixes like FeCr69% and Fe-Bo.I didn't put these materials into the coil, I just lowered the coil on their surface. Air = 89 mirco HenriesElectrolytic iron = 115Fe-Bo = 107Nickel = 100, Nickel shot = 104FeCr=98Cobalt = 96I took enough electrolytic iron to fill a 2.5 inch diameter cardboard mailing tube to a depth of 7 inches. I had wound 25 feet of #16 insulated wire around a plastic form of 2 5/8th inch inner diameter. I put the coil 1/2 way down a 2.5 inch diameter 7 inch length of cold rolled steel then on the tube filled with thick (1/8 inch) flakes of electrolytic iron, then on those two 3/8 inch diameter, 7 inch long ferrite antenna rods.I measured the inductance in micro HenriesAir = 87Cold Rolled Steel = 128, Factor of 128/87 = 1.47X better than airElectrolytic Iron = 198, Factor of 198/87=2.28X better than airFerrite Antenna rods = 166, Factor of 166/87=1.91X better than airSo, the electrolytic iron is 198/128=1.3X or 30% better than cold-rolled steel.Yet this is a far cry from the book values of 5,000/200 = 25X!Look at those crazy Ferrite Antenna Rods, hardly filling up 4% of the coil but giving a significant increase in inductance. I sure do wish that 30 years back I had bought a big handful of those ferrite antenna rods!!! Anyway, with what I got (electrolytic iron), it looks that I can at least raise the inductance of my arc stabilizer by 30% - and that is progress. Rick V
Reply:Science marches on, Rick V. Thanks for the interesting narrative.Perhaps you are penalizing yourself too much with such a long effective air gap. After all, your typical power inductor with laminated silicon steel core has an air gap of perhaps .050" or .10 to reduce the susceptibility to saturation. I suspect that all your measurements are of the air core with a little bit of stray higher permeability material in the neighborhood, rather than an evaluation of the real potential of the magnetic material in a practical core configuration.Take a look at the typical inductor in a machine of similar current capability to yours. awrightLast edited by awright; 02-07-2007 at 02:30 AM.
Reply:Hey awright,RE: "I suspect that all your measurements are of the air core with a little bit of stray higher permeability material in the neighborhood, rather than an evaluation of the real potential of the magnetic material in a practical core configuration."Actually, at first I was laying the 7/8th inch thick test coil down flat on a cardboard box and then filling up the empty central core (to a depth of 7/8th inch) with metallic material from the storage bins - but that was quite messy. So, I tried just placing the coil flat on top of the metallic materials while still in the bins. Yes, the readings were somewhat less but the inductance ratios between three materials that I measured 'in-core' were the same when I measured 'flat on top'. Thus, 'flat on top' measurements were good enough to identify which of the materials in the 20 odd bins had the material with highest magnetic permeability. That material turned out to be the electrolytic iron. So I feel I did it right - certainly no super magnetic material slipped by me. Rick V
Reply:Rick V, I agree fully that your tests of relative permeability of the materials is quite valid as an abstract material evaluation. I guess I was reacting to the miniscule inductance values you were getting and suggesting that they could be greatly increased by incorporation into a closed core with a distributed air gap (for granular or powdered materials) or a discrete, but relatively small, air gap (for solid materials). But I see now that you were way ahead of me in your evaluation process.awright
Reply:Upgrading to Arc stabilizer Coil #2Recall Arc Stabilizer Coil #1?That was 39 feet of #1 gauge equivalent wire (26 parallel stands of #15 enamel wire) wound on a 2.625 inch hollow steel core with steel side frames. The measured inductance was 138 micro-Henries (uH) and the coil resistance was 0.005 ohms. I also had an option to use this coil as if it were 78 feet of #4 gauge equivalent wire (two parallel bundles, each of 13 strands of #15 gauge wire); in that case the inductance was 490 uH and coil resistance was 0.02 ohms.Arc Stabilizer Coil #2Ready to build a Bigger and Better Coil.I found some more of the #15 gauge enamel wire, enough to add 19.5 feet to the original coil length of 39 feet, making a total wire length of 58.5 feet. The core was a 2-inch internal diameter plastic vacuum pipe (1/16 inch thick wall) of 12-inch length. Side frames were made of two layers of 1/2 inch thick flake board and the length of the coil winding was the same as before = 3 inches.By the way, compared to quickly welding a rigid thin-steel frame for coil #1, construction of coil #2 required working with plastic, wood, screws and duct tape to achieve a semi-wobbly frame! See Pic#1 for a photo of the coil form.The predicted inductance of coil #2 in air was 160 uH.However, the actual as-wound coil inductance in air was 145 uH.Why lower?This was likely do to meandering gaps in my winding of the bundled wire. I experienced problems trying to smoothly wind the previously wound cable (now a tight coil of work-hardened copper) on the new form. It was a fight. See Pic#2 for a photo of the wire about to be wrapped around the new coil form. You can see the previously used wire in a tight coil and the new wire in a much larger loop of ~2.5 turns.Pic#3 is a photo of the coil winding in progress - after the first couple of layers of wire was wound on the new form.When I filled the plastic tube with electrolytic-iron flakes, the inductance rose from 145 to 297 uH, a factor of 297/145 = 2.05. I found that if I placed some 3/32 inch thick x 1.5 inch wide steel plates around where the plastic tube emerges from the side frames, I could raise the inductance by about 5%, from 297 to 311 uH.Final coil, inductance =311 uH, Resistance = 0.0075 ohms,Pic#4 is a photo photo of the completed arc stabilizer coil #2, the plastic core filled with iron flakes and capped with duct tape.OK, so what voltage drop will inserting coil #2 cause, and how much heat will it generate?At 100 amps, the coil resistance drop will be V=IR = 100 x 0.0075 = 0.75 volt and the power dissipated in the coil will be I^2R =100^2 x 0.0075 = 75 watts. At 200 amps, the coil resistance drop will be V=IR = 200 x 0.0075 = 1.5 volt and the power dissipated will be I^2R = 200^2 x 0.0075 = 300 watts. Given the large mass of copper involved (58.5 feet of #1 gauge copper = ~15 pounds of copper), I suspect a person could weld at ~200 amps for quite a while before the coil got hot; at ~100 amps, there should be no noticeable heating of the coil for a long time.Warning: I had some unexpected problems with copper wire!Left over from my amateur radio days, I found a 120 foot length of #14 gauge copper wire. Great I thought, I could use that wire to make my coil longer and get more inductance. A quick look at a wire table showed that ~20 parallel lengths of #14 would be equivalent to #1 gauge wire. So, I could lengthen my coil by 120/20 = 6.0 feet. I actually cut the wires and started to bundle them but decided that I really better confirm the resistance of this wire just as I had for the #15 gauge enamel wire that made up the bulk of my coil.Dad-Gum! This wire was crap! The wire measured out at 165 feet per ohm, a far cry from the 396 feet per ohm of proper #14 copper wire! My crap #14 gauge wire had the resistance of a #17.5 gauge good copper wire! What a waste of time that was but a darn good thing I checked else I would have added onto my 58.5 feet of #1 gauge equivalent wire a 6 foot length 20 parallel strands of that #14, thinking it was like #1 gauge, when it would only have been equivalent to #5 gauge. That single piece of 6-foot crap wire alone would have dropped 0.364 volts with a 200 amps current flowing and generated 73 watts 20% of the total coil heat! Using this crap wire, to make a wire bundle equivalent to #1 gauge wire, it would have taken not 20 parallel lengths of wire but 49!!! So, my 120 feet of crap wire would have yielded just 2.44 feet of #1 equivalent wire and been very BIG & FAT!!Totally unusable! Lesson Learned = Check Everything / Trust Nothing ! Next post... I will set up and report back on how the new arc stabilizer works for welding... Rick V Attached Images
Reply:That's real dedication to experimentation.What kind of insulation is that on your wire? It looks like plastic coating instead of enamel. Is it limiting the density of the winding?Look forward to your next report.awright
Reply:So, I owe you lunch.I bought a new Ranger 250 and it had this silly knob on the front that was "arc force" and "inductance". I read all I could, and called Lincoln twice. I found a graph in a book that showed that inductance slowed the incease in voltage to help the arc stabilize. You explained it great in your first posts. I tried the knob here and there. Min, Max and in the middle. Lotsa amps, little amps. I DID figure out what it does. Your explanation of your arc stabilizer helped me out a whole bunch, so THANK YOU!The rest of the stuff is way over my head.DavidReal world weldin. When I grow up I want to be a tig weldor.
Reply:To make a material that works like ferrite- Take VERY FINE powdered iron mix with epoxy, potting material, body filler or your choice of poisonplace in mold to the shape that you want the inductor inbefore the epoxy sets, strongly magnetize one end as N, the other as Sone set, you have a solid material where iron is lined up efficientlynext....instead of a fixed core inductor, make a "sliding core" unitwind your inductor on PVC pipe make a ferrite or iron core that slides in and out of the PVC pipe & coilyou can now adjust your inductancewe used to do this in the early 80s to stabilize the carbon arc in movie projectors at drive-in theaters where the power source was a motor-generator set.our coil was #2 solid. I don't remember the length and diameter*A toroid will also work with a sliding core*You can also make an inductor with a secondary winding and make adjustments by shunting the secondary winding with a carbon pile or a very large variable resistor.
Reply:HA!90% of this thread is done in chinese i see,wish i could grab ahold of all this valuable lingo!Thank goodness for pics!
Reply:Hi Folks,Thanks for all your comments and suggestions - and observations. OK when we left off coil #2 had an inductance in air of 145 micro Henries (uH) and when the 12 inch long, 2 inch diameter hollow core was filled with electrolytic-iron flakes, the inductance rose from 145 to 297 uH, a factor of 297/145 = 2.05X.Some improvements:1 - Added threaded rodsOnly duct tape was holding the frames in place so I installed a threaded rod and nuts at each corner to secure the assembly together more rigidly.2 - Added steel plates over the side framesRecall that when I temporarily placed steel plates on the side frames, I could raise the inductance by about 5%, from 297 to 311 uH. I figured I could get more inductance if I could get the steel plates closer to the coil. Recall the side frames were made of two layers of 1/2-inch thick flake board? I removed one layer of flake board from each side frame and covered the side frames with steel plates - using wood screws to secure the plates to the wood frames.See Pic#5 and Pic#6 for photos of the completed assembly. With the addition of the steel plates to the wood side frames, the inductance rose from 297 to 327 uH, better than I expected - a gain of about 10%. 2 - Those two 7 inch long ferrite rods (~3/8 inch diameter) that I had...hmmmm. I decided to add them into the core with the mix of iron flakes. I dumped out enough iron flakes to allow me to position the rods in the middle of the core length, then I taped the rods to opposite sides of the plastic core tube. I then added back in the iron flakes. The inductance rose from 327 to 433 uH! a gain of about 32%! (Wow... I sure could use lots of good 'soft' ferrite!)3 - That iron-oxide powder I had tried earlier....hmmmm... I wondered if that could be used to fill the voids between the iron flakes and maybe raise the inductance? I added iron-oxide powder to the core, shaking and tapping the fine powder down as I added scoop after scoop into one end of the plastic core tube. The inductance rose from 433 to ~481 uH a gain of another 10%. See Pic#7 for a photo of measuring the inductance of the completed assembly.Note: Actually, the inductance rose as high as 485 uH, but by the time I took the photo I guess the iron-oxide powder has sifted to different locations - that's why you see only 481 uH on the BK Precision Meter. Still, I likely get that back and more by adding more iron-oxide powder later.OK, so we have Arc-Stabilizer Coil #2 (58 feet of #1 cable) with an inductance of ~480 uH. This is almost the same value (490 uH) as Coil #1 when it was operating as two coils in series equivalent to 78 feet #4 cable. The difference is that Coil #2 has 2.65 times less resistance (0.0075 instead of 0.020 ohms).At 100 amps, that means the voltage drop in the coil will be 0.75 volts instead of 2 volts or welding with 23.25 volts instead of 22 volts - likely good for another 10-15 amps of much needed welding current.The Proof is in the Pudding, so let's get welding.... next post. Rick V Attached Images
Reply:OK, here are the welding results!BatteriesThe batteries are equivalent to two 12 volt 300 amp-hour batteries connected in series.Note: As an emergency power supply, I have a bank of six 12 volt 100 amp-hour deep cycle marine batteries connected in parallel feeding a 12 volt DC to 115 volt AC power inverter.See Pic#8 for a photo of the battery bank. For this welding test, I just connected in series, two parallel banks of three batteries each. This gives me 24 volts at a 300 amp-hour rating.WiringThe two parallel banks of three batteries each were connected in series with a 16 inch length of #4 gauge booster cable, each end fitted with conventional lead battery connectors as used on a car. The negative cable between the batteries and the work piece was 20 feet of #1 gauge booster cable, each end fitted with spring-loaded jaw clamps (booster cable clamps). The positive cable was also 20 feet of #1 gauge booster cable, each end fitted with spring-loaded jaw clamps (booster cable clamps). The positive cable ran from the batteries to either:a) The spring-loaded jaw clamp of a 5-foot length of #4 gauge booster cable that connected to a Bernard screw-type electrode holder.b) The copper connector on the arc stabilizer coil, then out of the stabilizer coil to the spring-loaded jaw clamp of the 5-foot length of #4 gauge booster cable that connected to a Bernard screw-type electrode holder.See Pic#9 for a photo of the booster-cable clamps connected to coil #2 and rod holder. Welding RodAll welding in this 2 battery test was done using Lincoln 7018AC, 3/32 inch welding rod.Digital Camera Trick - Amperage RecordYou may recall that last time while I was welding, I had my wife call out the amperages from the voltmeter display. This time, I set my digital camera in 'Movie Mode' and the camera recorded the image of the voltmeter display while I welded - it worked great! I don't need the wife - no more! Welding with No Arc Stabilizer CoilWeld #1Pic#10 shows the result (weld #1)- not very pretty!Where I started the arc (right side in photo), there is a large blob of weld metal and after that, as I moved to the left, there was not enough weld metal. Welding was the same as before - a major trial: blowing the end off the rod, scraping the rod tip on the concrete floor to expose rod metal, trying again, and again, and again. It was very difficult to maintain the arc - lot of repetitive tip blow off and scrape rod on floor... I think the rod tip spent equal time on the floor and on the weld!Welding amperages: 108, 187, 132, 106, 109, 107, 113, 125, 145, 104, 128, 103, 118, 121, 104.Average current - maybe about 120 amps, varying from 104 to 189 amps.Welding with Arc Stabilizer Coil #2Welding with the arc stabilizer was much better; the arc only went out a couple of times.Weld #2Pic#11 shows a much smoother weld (weld #2a), still a little lumpy but not too bad.Welding amperages: 100, 108, 115.Pic#12 shows a second pass (weld #2b) over weld #2a - good fusion but somewhat ugly looking - blame that me.Hey wait a minute - I share the blame with my welding helmet; yeah it was real cold (winter) in the garage and my auto-darkening helmet didn't lighten up past say #11. Welding amperages: 115, 120, 150... continued on next post (5 picture limit per post)Rick V Attached Images
Reply:Weld #3Pic#13 shows another weld (#3a) - good fusion but my technique was poor - rather embarrassing but it was a fair lay down of metal for 1 pass.Welding amperages: 93, 166, 105, 125, 148, 166, 109, 124, 121, 143, 102, 149, 110, 120, 130Average current - maybe about 128 amps, varying from 93 to 166 amps.Pic#14 shows the second pass (weld #3b) over weld #3a - Ok so I wandered a bit. Yeah, my helmet again!Welding amperages: 156, 130, 140, 152, 138, 149, 114, 132, 115, 110, 144, 99, 115Average current - maybe about 130 amps, varying from 99 to 156 amps.Folks, the arc is 'Real Short' and you must maintain light contact (by feel) between the rod and the base metal. Hey for me, this is good - I am forced to keep the arc short! I had a lot of past trouble with my 80 amp AC welder with an open circuit voltage of ~40-60 volts which allowed me to maintain far too long an arc - like maybe a 1/2 to 3/4 inch (Hey, I didn't know - I was a newbie) - and my welds did not fuse well at all - like over maybe 15% of their length! Yep, I'd knock the hot, just finished welded pieces off the welding table onto the floor and they would come apart!!! Folks, this just does not happen with this short-arc DC welding technique; if you don't maintain a real short arc, you don't weld!Pic#15 shows another weld (#4a) - again good fusion, somewhat smoother than my norm.Welding amperages: 129, 128,145, 136, 113, 140, 110, 130, 160, 136, 145Average current - maybe about 133 amps, varying from 113 to 160 amps.Note: The spring-loaded jaw clamps of the booster cable clamps were a source of trouble. They seem to work well at the batteries were the copper teeth dig into the soft lead battery terminals. However, the jaw-clamps did not have a lot of contact area connecting to the coil or to the cable leading to the electrode holder, so conduction was not always so good (See Pic#9). I cooked a rod - rod tip stuck on work and the whole rod turned red! This doesn't happen when the connections are good as the higher amperage available just melts the rod tip. In future, I will have to get rid of those spring clamps and use proper welding cable connectors. After the rod cook-off, I repositioned the jaw-clamps trying for better connection - you can see the positive result ~20 amp increase in the welding currents observed in weld #4b - below.Pic#16 shows the second pass (weld4b) over weld #4a - just after welding with the slag still in place.Pic#17 show the second pass (weld#4b) with the slag removed - again good fusion & poor technique.Welding amperages: 135, 198, 143, 156, 106, 127, 121, 113, 197, 190, 180, 198, 127, 173, 121, 113 [See amperages are higher]Average current - maybe about 150 amps, varying from 113 to 198 amps.... continued on next post [5 picture limit]Rick V Attached ImagesYep, this is the Successful Sequel to the original posting, Welding With Car Batteries - http://www.weldingweb.com/vbb/showthread...ight=batteries.This sequel deals with quality welding with just Two 12-volt lead acid batteries BUT using an ARC STABILIZER!There are two main problems when trying to weld with just two 12-volt batteries:1 - The arc is very difficult to start.Striking a new rod creates a brief short circuit, but with a battery power supply this flows a lot more current than a typical arc welder whose constant current design tends to limit the current to a set value. In comparison, batteries deliver high current that tends to blow the end off the rod! leaving the metal of the rod tip recessed inside a non-conducting sleeve (with 6013, 7014 and 7018 rods). The gap between the recessed rod tip and the work piece cannot be jumped with an open circuit voltage of only 24 volts from a battery power supply unlike a typical welder that may have an open circuit voltage from 40 to 80 volts and be able to jump a sizeable gap maintaining an arc. Once the arc is out, one has to break off the surrounding sleeve, usually by scrapping (dragging) the rod tip along a concrete floor.2 The arc is difficult to maintain. e.g. With 3/32 inch diameter Lincoln 7018AC, one has to weld by repeatedly moving the rod IN/OUT. IN - for an instant short circuit and then OUT to prevent sticking the rod to the work piece do this in/out (push/pull) as you drag the rod along the work piece. You kind of feel the rod tip mush into the work (arc is very short) and pull back slightly (arc is longer but in danger of going out) to prevent fusing the rod solid to the work piece then repeat. It is a dance between sticking the rod and the arc going out. Arc stoppages are frequent - in my experience, about 8 times on a 2.5-inch-long weld! Each stoppage requires breaking/scrapping the sleeve off the rod tip. It is a real pain in the butt trying to weld like this!The solution to both problems is an ARC STABILZER!An arc stabilizer is an inductor (a coil of wire) inserted in series with one of the welding cables.An inductor opposes any rapid change in current through itself.3 Starting the ArcWhen you strike an arc, there is a short circuit between the rod tip and the work piece. With a battery power supply, a large current rushes through the welding cables usually blowing the tip of the rod. However, with the proper size inductor, that large rush of current is prevented and goes into creating a large magnetic field within the inductor. The arc strikes more normally (with less current) and a reasonable arc forms. Note: As soon as the arc is first established, the stored magnetic energy in the coil from the arc strike is released as an increase in voltage and current through the rod thus maintaining the arc.4 Maintaining the ArcAs you weld, the rod burns down and the distance between the rod tip and the work piece increases and you have to move the rod in closer to the work. Since no welder is in perfect control, the length of the arc varies and this changes the electrical resistance of the arc a longer arc having greater resistance and requiring a larger voltage to maintain a constant welding current. Arc welders are designed for this, being constant current machines they adjust the voltage to maintain a constant current across the arc. Also, a typical arc welder has a 40 to 80 volt open circuit voltage that is able to jump a substantial arc gap.With a battery power supply, you have a constant voltage machine of only 24 volts, so welding current varies greatly with the length of the arc. However, with the proper size inductor, the large variations in current with variations in arc length are prevented; the inductor storing magnetic energy on short arcs delivering less voltage and less welding current; the inductor releasing stored magnetic energy on long arcs delivering more voltage and more welding current. In short, the inductor acts to stabilize the arc and acts to maintain the arc.BUILDING THE ARC STABILZERI figured that with only 24 volts, I did not want to drop any more than 0.5 volt (at 100 amps) across an inductor/coil. I had good experience with an arc stabilizer using 50 feet of 3/8 inch steel cable that worked well with three 12-volt batteries, so I figured that something like 40 feet of #1 gauge copper wire would do. I modelled various coils using a formula for multi-layer wound coils and it seemed that a 3 inch long form, 6 inches in diameter with a core of 2.6 inch diameter with about 5 layers of 6 turns each would do it produce a coil with an inductance of ~70 microHenries in air. Stuffing the core with steel, I expected this to increase by 400 times!I didnt want to buy 40 feet of #1 gauge wire. Rummaging in the basement, I found a couple of chunky coils of #15 gauge enamel coated wire that I wound 25 years ago! [Keep your Junk!]The total wire available was about 1,000 feet. So, I used that I paralleled 26 stands, 40 feet long, of #15 gauge wire. This gave me the equivalent resistance of 40 feet of #1 gauge wire R=0.005 ohms (Measured as 0.00492). With 100 amps of expected welding current through the coil, the power loss would only be I^2R = 10,000 x 0.005 = 50 watts and the voltage drop V = IR = 100 x 0.005 = 0.5 volt.However, when I built the coil using an H form of steel, I didnt get anywhere near the expected 400 times increase by stuffing the core with steel. Instead I got only a factor of 2!I made a bad design! My air core (theory) of 70 micro Henries measured as only 138 micro Henries.Note: Playing later with other small coils, I could obtain a factor of 5 or 6X using steel to fill the core and a factor over 12X using ferrite in the core far better than my measly 2X! Ok, since I used 26 parallel enamel coated #15 wires, I modified my wire groupings making two bundles, each of 13 wires. In parallel, the two bundles measured inductance was 138 micro Henries and resistance was 0.005 ohms. In series, the two bundles (twice as many turns) measured inductance was four times as high at 490 micro Henries - but the resistance was also four times as high at 0.02 ohms. At 100 amps welding current, the higher resistance would cause of power loss in the coil of I^2R = 10,000 x 0.02 = 200 watts and the voltage drop V = IR = 100 x 0.02 = 2.0 volt. That 2 volt drop looked significant meaning my welding voltage would drop from 24 volts to only 22
and that looked really marginal.DID THE ARC STABILZER WORK?***** You bet! *****Note: Batteries were series connected with a 5-foot length of #4 gauge wire. Current from the negative terminal of the batteries to the work piece was through a 20-foot length of #1 gauge wire. Current from the positive terminal of the batteries to arc stabilizer was through a 20-foot length of #1 gauge wire. Current from the arc stabilizer to the rod holder was through a 5-foot length of #4 gauge wire. In all three cases below, the weld work pieces were 2.5-inch-long T configuration (for filet weld) of 1/4 inch-thick mild-steel plates.A)Two 12-volt batteries - 24 voltsAs described, this was a real pain starting the arc and trying to maintain it was a frustrating experience. After several (~8) arc stoppages, I managed to complete a 2.5-inch-long filet weld. Weld quality was so-so.B)Two 12-volt batteries and a 138 micro Henry coil dropping ~0.4 volts at 80 amps.This was pretty good. Starting the arc was much easier. Maintaining the arc was not bad but I had to restart the arc twice during an identical weld. Weld quality was not bad at all.C)Two 12-volt batteries and a 490 micro Henry coil dropping ~1.6 volts at 80 amps.This was surprisingly good even with the ~2 volt drop in welding voltage. Starting the arc was easy. Maintaining the arc was simple; I did not have to restart the arc at all during an identical weld. Weld quality was very good. This test was surprising in that it showed that upside of having more inductance was of greater importance than the downside of having more resistance. Where to go from here?It is obvious that increasing inductance is the road forward.I need to re-design and remake my poor H-core design. The magnetic field is being lost in the sides of the H originally made of steel to hold the wire turns in place; this need to be a non-conductor and non-magnetic. If I can find and use ferrite instead of steel for the core, my existing coils should be capable of 5-fold increase in inductance. That would mean that my parallel wound inductor would go from 138 to 690 micro Henries and my series wound configuration should go from 490 to 2,450 micro Henries. Recall, welding was good with ~500 micro Henries.Pretty neat stuff! I had a lot of fun!Rick V
Reply:Very interesting, Rick V. Thanks for the detailed analysis and discussion of your experiments. A couple of questions come to mind.Where in the world did you come up with an "H" core in steel? Is this made up of steel transformer core laminations? (Just curious why you chose that design. I'm not tempted to try it as you already told us it was not an efficient configuration.)While we know that the conventional wisdom mandates use of insulated (varnished) transformer steel laminations for a low-frequency inductor core, for a quasi-DC application like this is there any reason one cannot use plain steel bars to form an inductor core? My understanding is that the primary reason for using laminated (silicon) steel as the core is to avoid eddy currents in a solid steel core induced by the fluctuating magnetic flux. If the only AC components in the magnetic flux are due to current transients during the initiation and termination of the arc, and the fluctuations incidental to operator technique, would eddy currents in an uninsulated core really be a significant energy loss, heat problem or "Q" killer? Or is part of the problem in using ordinary hot rolled steel as a core that mild steel just doesn't have the properties to do a good job as a magnetic core?awright
Reply:Hey awright - nice to hear from you...as always.Where in the world did you come up with an "H" core in steel?The 'H' core...I had to make a spool to hold the coil wire. I didn't have anything suitable so I went 'a-welding'. I welded four 1.5 x 3 x 3/32 pieces of steel together to form a square core/hub 1.5 x 1.5 inches square and 3 inches long. Then I welded up 4 pieces of the same material to form the square sides ~6 x 6 inches (left side of the letter H, and right side of the letter H) with a hole matching the core in the middle. I welded the sides to the core - forming a flat sided wheel - looks like an 'H' - a square wire spool viewed on the side view, wires are wrapped around the horizontal bar of the letter 'H'. Anyway, I did it to make a rigid spool. My wires didn't like bending 90 degrees around the square hub - so I built it up with layers of tape and strips of steel on the four sides to approximate a round core.Later, playing with small coils, I discovered that the magnetic field formed in the ends of the core flows into the H sides and this is not so good for getting high inductance values. e.g. I should have made the form out of plastic or wood - then loaded the hollow core with steel/iron. Mine - you could load the hollow core and it made zip difference.Is this made up of steel transformer core laminations?Nope - mild steel... as described above.is there any reason one cannot use plain steel bars to form an inductor core? I didn't have a suitable junk transformer around; I used mild steel bar stock.Eddy currents - I agree with you... since we are not running 60 cycles here, I don't expect much eddy current heating, as in "the only AC components in the magnetic flux are due to current transients during the initiation and termination of the arc".Yep, I think " mild steel just doesn't have the properties to do a good job as a magnetic core" for this application. Playing with several smaller air wound coils, stuffing the core with steel would get me 5 to 7 times the inductance with the core empty (air). A couple of 6 inch long 1/4 inch diameter ferrite antenna rods (Old am antenna rods radio's) would push the inductance up 10X - and the rods were filling only say 20- 30% of the core.Anyway, I am checking into core materials now - Internet is great for looking for tech info! This application is a mite different and I'm not certain what will work best in reality. The ferrite looks good on the BK inductance meter but what happens when one wacks 100 amps around it? If a ferrite core is going to saturate, then it kicks back into being like air - and one looses the high inductance seen at the very low test currents used by the inductance meter.I want to run this down; this welder runs real sweet and could be sweeter yet! Hey every time two fellows (or gals) get together, there are likely two car batteries just waiting to be used! I am quite amazed that the arc stabilizer has such a positive effect of the arc quality from the batteries - turning the battery welding from next to useless into quite a viable approach.This this thing has real potential! It just need some optimization.Awright, if you got some pointers or suggestions, I'm listening.Rick V
Reply:Maybe a suitable core material a ferrite toriod core ( donut shaped ). I have used them in high powered RF amplifier projects in my Ham radio days. see the link. http://hyperphysics.phy-astr.gsu.edu...ic/indtor.htmlNot a beginner, not a pro !
Reply:Well the mystery is solved.. Just dont do it near me.. Just for sh!+s and giggles..Who is gonna bust this myth... and show results..Not me.. I've done my duty Lets go...your turn.. ...zap!I am not completely insane..Some parts are missing Professional Driver on a closed course....Do not attempt.Just because I'm a dumbass don't mean that you can be too.So DON'T try any of this **** l do at home.
Reply:Very interesting read, thanks for sharing. Can we see some pics of the accomplished work?John - fabricator extraordinaire, car nut!- bleeding Miller blue! http://www.weldfabzone.com
Reply:Hey Folks,Ok before I destroy and rebuild the existing arc stabilizer, I thought I should take some photos. Pic1 shows the arc stabilizer in use, the coil-wire bundles configured in parallel to acts as ~31.5 turns of 39 feet #1 cable with a measured resistance of 0.005 ohms and an inductance of 138 micro Henries. Pic1Pic2 shows the arc stabilizer with the coil-wire bundles configured in series to acts as ~63 turns of 78 feet #4 cable with a measured resistance of 0.020 ohms and an inductance of 490 micro Henries. Below the arc stabilizer are the first three weld specimens: left = coil in series, middle = coil in parallel, right = no coil. So, heres some closer up pictures of the welds.Pic3 - Weld with coil-wire bundle in seriesPic4 - Weld with coil-wire bundle in parallel Pic5 - Weld with no coil ... continued on next post (photo limit 5 pics per post)Rick V Attached Images
Reply:Contunied from previous post...Before destroying/rebuilding the existing arc stabilizer, I decided to repeat the whole exercise because the first welds looked pretty sad in the photos. Also, I wanted to know what sort of welding current I had - without and with the arc stabilizer from two 12-volt batteries in series.Measuring the current:Im using #1 gauge booster cables. According to the charts, the resistance R of #1 gauge wire is 0.1264 ohms per 1000 feet. That means that 9.91 feet of wire has a resistance of 0.00100 ohms.Recall Voltage V = Current I x Resistance R.If we pass 1 amp through that resistance, we will create a voltage drop of 0.001 volt (1 millivolt).More importantly, a welding a current of 100 amps in the booster cable will create 100 millivolts on a voltmeter. So, we can read millivolts just as if it was amps of current. So, I connected the leads from a digital voltmeter into the #1 gauge ground cable at two points 9.91 feet (~9 ft 11 inches) apart just pushed the needle-like tips into the booster cable (see Pic6). Folks, I found that I could not weld and read the voltmeter at the same time
so my wife bundled in her winter coat, suffered the cold temperature in my frosty-breath garage, and called out the voltage readings while I welded! Thanks Honey!Ok, so welding:Pic7 - No CoilAs before, it was stab, mush the rod and pull back repeated ad infinitum with several breaks to scrape the rod tip on the floor to remove the non-conducting sleeve. Amperage (millivolt) readings were all over the place but the average majority were in the range of 90 to 135 amps. Pic8 - Coil in parallel ~31.5 turns of 39 feet #1 cable, resistance of 0.005 ohms, inductance of 138 micro Henries.The arc was much easier to maintain, one could actual weld as per normal but with a very short arc; I had to hard drag the rod along the metal. The arc went out a few times and I had to scrape the rod on the floor to get ride of the sleeve. I reached the end of the weld and still had 1/3 of the rod left, so I cleaned the slag off the weld and laid in a light second pass using up the rod. Amperage readings were more stable than with no arc stabilizer, the average majority in the range of 105 to 120 amps. Pic9 - Coil in series~63 turns of 78 feet #4 cable, resistance of 0.020 ohms, inductance of 490 micro Henries.The arc was relatively easy to maintain easier that with #2 coil configuration. The system seemed to weld best by lightly dragging the rod along the metal - the arc was short. The arc went out a couple times and I did the scrape-the-floor thing till I discovered that I could just push the rod into metal and the arc would restart! I reached the end of the weld and still had 1/2 the rod left. I cleaned the slag off the weld and noted that the weld was rather thin and proud like I was a mite low on amperage. I decided to lay down a second pass over the first and I used up the rod. Amperage readings were more stable that previously, the average majority in the range of 95 to 115 amps. I apologize for the quality of the photos first time with digital camera (Ive got a lot to learn.)Rick V Attached Images
Reply:Pics are fine, Rick V. Remind us of what your plan is for rebuilding the stabilizer.I don't have any particular information or knowledge to pass on, but one thing you should be aware of is the possibility of saturation of the steel core due to high amp-turns product. In your case no damage will accrue because you are using the coil to stabilize the arc, rather than to control welding current. If you did go into saturation, you would just loose the benefit of the iron core in contributing to the inductance of the stabilizer and be back effectively to an air-core coil. Such things can be easily predicted with a little basic math if you know or can measure the magnetic properties of the steel, but I don't remember the calculations offhand. I think the experimental approach is more fun and probably just as educational at this stage. Doing the math first would allow you to minimize the error part of trial-and-error by predicting the inductance you will get and the current at which the core will saturate.Saturation will occur at lower amp-turns product for a smaller core cross-section and for a completely closed magnetic path around the core, although a closed core will give the highest inductance for a given core size and number of turns. That's why an air-gap is normally provided in higher current inductors - usually by putting insulating, non-magnetic shims between the E-stack and the I-stack of laminations (for a steel core). So part of the design effort is to come up with a core and winding configuration giving the desired inductance but which does not go into saturation over the range of currents you will be using. Notice that cores consisting of straight bundles of steel wire or straight bars of steel laminations have an enormous air gap from end to end and are thus saturation resistant (but provide low inductance for the amount of steel present).Such things as molded, powdered iron cores and ferrite cores have a certain amount of inherent protection against saturation because the magnetic particles are dispersed in a non-magnetic binder that provides the effect of a distributed air gap. These properties are very specific to the particular core material and dimensions, so you can't assume that any random ferrite or powdered iron core you find in your junk box will have the appropriate properties for your project without checking the manufacturer's specs on the core or experimenting. I suppose one good clue would be if the core came with windings on it of about the size you would be using.I don't know how you would determine that saturation has occurred other than performance of the welder or with incremental inductance measurements as DC current was increased. At some current you would see a fall-off of inductance, indicating some degree of saturation. Whether the saturation occurs suddenly or gradually is a function of core material and air gap. While many ferrite materials are specifically designed for abrupt saturation ("Square-Loop Cores"), you would want a softer, gradual saturation that is more common of ordinary line-powered transformer steel laminations. A junk yard might be a good source (although around here, the junque yards have gotten quite snotty about scrounging).That exhausts my rudimentary knowledge (?) of inductor design. Hope it is of interest. By the way, DVMs that hold max readings are fairly cheap these days and might be good marriage insurance.awrightLast edited by awright; 02-01-2007 at 02:32 PM.
Reply:Hey awright,RE: possibility of saturation of the steel core due to high amp-turns product. In my case, I don't think the combo of ~30 turns and 100 amps (3,000 amp-turns) is sufficient to cause magnetic saturation of the core - which is a pretty fair chunk of iron/steel.RE: a closed core will give the highest inductance for a given core size and number of turns. That's why an air-gap is normally provided in higher current inductors - In my H-coil design, I have a huge 3 inch air gap between the sides of the 'H' form - so I don't expect magnetic saturation.As to a closed form, the measured inductance of original coil, comprised of ~30.5 turns of 39 feet of #1 gauge wire, was L = 138 mirco Henries. I did try adding steel plates to close in the 'H' but I noted that the inductance only went up to 143 mirco Henries - hardly worth the effort.I think most power transformers/inductors use some form of closed core, not to increase inductance, but to confine the magnetic lines of force to the form - otherwise the magnetic field would affect things around it... disturbing electronics, attracting magnetic dust, etc. In my application, since I am using the arc stabilizer outside of a welder box, I don't think it matters.RE: So part of the design effort is to come up with a core and winding configuration giving the desired inductance but which does not go into saturationYes, I am going to try two approaches:1 - A simple steel core of 2.5 inch diameter x 6 inches long cold rolled steel.2 - A cardboard/plastic tube of ~2.5 inch diameter x 6 inches long filled with powdered iron oxide. (I have a source iron oxide powder/dust here used for magnetic-particle inspection.)I will wind a few turns of wire around each of these cores and use the inductance meter see which core has the higher inductance - then I'll use that core to make my Mark II arc stabilizer - same 30 turns of #1 wire wound over the center 3 inches of the core - leaving 1.5 inches of core sticking out either end.RE: By the way, DVMs that hold max readings are fairly cheap these days and might be good marriage insurance.Holding the max reading is not what I want here.e.g. Knowing I have say a max of 200 amps when I strike the arc is not as important as knowing that when actually welding the average current is 90-100 amps.I think I'll just place a small capacitor across the leads at the entry into the meter - that should slow up the change in voltage readings and let me (ok, let my wife) see the average readings. Anyway, the weekend awaits and I have all the new core material. Also, on the digital camera, I now know how to turn of the flash, use macro close up and manual focus! So pictures should be much better... it bothered me that the pictures of the coil welds looked worse that no coil weld - that was not the case. It's not so easy to capture the contour of a weld either.Rick V
Reply:A welding experimenter with an inductance meter is a pretty rare beast.I didn't want to add yet more ruminations to my too lengthy last post, but had thought about addressing just your comment on peak-hold meters. Some meters with reading hold function have an "Average" reading built in along with the min and max. Don't know about all meters, but my 15 year-old Fluke 87 would be easy to use for your measurement. With a long enough sample of welding current, the pauses at the ends of the welding period as you reach for the meter "HOLD" button and any high voltage starting transient could be very well cancelled out by a lengthy sample time.awright
Reply:Hey awright...RE: A welding experimenter with an inductance meter is a pretty rare beast.Yeah, I signed it out from work - a benefit of my work place (steel, alum, magnesium casting and rolling research, pipeline welding research and non-destructive testing practical examinaton centre).RE: Some meters with reading hold function have an "Average" reading built in along with the min and max.True... but I try to make do with what I have but I could sign another meter out from work - long as I don't burn holes in it!OK - LATEST RESULTS - NEW CORE MATERIALSSorry, I'm not having much luck with new cores
for the arc stabilizer coil.I had two new cores to try:a)2.5 inch diameter x 6 inch long round stock, cold-rolled steelb)2.5 inch diameter cardboard mailing tube filled with iron-oxide powderBefore going through the lengthy process of constructing a new stabilizer coil, I decided a quick comparison test would save time/effort. I took a spool of 25 feet of #16 gauge insulated wire and wrapped in around my prospective cores. I measured the inductance (L) on the cores and in air. To get a measure of the improvement, I calculated the Ratio of inductance = L(core) / L(air). Note: According to the books, maximum inductance occurs when the coil length = coil radius.Coil length usually means coil width, the width across the parallel windings.Coil radius means the distance from the centre of the coil to the middle of the windings.e.g. If the coil inner diameter = 2.5 inch (start of windings) and the outer diameter = 6 inch (last winding), then inner radius = 1.25, outer radius = 3, coil radius = 1.25 + (3 1.25)/2 = 2.125 inch.Anyway, all this really means is that maximum mutual inductance between wire occurs when the wire in the very centre of all the windings looks around and sees a maximum number of wires surrounding it in all directions! Core Results#1 Coil: 2.5 inch inner diameter, 31 turns (7 turns per layer x ~4.5 layers), ~1 inch longInductance (air) = 85 micro HenriesInductance (a - centre of 6 inch long steel round stock) = 134 micro Henries, Ratio = 1.57XInductance (b centre of 6 inch long iron-oxide tube) = 141 micro Henries, Ratio = 1.66XPretty Pathetic! My existing arc stabilizer coil already has a Ratio = ~2X.(I dont have an air measurement for my stabilizer coil, theory in air predicted 70 micro Henries and measured on steel was 138 micro Henries; hence a Ratio = ~2X.)I have to do way better than this!#2 Coil: I wound the 25 feet of #16 wire in a single layer (37 turns) around core a), the 2.5 inch diameter steel round stock. The measured inductance was L = 85 micro Henries. Crap! This the same as the 1st coil in air!I should have expected this
recall that max mutual inductance occurs when the central wire is equally surrounded by mutual friends (wires). Strung out in a single layer along the length of the core, each wire turn has only two neighbours one on either side and nobody living above or below
pretty lonely - limited mutual interaction.#3 Coil: 2 5/8 inch inner diameter, ~29 turns (7 turns per layer x ~4 layers), ~1 inch longInductance (air) = 79 micro HenriesInductance (a - centre of 6 inch long steel round stock) = 121 micro Henries, Ratio = 1.53XInductance (b centre of 6 inch long iron-oxide tube) = 128 micro Henries, Ratio = 1.62XThis is pretty much the same as #2 coil; the coil was only slightly larger to more easily slide over the two cores.Just for kicks, I placed two ferrite rods (old ARMACO FBC Ferrite Antenna Rod) in this coil. The two rods did not come even close to filling the core opening in the coil as each rod was only 3/8 inch diameter x 7 inch long. Anyway,Inductance = 166 micro Henries, Ratio = 2.1XFerrite wins!Coil #4:OK so I grabbed the two ferrite rods and tightly wrapped the 25 feet of wire around the rods - looked like a small OO-shaped form. Yeah, I kept the winding short and bunched up for higher mutual inductance.Inductance = 637 micro Henries!I carefully slipped the wire coil off the ferrite rods and measured the inductance in air.Inductance = 80 micro Henries.Hey, this gives a ratio of 637 / 80 = 8X !!Folks, I need FERRITE!Anything else for a core, like steel bars, steel round stock, laminated iron strips, iron-oxide powder, etc. is a waste of time give me that FERRITE !!!Now I can't wrap #1 gauge wire around those two small ferrite rods - not enough material to avoid saturation when flowing 100 amps about the rods. All I need is bundle of those ARMACO FBC Ferrite Antenna Rods - another 40 of those rods will do. Hmmm
Unfortunately, a quick search of the Internet indicates that ARMACO FBC Ferrite Antenna Rod distributed by R. MAC & CO. LTD. back in the mid 1970s is not to be found.So, any suggestions as to where I can find an inexpensive source of suitable ferrite rods or suitable core materials?Rick V
Reply:Ferrite?I tried inserting ferrite magnets into a coil of wire - and there was hardly any change in inductance.Important Lesson #1: magnets use 'hard' ferrite - useless for transformer/inductor applications.I need 'soft' magnetic material - able to be easily magnetized.I looked into Core Materials a little further.Here is a brief table of likely candidates.Material----------------Permeability------Saturation flux density B gaussCold rolled steel------ 180 - 2,000 ------- 21,000Iron------------------- 200 - 5,000 ------- 21,500Purified iron 99.95% ---5,000 - 180,000 --- 21,5004% Silicon-iron---------500 - 7,000 ------- 19,70045 Permalloy----------- 2,500 - 25,000 ---- 16,00045 Permalloy2---------- 4,000 - 50,000 ---- 16,00078 Permalloy----------- 8,000 -100,000 ---- 10,7004-79 Permalloy--------- 20,000-100,000 ---- 8,700Mu metal--------------- 20,000-100,000 ---- 6,500Supermalloy------------ 100,000-800,000 --- 8,000What we want firstly is high permeability and then high saturation flux.You can see that the permeability of cold rolled steel is not so great.Anyway, of all those materials I probably have quickest access to high purity iron from the casting department at work - likely in the form of shot (small spheres).Given the variety of materials I could try/use, I'm thinking to rebuild my coil using a hollow plastic tube as the center core around which I will wind the wire. That way, I can dump powder iron (I have than), high purity iron shot (likey have that soon), or whatever down the tube.Rick V
Reply:Recall that I was looking for a material with high magnetic permeability to use as the core of a new coil - hoping to increase the inductance of the coil/arc-stabilizer so that I would have a smoother and more stable arc when stick welding at ~100 amps using two 12-volt car batteries.I visited the casting department at work - armed with a test coil and the inductance meter. There were lots of bins full of materials such as electrolytic iron [Fe](99.99% pure), nickel [Ni], cobalt [Co], chromium [Cr], and mixes like FeCr69% and Fe-Bo.I didn't put these materials into the coil, I just lowered the coil on their surface. Air = 89 mirco HenriesElectrolytic iron = 115Fe-Bo = 107Nickel = 100, Nickel shot = 104FeCr=98Cobalt = 96I took enough electrolytic iron to fill a 2.5 inch diameter cardboard mailing tube to a depth of 7 inches. I had wound 25 feet of #16 insulated wire around a plastic form of 2 5/8th inch inner diameter. I put the coil 1/2 way down a 2.5 inch diameter 7 inch length of cold rolled steel then on the tube filled with thick (1/8 inch) flakes of electrolytic iron, then on those two 3/8 inch diameter, 7 inch long ferrite antenna rods.I measured the inductance in micro HenriesAir = 87Cold Rolled Steel = 128, Factor of 128/87 = 1.47X better than airElectrolytic Iron = 198, Factor of 198/87=2.28X better than airFerrite Antenna rods = 166, Factor of 166/87=1.91X better than airSo, the electrolytic iron is 198/128=1.3X or 30% better than cold-rolled steel.Yet this is a far cry from the book values of 5,000/200 = 25X!Look at those crazy Ferrite Antenna Rods, hardly filling up 4% of the coil but giving a significant increase in inductance. I sure do wish that 30 years back I had bought a big handful of those ferrite antenna rods!!! Anyway, with what I got (electrolytic iron), it looks that I can at least raise the inductance of my arc stabilizer by 30% - and that is progress. Rick V
Reply:Science marches on, Rick V. Thanks for the interesting narrative.Perhaps you are penalizing yourself too much with such a long effective air gap. After all, your typical power inductor with laminated silicon steel core has an air gap of perhaps .050" or .10 to reduce the susceptibility to saturation. I suspect that all your measurements are of the air core with a little bit of stray higher permeability material in the neighborhood, rather than an evaluation of the real potential of the magnetic material in a practical core configuration.Take a look at the typical inductor in a machine of similar current capability to yours. awrightLast edited by awright; 02-07-2007 at 02:30 AM.
Reply:Hey awright,RE: "I suspect that all your measurements are of the air core with a little bit of stray higher permeability material in the neighborhood, rather than an evaluation of the real potential of the magnetic material in a practical core configuration."Actually, at first I was laying the 7/8th inch thick test coil down flat on a cardboard box and then filling up the empty central core (to a depth of 7/8th inch) with metallic material from the storage bins - but that was quite messy. So, I tried just placing the coil flat on top of the metallic materials while still in the bins. Yes, the readings were somewhat less but the inductance ratios between three materials that I measured 'in-core' were the same when I measured 'flat on top'. Thus, 'flat on top' measurements were good enough to identify which of the materials in the 20 odd bins had the material with highest magnetic permeability. That material turned out to be the electrolytic iron. So I feel I did it right - certainly no super magnetic material slipped by me. Rick V
Reply:Rick V, I agree fully that your tests of relative permeability of the materials is quite valid as an abstract material evaluation. I guess I was reacting to the miniscule inductance values you were getting and suggesting that they could be greatly increased by incorporation into a closed core with a distributed air gap (for granular or powdered materials) or a discrete, but relatively small, air gap (for solid materials). But I see now that you were way ahead of me in your evaluation process.awright
Reply:Upgrading to Arc stabilizer Coil #2Recall Arc Stabilizer Coil #1?That was 39 feet of #1 gauge equivalent wire (26 parallel stands of #15 enamel wire) wound on a 2.625 inch hollow steel core with steel side frames. The measured inductance was 138 micro-Henries (uH) and the coil resistance was 0.005 ohms. I also had an option to use this coil as if it were 78 feet of #4 gauge equivalent wire (two parallel bundles, each of 13 strands of #15 gauge wire); in that case the inductance was 490 uH and coil resistance was 0.02 ohms.Arc Stabilizer Coil #2Ready to build a Bigger and Better Coil.I found some more of the #15 gauge enamel wire, enough to add 19.5 feet to the original coil length of 39 feet, making a total wire length of 58.5 feet. The core was a 2-inch internal diameter plastic vacuum pipe (1/16 inch thick wall) of 12-inch length. Side frames were made of two layers of 1/2 inch thick flake board and the length of the coil winding was the same as before = 3 inches.By the way, compared to quickly welding a rigid thin-steel frame for coil #1, construction of coil #2 required working with plastic, wood, screws and duct tape to achieve a semi-wobbly frame! See Pic#1 for a photo of the coil form.The predicted inductance of coil #2 in air was 160 uH.However, the actual as-wound coil inductance in air was 145 uH.Why lower?This was likely do to meandering gaps in my winding of the bundled wire. I experienced problems trying to smoothly wind the previously wound cable (now a tight coil of work-hardened copper) on the new form. It was a fight. See Pic#2 for a photo of the wire about to be wrapped around the new coil form. You can see the previously used wire in a tight coil and the new wire in a much larger loop of ~2.5 turns.Pic#3 is a photo of the coil winding in progress - after the first couple of layers of wire was wound on the new form.When I filled the plastic tube with electrolytic-iron flakes, the inductance rose from 145 to 297 uH, a factor of 297/145 = 2.05. I found that if I placed some 3/32 inch thick x 1.5 inch wide steel plates around where the plastic tube emerges from the side frames, I could raise the inductance by about 5%, from 297 to 311 uH.Final coil, inductance =311 uH, Resistance = 0.0075 ohms,Pic#4 is a photo photo of the completed arc stabilizer coil #2, the plastic core filled with iron flakes and capped with duct tape.OK, so what voltage drop will inserting coil #2 cause, and how much heat will it generate?At 100 amps, the coil resistance drop will be V=IR = 100 x 0.0075 = 0.75 volt and the power dissipated in the coil will be I^2R =100^2 x 0.0075 = 75 watts. At 200 amps, the coil resistance drop will be V=IR = 200 x 0.0075 = 1.5 volt and the power dissipated will be I^2R = 200^2 x 0.0075 = 300 watts. Given the large mass of copper involved (58.5 feet of #1 gauge copper = ~15 pounds of copper), I suspect a person could weld at ~200 amps for quite a while before the coil got hot; at ~100 amps, there should be no noticeable heating of the coil for a long time.Warning: I had some unexpected problems with copper wire!Left over from my amateur radio days, I found a 120 foot length of #14 gauge copper wire. Great I thought, I could use that wire to make my coil longer and get more inductance. A quick look at a wire table showed that ~20 parallel lengths of #14 would be equivalent to #1 gauge wire. So, I could lengthen my coil by 120/20 = 6.0 feet. I actually cut the wires and started to bundle them but decided that I really better confirm the resistance of this wire just as I had for the #15 gauge enamel wire that made up the bulk of my coil.Dad-Gum! This wire was crap! The wire measured out at 165 feet per ohm, a far cry from the 396 feet per ohm of proper #14 copper wire! My crap #14 gauge wire had the resistance of a #17.5 gauge good copper wire! What a waste of time that was but a darn good thing I checked else I would have added onto my 58.5 feet of #1 gauge equivalent wire a 6 foot length 20 parallel strands of that #14, thinking it was like #1 gauge, when it would only have been equivalent to #5 gauge. That single piece of 6-foot crap wire alone would have dropped 0.364 volts with a 200 amps current flowing and generated 73 watts 20% of the total coil heat! Using this crap wire, to make a wire bundle equivalent to #1 gauge wire, it would have taken not 20 parallel lengths of wire but 49!!! So, my 120 feet of crap wire would have yielded just 2.44 feet of #1 equivalent wire and been very BIG & FAT!!Totally unusable! Lesson Learned = Check Everything / Trust Nothing ! Next post... I will set up and report back on how the new arc stabilizer works for welding... Rick V Attached Images
Reply:That's real dedication to experimentation.What kind of insulation is that on your wire? It looks like plastic coating instead of enamel. Is it limiting the density of the winding?Look forward to your next report.awright
Reply:So, I owe you lunch.I bought a new Ranger 250 and it had this silly knob on the front that was "arc force" and "inductance". I read all I could, and called Lincoln twice. I found a graph in a book that showed that inductance slowed the incease in voltage to help the arc stabilize. You explained it great in your first posts. I tried the knob here and there. Min, Max and in the middle. Lotsa amps, little amps. I DID figure out what it does. Your explanation of your arc stabilizer helped me out a whole bunch, so THANK YOU!The rest of the stuff is way over my head.DavidReal world weldin. When I grow up I want to be a tig weldor.
Reply:To make a material that works like ferrite- Take VERY FINE powdered iron mix with epoxy, potting material, body filler or your choice of poisonplace in mold to the shape that you want the inductor inbefore the epoxy sets, strongly magnetize one end as N, the other as Sone set, you have a solid material where iron is lined up efficientlynext....instead of a fixed core inductor, make a "sliding core" unitwind your inductor on PVC pipe make a ferrite or iron core that slides in and out of the PVC pipe & coilyou can now adjust your inductancewe used to do this in the early 80s to stabilize the carbon arc in movie projectors at drive-in theaters where the power source was a motor-generator set.our coil was #2 solid. I don't remember the length and diameter*A toroid will also work with a sliding core*You can also make an inductor with a secondary winding and make adjustments by shunting the secondary winding with a carbon pile or a very large variable resistor.
Reply:HA!90% of this thread is done in chinese i see,wish i could grab ahold of all this valuable lingo!Thank goodness for pics!
Reply:Hi Folks,Thanks for all your comments and suggestions - and observations. OK when we left off coil #2 had an inductance in air of 145 micro Henries (uH) and when the 12 inch long, 2 inch diameter hollow core was filled with electrolytic-iron flakes, the inductance rose from 145 to 297 uH, a factor of 297/145 = 2.05X.Some improvements:1 - Added threaded rodsOnly duct tape was holding the frames in place so I installed a threaded rod and nuts at each corner to secure the assembly together more rigidly.2 - Added steel plates over the side framesRecall that when I temporarily placed steel plates on the side frames, I could raise the inductance by about 5%, from 297 to 311 uH. I figured I could get more inductance if I could get the steel plates closer to the coil. Recall the side frames were made of two layers of 1/2-inch thick flake board? I removed one layer of flake board from each side frame and covered the side frames with steel plates - using wood screws to secure the plates to the wood frames.See Pic#5 and Pic#6 for photos of the completed assembly. With the addition of the steel plates to the wood side frames, the inductance rose from 297 to 327 uH, better than I expected - a gain of about 10%. 2 - Those two 7 inch long ferrite rods (~3/8 inch diameter) that I had...hmmmm. I decided to add them into the core with the mix of iron flakes. I dumped out enough iron flakes to allow me to position the rods in the middle of the core length, then I taped the rods to opposite sides of the plastic core tube. I then added back in the iron flakes. The inductance rose from 327 to 433 uH! a gain of about 32%! (Wow... I sure could use lots of good 'soft' ferrite!)3 - That iron-oxide powder I had tried earlier....hmmmm... I wondered if that could be used to fill the voids between the iron flakes and maybe raise the inductance? I added iron-oxide powder to the core, shaking and tapping the fine powder down as I added scoop after scoop into one end of the plastic core tube. The inductance rose from 433 to ~481 uH a gain of another 10%. See Pic#7 for a photo of measuring the inductance of the completed assembly.Note: Actually, the inductance rose as high as 485 uH, but by the time I took the photo I guess the iron-oxide powder has sifted to different locations - that's why you see only 481 uH on the BK Precision Meter. Still, I likely get that back and more by adding more iron-oxide powder later.OK, so we have Arc-Stabilizer Coil #2 (58 feet of #1 cable) with an inductance of ~480 uH. This is almost the same value (490 uH) as Coil #1 when it was operating as two coils in series equivalent to 78 feet #4 cable. The difference is that Coil #2 has 2.65 times less resistance (0.0075 instead of 0.020 ohms).At 100 amps, that means the voltage drop in the coil will be 0.75 volts instead of 2 volts or welding with 23.25 volts instead of 22 volts - likely good for another 10-15 amps of much needed welding current.The Proof is in the Pudding, so let's get welding.... next post. Rick V Attached Images
Reply:OK, here are the welding results!BatteriesThe batteries are equivalent to two 12 volt 300 amp-hour batteries connected in series.Note: As an emergency power supply, I have a bank of six 12 volt 100 amp-hour deep cycle marine batteries connected in parallel feeding a 12 volt DC to 115 volt AC power inverter.See Pic#8 for a photo of the battery bank. For this welding test, I just connected in series, two parallel banks of three batteries each. This gives me 24 volts at a 300 amp-hour rating.WiringThe two parallel banks of three batteries each were connected in series with a 16 inch length of #4 gauge booster cable, each end fitted with conventional lead battery connectors as used on a car. The negative cable between the batteries and the work piece was 20 feet of #1 gauge booster cable, each end fitted with spring-loaded jaw clamps (booster cable clamps). The positive cable was also 20 feet of #1 gauge booster cable, each end fitted with spring-loaded jaw clamps (booster cable clamps). The positive cable ran from the batteries to either:a) The spring-loaded jaw clamp of a 5-foot length of #4 gauge booster cable that connected to a Bernard screw-type electrode holder.b) The copper connector on the arc stabilizer coil, then out of the stabilizer coil to the spring-loaded jaw clamp of the 5-foot length of #4 gauge booster cable that connected to a Bernard screw-type electrode holder.See Pic#9 for a photo of the booster-cable clamps connected to coil #2 and rod holder. Welding RodAll welding in this 2 battery test was done using Lincoln 7018AC, 3/32 inch welding rod.Digital Camera Trick - Amperage RecordYou may recall that last time while I was welding, I had my wife call out the amperages from the voltmeter display. This time, I set my digital camera in 'Movie Mode' and the camera recorded the image of the voltmeter display while I welded - it worked great! I don't need the wife - no more! Welding with No Arc Stabilizer CoilWeld #1Pic#10 shows the result (weld #1)- not very pretty!Where I started the arc (right side in photo), there is a large blob of weld metal and after that, as I moved to the left, there was not enough weld metal. Welding was the same as before - a major trial: blowing the end off the rod, scraping the rod tip on the concrete floor to expose rod metal, trying again, and again, and again. It was very difficult to maintain the arc - lot of repetitive tip blow off and scrape rod on floor... I think the rod tip spent equal time on the floor and on the weld!Welding amperages: 108, 187, 132, 106, 109, 107, 113, 125, 145, 104, 128, 103, 118, 121, 104.Average current - maybe about 120 amps, varying from 104 to 189 amps.Welding with Arc Stabilizer Coil #2Welding with the arc stabilizer was much better; the arc only went out a couple of times.Weld #2Pic#11 shows a much smoother weld (weld #2a), still a little lumpy but not too bad.Welding amperages: 100, 108, 115.Pic#12 shows a second pass (weld #2b) over weld #2a - good fusion but somewhat ugly looking - blame that me.Hey wait a minute - I share the blame with my welding helmet; yeah it was real cold (winter) in the garage and my auto-darkening helmet didn't lighten up past say #11. Welding amperages: 115, 120, 150... continued on next post (5 picture limit per post)Rick V Attached Images
Reply:Weld #3Pic#13 shows another weld (#3a) - good fusion but my technique was poor - rather embarrassing but it was a fair lay down of metal for 1 pass.Welding amperages: 93, 166, 105, 125, 148, 166, 109, 124, 121, 143, 102, 149, 110, 120, 130Average current - maybe about 128 amps, varying from 93 to 166 amps.Pic#14 shows the second pass (weld #3b) over weld #3a - Ok so I wandered a bit. Yeah, my helmet again!Welding amperages: 156, 130, 140, 152, 138, 149, 114, 132, 115, 110, 144, 99, 115Average current - maybe about 130 amps, varying from 99 to 156 amps.Folks, the arc is 'Real Short' and you must maintain light contact (by feel) between the rod and the base metal. Hey for me, this is good - I am forced to keep the arc short! I had a lot of past trouble with my 80 amp AC welder with an open circuit voltage of ~40-60 volts which allowed me to maintain far too long an arc - like maybe a 1/2 to 3/4 inch (Hey, I didn't know - I was a newbie) - and my welds did not fuse well at all - like over maybe 15% of their length! Yep, I'd knock the hot, just finished welded pieces off the welding table onto the floor and they would come apart!!! Folks, this just does not happen with this short-arc DC welding technique; if you don't maintain a real short arc, you don't weld!Pic#15 shows another weld (#4a) - again good fusion, somewhat smoother than my norm.Welding amperages: 129, 128,145, 136, 113, 140, 110, 130, 160, 136, 145Average current - maybe about 133 amps, varying from 113 to 160 amps.Note: The spring-loaded jaw clamps of the booster cable clamps were a source of trouble. They seem to work well at the batteries were the copper teeth dig into the soft lead battery terminals. However, the jaw-clamps did not have a lot of contact area connecting to the coil or to the cable leading to the electrode holder, so conduction was not always so good (See Pic#9). I cooked a rod - rod tip stuck on work and the whole rod turned red! This doesn't happen when the connections are good as the higher amperage available just melts the rod tip. In future, I will have to get rid of those spring clamps and use proper welding cable connectors. After the rod cook-off, I repositioned the jaw-clamps trying for better connection - you can see the positive result ~20 amp increase in the welding currents observed in weld #4b - below.Pic#16 shows the second pass (weld4b) over weld #4a - just after welding with the slag still in place.Pic#17 show the second pass (weld#4b) with the slag removed - again good fusion & poor technique.Welding amperages: 135, 198, 143, 156, 106, 127, 121, 113, 197, 190, 180, 198, 127, 173, 121, 113 [See amperages are higher]Average current - maybe about 150 amps, varying from 113 to 198 amps.... continued on next post [5 picture limit]Rick V Attached Images |
|
发表于 2021-9-1 00:14:47