Y connection for mixing gases (maybe fairly accurate with a little physics)

admin

Hi all,This is my first post on here, but I have been reading for quite some time.  Little background about myself, I am a beginnger hobby welder.  Started a year ago or so with a arc, then moved to flux with mig, and recently just bought my first 220V 'real' mig.  This post has really nothing to do with my skills.Anyway onto the subject of this post - let me preface this by saying it went to law school, barely had and barely can remember physics.  So, just in reading around on here I happened to read the debate about using a y connector to mix gases compared with an expensive gas mixer (i.e. Smith proportional gas mixer).  Now, I am not advocating one over the other, I was merely just bored and thinking, so here it goes.From my understanding, there are two issues:  1)  Accuracy of flow meter 2)  Gas mixers include the weight of the gas.Well, in screwing around on the internet (mind you, I am an idiot with this stuff) I looked up Graham's Law, which states the rate of effusion of a gas is inversely proportional to the square root of its molecular weight (effusion being  the process in which individual molecules flow through a hole without collisions between molecules).  This formula is only approximate for diffusion.  Now, I am not sure if diffusion is what happens when a gas is flowing from a bottle, or if it is effusion, or neither or both.  However, Graham's law does take into account the mass of a gas.  Graham's law states Rate1/Rate 2 = Square root of M2/Square root of M1.  M being the molar mass of the gas.  Therefore we can say (I took these numbers from the internet) the molar mass of Argon is 39.95 and the molar mass of CO2 is 44.00964.  The square roots, respectively, are 6.321 (Ar) and 6.6339 (CO2).  6.321/6.6339 = 0.95.Therefore, Argon effuses at 0.95 the rate of CO2.Based on this, since Argon does not effuse at the same rate of CO2, if one wanted to achieve a C25 mixture, one would need to take this into consideration when setting the flow into the y connection.  Meaning a flowrate of 25% to 75% isn't exactly accurate due to the different rates of effusion of the gases.  The Argon would actually need to flow a little more than 25% and CO2 would need to be a bit less that 75% (proportional to Argon) for the 25% to 75% mix.However, that does not me the y connection can still be that accurate because the flow gauges aren't accurate enough to get that precision.  Additionally, the gases might not have time to mix that well in the short amount of time it travels down the hoses to the end of the mig gun.But my conclusion is this, no a y connection will not get you a perfect C25 mix.  I am not sure what +/- allowance a gas supplier is allowed in their mixes from one bottle to the next (probably some welding code on this and I have no idea), but I imagine a y connection with as accurate flowmeters as you can get, would get you fairly close for at least a C25 mix with a +/- fudge factor some person could probably figure.  The point is that one could actually take these issues into consideration, but in all would still be hindered by their equipment.  Any sort of precision type mixture would be out of the question, but I think one could get pretty close to a C25 mix.Anyway, just wanted to share with everyone what I was thinking about today.  LOL I am sure I am going to get my reasoning slammed, but that's the point.  It's fun for me to think about these things.
Reply:I can tell you that a Smith mixer is nothing more than a big pressure equalizer.  It has a large, maybe 6" diameter diaphram that is shared on one side with CO2 and the other side with argon.  If memory serves, there is a middle reference chamber. Mine is balanced to less than .1 psiSo two 100psi sources (single stage regulator) feed into a precision balanceing regulator (about 50psi) which goes to a mixing screw.The actual mixing is done by a srew that opens one side up or the other.  Turn the screw one way its 100% argon, the other way, its 100% CO2.  The difference in calibration from one to the other is the dial itself.  Slightly different markings for different gas combinations.I've added a ball type flow meter on the output.I have an automotive 5 gas analyzer.  It can measure CO2 levels to a max of 20%.  It seems to detect argon as O2.  So by looking at the O2 vs. CO2 levels I can dial in anything I want.I'm actually mixing C25 with 100% argon.  Both are easy enough to get, but that makes my mixture between C25 and C0    vs. C100 to C0.  So the dial means nothing (except for the black sharpie marks on it).   That's a calibrated Sharpie!Con Fuse!Miller Dynasty 350Millermatic 350P-Spoolmatic 30AMiller Multimatic 200Hypertherm PowerMax 1000G3Miller Maxstar 200DX
Reply:Have to add, there are some home-made beer equipment manufactures who are lookng at making cheaper equipment to measure CO2 content.  One that I talked to was thinking about welding gas checker.I have to warn you though, Smith Mixer's, used can run $500 (new something like $1700), two regulators (call if $50 each), whole bunch of hoses, fittings etc.  Call it another $50.  Flow meter at the end - another $100.   Next question, how much gas can you buy, pre-mixed for $750?  Now, do you weld enough to tell if 20% CO2 is better than 16%?  what about 14%?   How about 10% vs. 8%?Not me.  I can tell you that my LWS charges 2x when I ask for 90/10 in a 80 bottle and makes me wait a couple of days (call it a week) to get it.  So its roughly a $40 hit every time.  In two years I think I have saved maybe $200.Con Fuse!Miller Dynasty 350Millermatic 350P-Spoolmatic 30AMiller Multimatic 200Hypertherm PowerMax 1000G3Miller Maxstar 200DX
Reply:I actually did read that thread where someone opened up the Smith and showed the diaphram etc...I was interesting to see how little there was inside.  I, for one, am not experienced enough of a welder to be able to see that much of a difference between the welding gases.  The one thing a y connection could not provide you with is consistency.  Once you have a welder set with a single bottle, you should be good (some variables) to know what is coming the next time you weld with that same bottle.  A y connection would probably make you have to readjust every time.  However, most of the time the metal has different make ups (unless you are welding with the same exact metal/thinckness) every time - so some fiddling would probably be necessary (for me, fiddling will be necessary almost every time I weld).  In my mind I was really just trying to figure out what the expense was with the mixers.  Obviously there are worth the money, but your average hobby guy is never going to buy one.  Not only that, I find myself wanting to weld enough different materials in the small quantities I do to not really want 2 or 3 bottles.  But you are definitely correct when it comes to the cost v. benefit analysis.  Buying some good quality guauges can really set you back, and with the amount of used bottles on Craigslist, it is fairly easy to buy 2 or 3 bottles in different sizes to be able get what you want.  A y connection would definitely not be good for someone who is having welds inspected (I imagine not even in the code).  But, if someone who was just welding as a hobby wanted to pick up 2 medium sized bottles of CO2 and Argon for different metals, as long as you were only going as precise as C25, you could get pretty close and keep the amount of equipment pretty low.  Sure, the bottles will empty at different rates etc...My only point really was that with a y connection, some precision equipment, and some number crunching you could come close to nailing most the variables a machine a mixer does.  Maybe the price is partially for convenience.  I have some precision markings I need to make, can I borrow that sharpie!?!
Reply:It might have been my post with pictures of the internals.Right now on Ebay, there are two mixers that I see.  Search term "Smith Mixer"http://www.ebay.com/itm/SMITH-EQUIPM...item4840f95a87andhttp://www.ebay.com/itm/Smith-299-01...item230ed0ec2aI'm not 100% sure, but it is clear from the pictures that one of the key differences on the mixers is the dial on the front.  Note that the scales are not linear.  One of the mixers is Hydrogen/Argon.  I can imagine that something as light as hydrogen might be significantly different than handling equipment for Argon. Anyway, one of the asking prices is $1000 obo.   I got mine from Ebay for $500.  $500 seems to be a fair enough offer if you want to go that route.  Again, do the math, cost of custom blends vs. blending it yourself.   I'm a bit on the crazy side and have often spent $100s on tools to avoid paying someone to do the job for me.Con Fuse!Miller Dynasty 350Millermatic 350P-Spoolmatic 30AMiller Multimatic 200Hypertherm PowerMax 1000G3Miller Maxstar 200DX
Reply:bigger question is who cares if there is a 1% error? In the practical world where people work and make money, my buddy gets entirely satisfactory results from two flowmeters both plumbed to check valves, run together with a Y fixture. He has another one which uses 3 flowmeters, 3 check valves and a 3-way Y which started out life as an air fitting. Works great. But laboratory accuracy? In a weld shop? Gimme a break.metalmagpie
Reply:There is this old thread by zapster with the inside pics of the mixer that the OP may be referring to:http://weldingweb.com/vbb/showthread.php?t=10050
Reply:Gases mix in a Y connection due to turbulence and diffusion ( effusion refers to gases flowing through very small holes or pores and is really not applicable to this situation.)  I worked in one of the leading welding R&D labs in my early carreer.  One of the engineers who evaluated the best mixes for short arc MIG welding said he found 22 1/2% CO2 in argon was best but they were using two cylinders and regulator flowmeters and a Y connection to the wire feeder and it was easier to set 25% so that is how 25% was defined as the mixture we sold when we started to sell mixed gas in cylinders!  At the time the company I worked for produced about 1/2 of the argon in the U.S.  It may not have been a perfect way to mix the gas but it worked.  I know some large users of argon helium mixtures that still use that approach.  If mixing low percentages go gases, like 2 or 5% O2 then a quality mixer is needed.A regulator flowmeter operates with a fixed pressure inlet to the flow tube followed by  a needle valve.  Typical fixed pressures are a minimum of 25 psi to 50 psi and for CO2 80 psi is often used.  As long as the pressure is above 25 psi the flow is set by the gas velocity in the needle valve.  The velocity reaches the speed of sound and can not increase any further.  That is why you hear thunder after you see lighting! This is called choked flow or compensated flow.  The pressure will drop across the needle valve to what is needed to sustain the gas volume coming through the needle valve.  The reading on the flowmeter tube will be accurate and the gas flow on the two flow tubes is additive.Last edited by Jerry U; 08-12-2012 at 12:14 PM.
Reply:Welcome Jerry!It is important if you mix with a "Y" valve or fitting that the flowmeters are identical in output pressure. Would that be correct? I prefer Victor HRF2425 for use on my tig applications because the output pressures are preset to 25psi. I do pefer the dominant regulator on the Argon bottle to allow the cheaper gas to surge out first. Bleeding in Helium via the "Y" fitting is simple at that point.What I don't understand is how a premixed Argon75/Helium25 premix bottle stays premixed.  Some of my bottles are stored and I worry that they will separate. Do you have any insight on this?Weld like a "WELDOR", not a wel-"DERR" MillerDynasty700DX,Dynasty350DX4ea,Dynasty200DX,Li  ncolnSW200-2ea.,MillerMatic350P,MillerMatic200w/spoolgun,MKCobraMig260,Lincoln SP-170T,PlasmaCam/Hypertherm1250,HFProTig2ea,MigMax1ea.
Reply:Entropy of Mixinghttp://en.wikipedia.org/wiki/Entropy_of_mixingExperience is something you get right after you need it
Reply:Originally Posted by Fred sEntropy of Mixinghttp://en.wikipedia.org/wiki/Entropy_of_mixing
Reply:I would guess that the departure of the system from ideal is small for this particular situation.  The rule of thumb we use is "once mixed, always mixed".  I have tested 75/25 (Ar/CO2)bottles that were ~8 years old, and they were within original filling tolerances.99% of the time gas mix problems are due to incorrect filling procedures in the first place.Experience is something you get right after you need it
Reply:Interesting article on gas stratification in cylinders:http://www.praxair.com/praxair.nsf/0/ff814e89a9b9895185256ff60054af81/$FILE/P-8587.pdf"USMCPOP" First-born son: KIA  Iraq 1/26/05Syncrowave 250 w/ Coolmate 3Dialarc 250, Idealarc 250SP-175 +Firepower TIG 160S (gave the TA 161 STL to the son)Lincwelder AC180C (1952)Victor & Smith O/A torchesMiller spot welder
Reply:Hey guys,  interesting thread.  I am getting frustrated that I cannot get ANY spray transfer gas at any LWS in less than 300CF.  I used to have a couple 30 CF tanks and sold them because I hated moving them around, loading/ unloading etc.    So now I am thinking of mixing my own gas.  I have 3 of the 150 CF tanks and don't mind keeping one as 75/25, but I often want to spray transfer MIG.  I am considering the Y connection method of Mixed gas welding for this.  its for home hobby use only and not a WPS proceedureso it doesn't have to be exact.  95/5 to 84/16 (airgas gold steelmix that I have now)  Do you think if I got a tank of CO2 and an CO2 regulator that this method would work fine for my needs?   as long as I can get into the Spray Transfer range I don't really care if its 8 or 15% CO2Tiger Sales:  AHP Distributor    www.tigersalesco.comAHP200x; AHP 160ST; MM350P,  Spoolmatic 30A; Everlast PowerTig 185; Thermal Dynamics 60i plasma.  For Sale:  Cobra Mig 250 w/ Push-pull gun.  Lincoln Wirematic 250
Reply:There's another thread(s) out there that covers this subject.  FWIW, I built a Y-mixer and have floating ball flowmeters on my two tanks; C25 and Ar.  To get about 30 CFH of C8, I set the C25 to read about 10 CFH and the Ar to 20 CFH.  I was able to make nice spray arc welds at about 27.5V and 550 fpm WS using .030 wire.  Not exactly kosher, but it works fine for me.- "If ya can't be handsome, ya may as well be handy!"   HTP Invertig 201Lincoln Power Mig 255CLincoln SP125+Thermal Dynamics Cutmaster 38
Reply:the idea of getting an exhaust gas analyzer seems a good idea too.  wanted one for years anyway.  then just set your co2 level for that session at whatever % you want after setting the Argon flow.   Then its just opening the valve and dialing up the CO2 valve to the needed flowTiger Sales:  AHP Distributor    www.tigersalesco.comAHP200x; AHP 160ST; MM350P,  Spoolmatic 30A; Everlast PowerTig 185; Thermal Dynamics 60i plasma.  For Sale:  Cobra Mig 250 w/ Push-pull gun.  Lincoln Wirematic 250
Reply:Originally Posted by shovelonWelcome Jerry!It is important if you mix with a "Y" valve or fitting that the flowmeters are identical in output pressure. Would that be correct? I prefer Victor HRF2425 for use on my tig applications because the output pressures are preset to 25psi. I do pefer the dominant regulator on the Argon bottle to allow the cheaper gas to surge out first. Bleeding in Helium via the "Y" fitting is simple at that point.What I don't understand is how a premixed Argon75/Helium25 premix bottle stays premixed.  Some of my bottles are stored and I worry that they will separate. Do you have any insight on this?