Appx wattage needed for solding aluminum tubing with soldering iron?

[pipe2null]

I'm looking for a soldering iron with a high enough wattage and temperature to solder aluminum tube, simple square tubing 8mm x 8mm with 1mm walls.  The purpose of the soldering is to get air/water tight joints, but not under any significant pressure.

Some of the solder/brazing rods I've used melts around 440 degC, others is more into the 550 degC range.  Aluminum melts around 660 degC.  I've attempted using my little soldering station, but all the heat gets leached away from the joint I'm soldering, and the iron can't provide enough heat energy fast enough to get the joint up to temperature.  Previously I've used propane/mapp/oxy-propane torches to braze joints in aluminum tubing, but I'm out of gas at the moment so it's a good time to explore other options, preferably an all electric one.  As much as I like playing with fire, literally, maintaining tanks and a safe area to use them is a bit much when the sole purpose for the whole setup is to braze small aluminum parts.  I've considered getting a reflow heat plate to pre-heat the part, but that would still need a high enough wattage soldering iron to overcome the leaching, and I'd prefer to only need an iron (a single tool versus heat plate station plus tool versus tanks and torches and torch tips and.......).


Any advice on soldering these types of joints with a soldering iron?  Any idea what wattage is needed to overcome the heat leaching and still have a high enough temperature to get a good enough soldered joint?

Thanks

[Fraser]

This article states 150W may be enough. I have only ever seen a gas torch used though. It will be the size of the workpiece and it’s ability to conduct the heat energy away from the joint that dictates whether an electric soldering iron is even feasible.

https://www.wikihow.com/Solder-Aluminum#:~:text=You%20can%20use%20a%20soldering,a%20150%20watt%20soldering%20iron.

Fraser

[Fraser]

Useful article…

https://superiorflux.com/papers/TWJ.May.2018.Tech.Solder.Al.pdf

[mag_therm]

The thermal conductivity of Al is so high that the whole object tends to rise to nearly the brazing temperature.
So the heat source has to be of sufficient power to do that with a bit left over to do the braze.
It might be possible to use a soldering iron rated for 400+ C if the object is placed on a heating pad that can hold it at about 360C.

I use oxy acet #1 tip - quite a lot of heating power for small jobs compared to steel.
 The brazed joints need some stirring of the melt to get the oxide to float out, necessitating rigid clamping that also sucks the heat away.

As an aside, I found that Al sections from the rack in the home hardware stores have some impurity that does not allow reliable brazed joints.

So now I only braze 6061 alloy sections purchased from metal suppliers.

I can braze moly bike tubing quite well, but this Al stuff is not easy for me!
https://app.box.com/s/7q6jydidh7e4p94tfgnece3sropneq46

[pipe2null]

Great article on all the various methods for soldering Al.  And nice joints on that chassis (I swear I'm not being lewd).

The stock I'm using is at least advertised as Al 6063.  When I was previously using the oxy-propane torch I could get a decent braze on one joint, but the other joints were too close so when torching the second joint it would end up with enough heat leached back to the first joint to melt the solder and screw up the joint.  Which directed me into a rabbit hole of attempting to learn TIG welding of thin Al from youtube videos using a welding machine at my local Maker's Space, but then I got busy with life and the whole project got back burnered for a long while.  Some of my original prototype heat exchangers have been in daily use for over a year and a half without leaking and frying motherboards or GPUs (knock on wood).  I have extra time ATM, thought it was time to finally get the overall project done.

Thought it might be possible to take care of the problem of joints being too close plus eliminate the need for a torch by going all electric, but the fundamental heat conductivity problem looks to be unavoidable for tools within my price range.  I am curious about induction soldering, but that seems like it would be expensive.

Perhaps it's time to refill my tanks and then finally buy one of the TIG machines I've been drooling over for a long time.  Me tinks a few TIG tack welds plus brazing should get pretty decent very usable results, at least until my TIG skills evolve beyond burning holes in metal coupons.  (I jokingly refer to TIG welders as plasma cutters, but hopefully not for much longer.)

Thanks for the feedback!

[wizard69]

Sadly I'm going to have to say that most soldering irons will not even come close.   Not know the specifics of this joint I do have a suggestion worth looking into, and that would be inductive bolt heaters.   You could also look into induction heaters specifically for this usage: https://ultraflexpower.com/induction-products/?campaign=853295958&adgroup=44243703555&keyword=induction%20heating%20equipment&creative=216812048600&gclid=EAIaIQobChMI4-LQwYHI-AIVtTizAB33GQY1EAAYASAAEgLhAfD_BwE.   The induciton bolt heaters are like these: https://www.amazon.com/s?k=heat+induction+bolt+remover&gclid=EAIaIQobChMI4-LQwYHI-AIVtTizAB33GQY1EAAYAyAAEgIjS_D_BwE&hvadid=521278154588&hvdev=c&hvlocphy=9067609&hvnetw=g&hvqmt=e&hvrand=3195435156427248570&hvtargid=kwd-915459744187&hydadcr=29041_10165998&tag=googhydr-20&ref=pd_sl_8t27f3xahf_e which gives you 1000 watts of power.   I only mention these units because the dimensions mentioned might allow for success.   If you are dealing with closed circuits then you might have to become creative with the induction loop.   I'm sure if you look around a bit you will find more of these and similar designs for brazing tubing.

As for the issue of joint being pulled apart when moving to a different side, my suggestion is don't!   come up with a system that heats up the entire perimeter all at once and melts the braze filler into the joint all at once.   One of the advantages of induction heating is that you can heat an entire diameter evenly.   With the joint prepped with flux and a snip of wire wrapped around the entire joint, getting good flow of the braze material should be easy.   Done right the braze materiel should all flow into the joint at the same time.

Given all of that it might be easier just to go out and buy another gas bottle for the torch you have.   If you need more back ground this might help: https://www.efd-induction.com/en/induction-heating-applications, maybe a bit too high end but good info.   I actually believe that inductions systems are now avaiable for plumbers and pipe fitters to do braze or soldered joints.    At least I seen such advertised at one time.

[mag_therm]

I am using Forney 46111 rods which seem the same as Harbor freight "Alumiweld"
387 C 730 F
There is very little mushy zone and the melt flows like , say, milk.

The joints have to be nearly horizontal. 20 degrees off and the braze would run away.

The "close joint" problem means that only one side of a braze can be done.
Both side of an inverted T joint can be done if the base is horizontal.

If trying to do a double bevel butt joint, when turning it over to fill the other side, the first side just runs away.

To prepare a butt joint then, it means that a single bevel is made and then filled up from one side.
But that leaves a stress concentration on the other side.

Also, to get a good joint, the surfaces under melt have to be stirred or "scarified" to release flakes of oxide which end up as a dull finish on top of the braze.
(a shiny finish might mean a bad joint)
 I find that trying to use the rod to do that, results in too much filler,
because , of course the rod continues to melt. So I use a sharpened tool steel held in the vice grips to stir.
The parts have to be solidly clamped otherwise they will move during the stirring.

Those issues, and the metallurgical problems described in following paper
mean that (for my usage and methods) this type of Al brazing is only useful for low stress jobs , like non critical electronics hardware etc.
search: "Dai_et_al_Microstructure_and_Properties_of_6061_Aluminum_Alloy_Brazing_Joint_with_Al–Si–Zn_Filler_Metal_en.pdf"

[Doctorandus_P]

One of the problems with trying to work with an iron is to get the heat transfered form the tip to the aluminium, and I doubt it's going to work.

What you could try is to put some wet rags on (or behind) those other joints to cool them down enough so the solder won't melt again. Thiw sill create big temperature gradients and thus you will need a bigger flame to compensate for that.

Yet another option is to make a jig that holds all the parts in place, then solder the joints and take it out of the jig after cooling.

Another option is to use a smaller but hotter flame to put more localized heat right at the right spots.

[BrokenYugo]

The electric tool for this job is a TIG welder, not a soldering iron.

If you can't see the puddle, get a good enough adjustable auto darkening helmet (sensors above and below viewing window, user serviceable battery) so you can. Biggest thing about learning to weld well is to see what you're doing, if you can't see the puddle, metal it is or is not wetting on, crack you're following, etc. you can't make corrections on the fly.

Those brazing rods are handy, but they're only so useful.

[langwadt]

The electric tool for this job is a TIG welder, not a soldering iron.

If you can't see the puddle, get a good enough adjustable auto darkening helmet (sensors above and below viewing window, user serviceable battery) so you can. Biggest thing about learning to weld well is to see what you're doing, if you can't see the puddle, metal it is or is not wetting on, crack you're following, etc. you can't make corrections on the fly.

Those brazing rods are handy, but they're only so useful.

needs to be AC TIG and not all types of aluminium can be welded

[pipe2null]

Thanks for all the feedback.

Looks like I am officially in the market for my first TIG machine to tack joints together and refilling my tanks to use my existing mini jewelers mixed gas torch (Smith's "Little Torch") to braze joints to get water-tight-ness.  My thought is that as long as the joints are tacked together with TIG (solid up to 660 degC), soldering many joints should not be an issue since even if solder in previous joints re-melt when the whole part goes above 450 degC, surface tension will hold molten solder in the joint as long as the pieces are immobilized by the TIG tack welds.  Obviously, it would be better if my TIG skill were good enough just to weld the whole piece, but that's nowhere near realistic yet.


Currently diving into the many variations of welding machines that include TIG.

[Marco]

Can't TIG welding work pulsed too BTW? Basically doing overlapping spot welds which don't heat up much of the material.

This doesn't seem to generate much heat in the workpiece :
https://www.youtube.com/watch?v=_3IwnplHltg&t=530s

[langwadt]

Can't TIG welding work pulsed too BTW? Basically doing overlapping spot welds which don't heat up much of the material.

This doesn't seem to generate much heat in the workpiece :
https://www.youtube.com/watch?v=_3IwnplHltg&t=530s

https://youtu.be/V7-mcjR59a4 ...

[Marco]

It only needed to be water tight, not structurally sound.

[Ground_Loop]

You might want to check out irons used for stained glass such as Weller W100P or 1175.

[pipe2null]

I'm curious about the so-called "Cold Weld" mode, although as The Fabrication Series points out, it is just a psuedo pulsed tack weld mode that goes full off between pulses, and "Cold Weld" is literally a welding defect due to really poor penetration usually referred to as a cold weld. 

I still would not trust overlapping tack welds to be water tight, but I am curious how to simulate that "Cold Weld" mode with a real TIG welding machine... Might still be useful for me until I get my TIG skill good enough and still rely on brazing for water tightness.  Do you just unplug the pedal and tape a clicker to the torch?  Is it still using Argon at normal flow?


Any recommendations on web forums for buyer's advice for TIG/multi-process welding machines?  There's always people on Eevblog for any subject you can think of, but I hesitate to ask general questions here that are as far removed from electronics as multi-process welding machine purchase advice.

[Someone]

If the object can withstand some heat, why bother with a high-skill method like welding? Adhesives such as Permabond ES558 can be cheaper, or marine silicone....

[BrokenYugo]

IIRC it is pretty typical for modern inverter TIG machines to offer at least some waveform shaping options, check the manual.

[pipe2null]

I did some early prototype parts using JB weld.  Most of the parts ended up working out ok, BUT: One prototype appeared to have good, fully cured joints.  I pressurized the part with an air compressor, submerged it, and it passed the test with no bubbles.  I ran it with coolant for a couple days without any power applied to the electronics, no problem.  2 days after I started using the JB welded part with power applied to the electronics, the bloody thing developed a small leak and literally fried a dual CPU server motherboard, complete with slag surrounding the VRM that blew up as result of the puddle.  Fortunately both CPU's and the RAM survived, but it was a hard lesson to learn, so, no more JB Weld type compounds for doing these joints.  Thus my transition to brazing and now adding TIG variations to my fabrication skill set.  Theoretically the same problem could occur with TIG-tacked and brazed joints, but my confidence level is much higher with an all metal joint that passes the same pressurized air submersion test for water-tightness.

Plus, I must admit that part of my desire to go the TIG route is simply that I've wanted to buy welding equipment to add to my shop for a while, but lacked sufficient justification for the purchase.  Still, for this project a nice and simple solution would be preferable, but it seems the TIG route would get the most reliable result with equipment purchases that are more general purpose, general use welding equipement compared to something more specialized with little use for me other than this specific project.

[mag_therm]

Sorry to hear about that disaster.
Are you aware that thin wall copper tubing is available?  Copper greatly eases the difficulty of joining Aluminum and also has  corrosion resistance  with water.

Thin and thick wall rectangular copper tube is  used for high pressure water cooled coils and conductors in induction heating.

[BrokenYugo]

A good point, if we're just doing typical computer water cooling then soft soldered copper is certainly an option, more costly material, but higher performance and easier/cheaper to work with.

[Old Goat]

Thanks for all the feedback.
...

Currently diving into the many variations of welding machines that include TIG.

Take a look at Kevin Caron on youtube. He's a metal artist, and offers reviews of various TIG welders as well as many tips and procedures for welding various materials.

[wizard69]

If that is what you are doing I'd have to say STOP.   If aluminum tubing and brazing was a reasonable solution for any plumbing need we would be seeing more of it in use.   There are some systems for airlines out there that use aluminum successfully but they also leverage compression fittings.

I've seen a lot of these computer chilling systems over the years and have to wonder why they don't use common industrial solutions that are known to work well.   Swagelok is one approach and there are all sorts of plumbing type components for refrigeration systems.   Swagelok is frankly pretty easy to use and can interface to stainless, copper and a variety of plastic hoses.   

If you go with copper there are so many reliable ways to make up connections that there is almost always a reliable commercial solution available.   That could be soldered or brazed connections or some sort of compression fitting.

I did some early prototype parts using JB weld.  Most of the parts ended up working out ok, BUT: One prototype appeared to have good, fully cured joints.  I pressurized the part with an air compressor, submerged it, and it passed the test with no bubbles.  I ran it with coolant for a couple days without any power applied to the electronics, no problem.  2 days after I started using the JB welded part with power applied to the electronics, the bloody thing developed a small leak and literally fried a dual CPU server motherboard, complete with slag surrounding the VRM that blew up as result of the puddle.  Fortunately both CPU's and the RAM survived, but it was a hard lesson to learn, so, no more JB Weld type compounds for doing these joints.  Thus my transition to brazing and now adding TIG variations to my fabrication skill set.  Theoretically the same problem could occur with TIG-tacked and brazed joints, but my confidence level is much higher with an all metal joint that passes the same pressurized air submersion test for water-tightness.

Plus, I must admit that part of my desire to go the TIG route is simply that I've wanted to buy welding equipment to add to my shop for a while, but lacked sufficient justification for the purchase.  Still, for this project a nice and simple solution would be preferable, but it seems the TIG route would get the most reliable result with equipment purchases that are more general purpose, general use welding equipement compared to something more specialized with little use for me other than this specific project.

[langwadt]

If that is what you are doing I'd have to say STOP.   If aluminum tubing and brazing was a reasonable solution for any plumbing need we would be seeing more of it in use.   There are some systems for airlines out there that use aluminum successfully but they also leverage compression fittings.

I've seen a lot of these computer chilling systems over the years and have to wonder why they don't use common industrial solutions that are known to work well.   Swagelok is one approach and there are all sorts of plumbing type components for refrigeration systems.   Swagelok is frankly pretty easy to use and can interface to stainless, copper and a variety of plastic hoses.   

yes swagelok is very nice, but also very expensive

[beanflying]

Won't try and talk you out of a good Tig welder as I talked myself into one late last year 

For Aluminium Brazing it's all about heat volume more than absolute temperature. Have a trawl through this repair I did a while ago on a badly broken Cast Ali frame. https://www.eevblog.com/forum/testgear/hewlett-packard-740b-dc-standard-digital-voltmeter-(and-740a)/msg1420035/#msg1420035

[mikerj]

Currently diving into the many variations of welding machines that include TIG.

The cost of the machine is just one aspect, make sure you figure in the cost of and availability of the shielding gas (bottle rental and refills) and a suitable grinder/grinding wheels for the electrodes (you will contaminate frequently when starting out).

[beanflying]

The Rental is far less an issue now than it used to be. Here in Oz finally we have several pay up front for a bottle and no rental, this was always a major deterrent in the past and even more so now. Bottle payback to rental costs are under 4 years then you are clear. You do pay a little more for Gas over Rental rates but for home or low volume use it still makes more sense to buy.

Way back in one of my first jobs I worked for a Welding Gas and Gear centre and bottle rental and handling was a duopoly locally and mass extortion of tradies for lack of options (30+ years ago)

[pipe2null]

Won't try and talk you out of a good Tig welder as I talked myself into one late last year 

Oh...  Don't you worry, I need no encouragement, just (barely) sufficient (excuse) well researched justification (really just a good excuse) for diving into the welding side of things.  I already had some of the TIG basic gear, and earlier today I hit a holiday sale at harbor freight to pick up most of the rest: heavy duty welding cart, baby starter welding table, basic magnets/clamps, and misc other stuff.  Added a few MIG and stick specific items to the mix as well (not for my current project, but future use).

@mikerj:  Good points.  I already have most of the basics for TIG, including one of those tungsten grinder attachments for a dremel, from when I was messing around with a TIG machine at a local Maker's Space.  The previous gas supplier I worked with for my oxy-propane torch charged full price for tanks but you just brought them in and swapped them and only pay for the gas, or just hang out for a bit while they refill the same bottle.  I have since moved and haven't selected a new supplier so no sure yet what the locals offer.

@wizard69:  Thanks for the tip on Swagelok, I'll check them out.  My water cooling project is mostly about seeing if I can do it successfully when no off-the-shelf part exists or is ridiculously expensive (commercial CPU water blocks for narrow ILM server CPUs are about $120 to $200 per CPU socket per dual CPU server last time I priced them), and with the exception of one fried motherboard out of the stack of 4 servers, it has worked out more or less ok.  All these machines are old retired enterprise servers and no one in their right mind would put in the massive effort (CAD, 3DP, fabrication, etc...) to water cool them unless it's a project, or like me, want my servers running very quietly next to my workbench instead of a sound-proofed closet far far away...
If these were newer servers or any consumer grade desktop with plenty of off the shelf solutions, I would definitely NOT go this route.

General FYI: I went with Aluminum literally because I could get 8mm square aluminum tubing on Amazon and didn't easily find available copper square tube around that size.  Using square tubing because the flat sides are convenient for heat transfer from other flat things.  True, connecting square metal tube directly to round silicone tubing is a PITA, but doable.  Plus going the aluminum route is CHEAP at least for materials.  Example, I built one heat exchanger that uses an Al plate as heat spreader brazed to the square tube, and this cools 2x Dual Ten gigabit ethernet chips (4x 10 GbE).  I had to 3DP a bracket to mount the exchanger on the riser card that had no heat sink mounting holes since the original heat sink was epoxied onto the glass flip chips (PITA to remove). Total part cost was about $5.


Please feel free to suggest welding machines I should look at before I finalize my purchase...  I'm currently looking at the Weldpro Omni 210, but there are other machines I'm considering...  My budget is somewhat flexible depending on bang for buck, but I'm not buying any cheapo machines (<insert FIRE emoji here>).  The Weldpro Omni 210 is about $1500, but I could probably go up or down if there is good reason.  Miller Multimatic's are attractive, but I can't justify $4k.  I messed with a Lincoln Electric Square Wave TIG, but I can't justify $2.4k for a machine that only does TIG...  Etc...

Since I'm diving in the welding side and not just upgrading my lab/shop the absolute minimum for the project at hand, I want to make sure the welding machine(s) I get can handle multiple processes and all materials reasonably accessible to consumers, realistically for me would be mostly aluminum, but steel and possibly copper would make sense as well.  I gave up trying to find a multiprocess machine that does AC/DC TIG, MIG + spool gun, stick, and Plasma cutting, (flux core as nice to have).  They have machines that do just about every combination of those processes but they all lack at least one.  For example, there are multiple makes and models that do all of the above processes including plasma cutting but only can do DC TIG not AC...  It's irritating AF to find a bunch of machines that ARE SO CLOSE to a complete multi process machine but stop short.  So, I'll get a plasma cutter separately as a future fun purchase (with distinct urge to pair with CNC table) and continue looking for a single ACDC TIG / MIG + spoolgun /stick multiprocess machine that performs decently all the way down at 5 - 25 Amp AC TIG for very thin aluminum (<= 1mm), ideally with pulse and some of the more interesting waveform manipulation features that might be useful for really thin aluminum.  But still beefy enough for normal material thicknesses so the investment is still useful for other projects already bouncing around the back of my head.

Sorry about long post, I'm a little amped up about being so close to buying a whole new genre of toys (horrible pun intended)...  heh.

[beanflying]

As you are in the US have a good look at the Everlast range too. Plenty of reviews on Youtube for the various offerings.

For my uses I went with a more dedicated Tig unit that will do Stick and chose not to get anything MIG capable in the box. The thinking was that if I ever found myself in the position to need a MIG for bigger jobs then there is plenty of used quality more industrial gear (15kg spools) to consider. Flux core MIG and small spools is something I decided I can do without and initially I didn't want a second bottle of gas either.

Unloved thread I started some time ago I must post some of the adventures in. But it has some good links in the first post that might help https://www.eevblog.com/forum/mechanical-engineering/welding-stick-tig-and-mig/

[Siwastaja]

Use copper for computer cooling stuff. Better thermal conductivity, much easier to work with, can be soldered with standard tin. Can be welded with DC TIG which is cheaper than AC. Brazing with copper phosphorous alloy stick is also quite easy, I haven't done aluminum but I'd assume it's more difficult. So with copper, everything is better, except the price of the raw material itself, but at the scale of some small computer cooling system, does it matter?

[PaulAm]

If you have the space (and power), you may be able to find older TIG gear cheap.  I picked up a Miller Syncrowave 300 for $85 at an auction a couple years ago.  That's an awesome machine.

You could probably find a spool gun and use MIG, even one of those cheapie inverters, and get decent results for what you're doing.

[pipe2null]

As you are in the US have a good look at the Everlast range too. Plenty of reviews on Youtube for the various offerings.

For my uses I went with a more dedicated Tig unit that will do Stick and chose not to get anything MIG capable in the box. The thinking was that if I ever found myself in the position to need a MIG for bigger jobs then there is plenty of used quality more industrial gear (15kg spools) to consider. Flux core MIG and small spools is something I decided I can do without and initially I didn't want a second bottle of gas either.

Unloved thread I started some time ago I must post some of the adventures in. But it has some good links in the first post that might help https://www.eevblog.com/forum/mechanical-engineering/welding-stick-tig-and-mig/

After digging into it and resisting the urge to go with cheapie burn-down-the-house brands, and being too poor for brands like Miller or Lincoln Electric, I ended up buying Everlast:
- TIG/stick: PowerTIG 255EXT
- MIG/stick/Plasma: Storm 215C w/Spool gun for aluminum.
Will be delivered in a few business days.

Should be pretty solid across the board for TIG from thicker stuff all the way down to ultra thin stuff with it's minimum AC start at 5A, plus other bells and whistles that I've read help with thin material.  I did not "purposely" go for a dedicated TIG machine over a multi process one like the Everlast PowerPro or LightningMTS, but it ended up that was since the PowerTIG 210EXT and 255EXT are a little better for the thin stuff. 

I picked up the Storm to round out welding capabilities with MIG and make sure I had entry level plasma to cut up scrap for practicing.  Plus I have nephews interested in welding, thought MIG would be much better for them to start with so they don't lose interest too fast due to the pain of learning TIG and then go back to playing video games 247.  Overall should be a pretty solid starter setup.  I would have actually preferred to go with separate machines for everything, but the combo unit ended up being a little cheaper than dedicated units and I care less about MIG than TIG.

If you have the space (and power), you may be able to find older TIG gear cheap.  I picked up a Miller Syncrowave 300 for $85 at an auction a couple years ago.  That's an awesome machine.

You could probably find a spool gun and use MIG, even one of those cheapie inverters, and get decent results for what you're doing.

Yea...  I blew up the budget buying everything at once, new, but now it's also done with all at once, so that's a big bonus.  I will definitely keep on the lookout for used Miller units, but when you're a beginner like myself, it's sketchy to buy anything used without taking a welder buddy along who knows what they're doing, which I don't really have at the moment.  Good tip on the spool gun, I picked one up and will try it out.  I am in the process of installing 240V in the garage, but these dual voltage machines will run at half their max amps on 120V until I finish the install and get someone out to add the new 240V circuit to the breaker box.  Well, we'll see what amps are actually usable on these dual voltage machines when I start tripping 120V breakers...   

Use copper for computer cooling stuff. Better thermal conductivity, much easier to work with, can be soldered with standard tin. Can be welded with DC TIG which is cheaper than AC. Brazing with copper phosphorous alloy stick is also quite easy, I haven't done aluminum but I'd assume it's more difficult. So with copper, everything is better, except the price of the raw material itself, but at the scale of some small computer cooling system, does it matter?

I have a helllllllllloooooooooovvvvvvvvaaaaa lot of TIG practice to do before I can actually finish my mini aluminum heat exchanger projects, but I'm moving forward on it again at least.  Will also play with copper too as my TIG skills improve.  And brazing with TIG instead of gas torches...  So many things to try with new toys.     I've added copper welding coupons to my shopping list.  Originally went with aluminum for cooling project due to cost versus thermal performance, then I decided I would rather use copper to get the extra 20-25% cooling perf, then I couldn't easily find copper square tube and ended up going back to aluminum since it was easily available on Amazon.

[PaulAm]

You can use argon for TIG and the aluminum spoolgun MIG welding, but you'll probably want a MIG blend for steel if you do any significant amount.  They add some CO2 and O2 to get better penetration and improve arc stability..  Two gas bottles, oh joy 

I use Q (80 ft^3) tanks for my MIG and an S size (120 ft^3) tank for argon.  I wouldn't get anything smaller than the Q size, but it will last for quite a bit of welding.

[Ground_Loop]

Wow! You blew right past the soldering iron idea. I have a similar collection and find more need to weld things than I would have ever thought. I think my mower deck has gained about five pounds in weld bead over the years.  Anyway, if you don't already have one, a metal cutting bandsaw is indispensable.

[pipe2null]

Wow! You blew right past the soldering iron idea. I have a similar collection and find more need to weld things than I would have ever thought. I think my mower deck has gained about five pounds in weld bead over the years.  Anyway, if you don't already have one, a metal cutting bandsaw is indispensable.

Yea, the high powered soldering iron idea was sort of a last ditch effort to find a much cheaper indoor-friendly solution I could use in my carpeted lab in nice comfy air conditioning, and just slap down a large silicone mat as the only thing needed for safety and to protect my wooden workbench.  At the end of the day, I just needed to exhaust all reasonable alternatives to justify the excuse to buy new more expensive toys, errrr, I mean TOOLS, because I've wanted to buy them for a long time, errr, I mean because they're the right tools for the job.  Eventually I might come up with a safe, well ventilated, and fire shielded mini work space for indoor low-amperage TIG of thin aluminum.  But yea...  not going anywhere near that until I'm proficient with the process, or just quit whining and get used to the heat of working in the garage.  Wah.  Horizontal bandsaw is definitely on the list once a budget exists again, and poor poor me, I'll just HAVE to struggle through and make due with plasma cutting in the meantime.   

You are quite right on the "how did I survive without a welder before" notion.  The moment I informed my dad that I was buying welding equipment, suddenly now he needs an oversized  metal toolbox for the back of his John Deer.  Heh.  And of course, every new project idea I've had in the past week somehow includes excuses to weld some part of it.  Do I really need to use a welded aluminum enclosure for my 3D printer resin curing box project?!?!?  NO, but I'll probably weld an enclosure anyway or at least aluminum UV reflectors.

You can use argon for TIG and the aluminum spoolgun MIG welding, but you'll probably want a MIG blend for steel if you do any significant amount.  They add some CO2 and O2 to get better penetration and improve arc stability..  Two gas bottles, oh joy 

I use Q (80 ft^3) tanks for my MIG and an S size (120 ft^3) tank for argon.  I wouldn't get anything smaller than the Q size, but it will last for quite a bit of welding.

Thanks for the tips on tank size, that purchase/rental is happening in the next few days, so perfect timing on the tip.  I'll probably try to avoid rentals, but don't know what the local suppliers offer yet.  I'll probably stick with aluminum to start for both TIG and MIG so I don't need to deal with the second tank yet or figure out what gas mix will be ideal for general purpose MIG, many many learning curves.  But perhaps I can work out a deal with the supplier?!?

Crossing fingers the gas supplier is beginner friendly.  The last time I went into a welding supply shop, as a total newb, to pick up oxy-propane torch parts, the good ol' boys running the place sent me to another shop down the street for the tank adapter, and the next place said they were out of stock and sent me to a third place a few blocks away, and the third place said I could mail order the adapter or go pick one up at their other shop in the town next over.  Before going out of town to pick up the adapter, I finally fiddled with my regulator and the propane tank enough to finally figure out that pretty much EVERY propane tank's valve stem has external threads for your gas grill kinda connector, but it also has INTERNAL threads that will DIRECTLY connect to an acetylene/propane regulator (pretty sure CGA510), no adapter required.

[coppercone2]

this is ridiculous, use a flaring tool to make a decent fit and solder it with a damn blow torch. If you are just soldering with aluminum solder its not so bad. braze is bad for Al, solder is ok. Solder it with the special flux that comes with the aluminum solder.

[PaulAm]

AFAIK, gas tanks in the US are purchase and then refill by exchange.  I tend to use AIrgas, but my local suppliers have no problems exchanging tanks.  Heck you can even buy them through Tractor Supply, but you'll probably get a better deal from a local place.  They might call it a one time lease payment or something.   It's been a few years since I had to get an additional tank.  The good news is you never have to worry about recertifying , just swap the tank out..

Try to be educated before going into a welding supplies store.  They have to deal with the unwashed public who insist on loading filled propane or acetylene cylinders in passenger cars (not quite so bad with a nonflammable gas, you only suffocate in that case instead of getting blown to bits), so try to not look like a total noob.  Get 5 points for showing up with a pickup.

Years ago I had to lease acetylene and oxygen tanks; that adds up after a couple years..

Oh, btw, they generally do not like to swap say an argon tank for a Mig mix, even though they look the same.  I did swap an extra argon tank once for a Mig mix, but it took a little cajoling, so eventually you'll want 2 tanks.  Mig is great for quick work, Tig is much slower.  That tool box will go a lot faster with Mig, but it might have prettier welds with Tig.

[pipe2null]

@PaulAm:

Very good advice, thanks!  I will put that to good use in next couple days.

this is ridiculous, use a flaring tool to make a decent fit and solder it with a damn blow torch. If you are just soldering with aluminum solder its not so bad. braze is bad for Al, solder is ok. Solder it with the special flux that comes with the aluminum solder.

You can chalk it up to inadequate skill on my part.  You are correct about the blow torch soldering, which is exactly how I was doing it for some parts, I even got a cute widdle jewler's oxy-propane torch for it.  Torch soldering a piece with one joint worked well even when the whole piece was also soldered to a heat spreader plate.  But for pieces that had multiple joints 2cm apart, with each joint needing solder on 3 or 4 sides of the 8mm square tube, this is where I had problems with that process.  The moment I started working on the second joint, the first joint filler (less than 2cm away) would remelt and form holes as well as leak into the inside of the tube creating difficult to remove obstructions.  I tried adding heat sinks between joints, but that just made really hot bits of metal stick out from the workpiece as previous joints remelted anyway.  Perhaps I could have done a better job prepping and jigging, but using TIG to at the very least tack the joints together should yield a much better result than solely soldering/brazing, when the joints are so close together and need solder on multiple sides.  And as my skill gets better, TIG welding the entire joint is ideal.  Flaring tool is still necessary for some of the connections.

[coppercone2]

i thought you wanted to solder a tube inside of another tube, this is impossible to understand without pictures (this is standard plumbing even if its square)

if you want it at a right angle or something its gonna piss you off and be weak. the only way to make it reliable is to basically remake it so its a tube inside of tube in some way (milled slot) IMO. I think it will need to 'nest' to be any good at all.

[mag_therm]

this is ridiculous, use a flaring tool to make a decent fit and solder it with a damn blow torch. If you are just soldering with aluminum solder its not so bad. braze is bad for Al, solder is ok. Solder it with the special flux that comes with the aluminum solder.

Can you put up the name of the special flux and the name of the aluminum solder you use?
What 'blow torch" are you using?

This link is maybe useful and refers to > 300 C parent temperature.
https://superiorflux.com/techniques-for-soldering-aluminum/

[coppercone2]

the one I used is mg460 by messer welding, can't seem to find

i think this is the same one, at least temperature wise
https://www.harrisweldingsupplies.com/harris-al-solder-500-aluminum-solder-kit-500k/?sku=500K&gclid=Cj0KCQjw8amWBhCYARIsADqZJoUaf-2WaUmhgJ0aGdVOUwQ96q3KwH77Qxmpy3PzcYKg45N1FkrMM_YaAmWGEALw_wcB

[mag_therm]

Thanks Copper,
At $30/ounce it probably has high silver content where small amount is sweated in a joint as you mention.

To do a chassis with heatsinks I do butt joints needing about 1 to 2 filler rods per chassis.
Forney Al brazing rod is about $22 per half pound of 10 rods, so I use $2 to $4 per chassis.
The 6061 Al angle price has risen too so I would use about $12 of Angle  per chassis.

I could buy  nice, smooth, square, Hammond box for a few $ more !

[coppercone2]

well the secret to aluminum solder I think is Zinc, not silver. A high silver solder has too high  melting point, that is 500F. The idea is also it makes stronger joints by not being so hot because if you are careful heating the metal up the grain structure is not deformed as much, so for real world use, the difference between KPSI is not the only factor to look at. I bet there is a lot that goes into it.

Do you really need that much solder? Can you work on the fit up being better to use less? Its less strong as you get more.

The brazing rod for aluminum works like 750F, the solder I linked is significantly lower temperature, 250F is nothing to scoff at with aluminum. Saves gas/time too, and there is less dimensional change.

Like I said, I had excellent results with proper fit aluminum solder joints, but basically dog shit when it came to aluminum brazing, even with harris flux.

Again this is just for Al, for structural copper stuff, braze. Plumbing appears to have its nuances but engineers IMO will benefit from silver braze.

[coppercone2]

Like if you were tortured by lack of flow that led to excessive braze use, what I am saying is that aluminum SOLDER (500F kind) actually seems to flow. With braze I was eating rods too because that shit would just not go in . The solder should literarly creep in between the joint properly, you should not be making much of a fillet with this stuff! So long your joints are clean, tight, and you follow the procedure (do not over heat the flux), it should actually wick in. Once you start burning that flux (it has lots of organics if you look at the SDS), then you get soot and it basically turns off wicking. You gotta be gentle and thoughtful with the torch.

I think you should give it a try to make sure your solder consumption is as high as you think. The strength of the joint comes from the lap area, the 'fillet' IMO is brittle crap that gives a false sense of security.

 Again, I just COULD NOT get the braze to work like the solder does. I will say it a million times because I basically get heart palpitations thinking about doing an aluminum BRAZE job. When I got it to fill a clean hole by 65% that I felt good about putting a epoxy patch over it it felt like a miracle. I had the braze rods from the welding store that claimed to be prefluxed with some arabic styling on the text. I still have them and I basically know if I want to end the day pissed off at 2am to give it a try.

And for heat sinks, I think the high flow of aluminum solder will help you. The hard part will be good even regulated heating to prevent flux burning. Make sure the stuff is sand paper on glass flat before you solder to minimize the amount of solder you need. And it needs to be cleaned from oxides using a clean stainless steel wire brush before soldering and wiped with alcohol before fluxing.

[mag_therm]

Hi Copper, I am doing the classical single V or single U butt joints , both as shown in this welding link.
https://welditu.com/welding/tips/welding-joints-welds/
These joints are intended to be filled up with filler.
I don't have  a lack of flow, the stuff is like milk, so for example I can not  do a double V because the first one would run away when heating the second side.

The only Al brazed joints I can do are when the parts are nearly horizontal.

I know tig would be better than oxy acet for this Al, but for hobby use, I use the torch mainly for brazing steel eg  4130 bike frame sections with nickel bronze filler.
I used oxy acet for 40 years and no point changing now.

[coppercone2]

those joints are not good for brazing. You don't get strength doing it that way. Overlap not filler. Those kinds of joints are good for fill welding. Every joint suitable for brazing needs a high surface area of contact between flats. Double V is precisely the opposite of what you want in a braze job. For filling holes for repairs OK, but its not structural. And butt joint is acceptable only on large surface areas. Really you want to imagine you were going to rivet the parts together but use braze instead of rivets. So you fold metal over into to L shapes to make a box, not mate it with two thin flats. I.e. those corner joints (in MY picture) are going to be weak, you really should get a brake and bend it like you were gonna rivet. I.E. in the picture I linked, the dimensions need to scale. If you take two 1/8 inch metal plates and attempt to braze them perpendicular its gonna suck/be brittle. If you bend a 1/4 inch flat on one of the plates and braze it to a flat on the other plate, it will be strong. Same for soldering.

I.e. braze a angle piece to a plate and it will be OK. Butt joint two thin plates and its gonna be crap. Scarf joint is what you want for joining two thin flat bars *but its hard to grind it right*. Think of it like wood glue.




like this

[CatalinaWOW]

For those of us who don't weld or braze professionally (and thereby get the training and practice to achieve good joints consistently), the advice should be the same as is given for electrical wired joints.  There you are advised to make a mechanically sound joint, and then solder it.

Several of my reference books give the same advice for thin metal joints.  No butt joints or tee joints.  Lapped or rolled joints followed by solder, braze or welding if the application demands it.

[beanflying]

Straight butt joints for Aluminium brazing are a nono as there is no braze/weld penetration so you partially overcome that with v grooving. Even with Oxy/Acetylene and Bronze brazing into castings or thicker metals this is how it is done.

You cannot use a 'scarf' joint with Aluminium rods as they will not flow into a narrow close fitting slit it is nothing like wood glue you need to bevel the base metal on each side to increase your contact surface area.

As you can not groove thin sheet for a simple butt joint, brazing or soldering no method is any stronger than the contact area of the braze pool or the thickness of that pool. So this is not a place for this process if you need any sort of strength across that point. 99% Chances are any resultant joint will fracture the Aluminium Braze material if you breath on it to hard.

The Slip or grooved seam option below is what was taught for soft soldering sheetmetal way back in the day and is still valid for watertight roofing now as an example (silicone SUCKS). Providing you don't get cracking of the sheet at the sharp bends (use the correct grade will help) this will get you a really solid joint in Aluminium although cosmetically less desirable joint.

For Aluminium in particular Good technique and post braze normalization and PRACTICE is what is needed as much as Brand X or Y filler and 'magic' fluxes from obscure company B. Snake oil and misinformation is the one thing that is persistent in Aluminium Brazing consumables.

[coppercone2]

the Al solder should flow in a scarf joint, thats why I am impressed with it, compared to "mesh the cake in the joint Al braze"

and youtube video tests indicates that there are at least a few bad braze manufacturers out there that have specially poor wetting/flow. (not related to price 100%). If you really think your going nuts because the youtube guy seems like a absolute magician after 10 attempts try changing manufacturers.

I also notice the solder is actually flexible, the braze rods basically snap like dry pasta noodles

[beanflying]

A scarf joint is a zero clearance joint. Leaving a 'gap' between the two faces is not a scarf joint that is a badly prepared one. The correct preparation process is a V groove even with MIG and TIG that is how it is done and it is done for a reason.

There are instances where a 'scarf' joint can be done in metal like with silver soldering and more or less tinning one or both sides then remelting that meterial with flux but it won't work with Aluminium brazing.

How about instead of claiming 'magic' post a link to it?

[coppercone2]

I don't see the scarf joint in the picture looking substantially different then a lap joint, maybe you mean a more advanced scarf joint that was not in the picture

[beanflying]

There was no 'scarf' joint in the picture on my post a few back as it is NOT a joint used in sheetmetal.

A more appropriate joint prep for brazing or welding is done like this in the case of bar or tube. The longer surface area (length of joint) does make it stronger than the 90 degree butted version with the same prep but sitting the filler in the case of a brazed joint on the surface will get you a rubbish joint. Straight up 90 degree butt joints are done in the majority as a compromise in simplicity of prep and welding.

[mag_therm]

Comments about my interpretaion of posts by Copper and Bean.

A single bevel butt is a mainstream structural joint probably referenced in all the structural welding standards -reliable because it is done from only one side, and because there is less step in the cross section.
A lap joint has about 2 * step in area at the weld which becomes a stress concentration in bending and tension.

The term " Brazing" defined by AWS is fuzzy and wide, extending to the wicked/sweated plumbing joints Copper has done, to a joint made with filler rod.
"The Oxy Acetylene Handbook" by Linde differentiates  "brazing" as a thin film of filler drawn into a joint.
Then "bronze welding" as having  joint design the same as if it would be fusion welded, but the parent is not melted. After "tinning" (now called alloying), the joint is proceeds with filler rod and "the results are fully compatible to those obtained through fusion welding"
Chapter 19 covers "Principles of Bronze Welding"

I would classify the Harris solder wire 1 ounce and flux as linked by CopperCone, as "brazing" and the Alumiweld which comes in 3mm rods , half pound tubes,as being more like "bronze welding".
The big difference is that Alumiweld has almost no plastic or "mush" zone between solid and liquid.
And  gas torch, due to the high thermal conductivity of Aluminum, does not allow a close control of the melt that TIG can do.
Therefore gas torch joints can not be made close together. So many of the fusion weld joints shown in the posts become difficult, unreliable and impossible, as I have experienced and I think also 2 others on this thread have experienced.

[beanflying]

Both single and double bevels are used in welding including a double v groove (grooved both sides) to reduce buckling during welding plate as an example. Single Bevel is just one option and not a solution for all cases, used in the wrong place and a single bevel causes material to twist and bend. We were not however talking about 'welding' in this case as 'brazing' is a different animal having effectively zero penetration of the base metal.

The point to v-grooving a brazed joint is to increase the weld metal thickness and hence get more strength in that joint, the V will also get increased surface area for the braze to bond to. There is no mixing or penetration of the filler into the parent metal so this is a braze and there is no ambiguity here at all regardless of the Snake Oil consumable used.

[CatalinaWOW]

Something you might consider would be to make (or buy) an expander to allow you allow you to slip the end of your square tubes together by a few millimeters.  I don't know the temper of the 6013 tube you are buying, but the numbers say that it will allow that much expansion for most if not all tempers.  Doing this would increase the braze surface and also greatly simplify alignment during assembly.  With a little thought you can see how to accomplish this overlap when making corners also.  I don't see much chance of bending this tube into reasonable size arcs though.

Doing such joints would make epoxies much more practical.  I know you had a bad experience, but JB Weld is certainly not the apex of epoxy performance and lapping joints dramatically reduces the strength requirements.  Also note that there are mechanisms for brazed joints to pass a pressure test and then fail under use.  Flux inclusions being the most obvious example.

None of this has to preclude you buying the welder you want.  The welder will find its uses however you make your heat exchangers.

[mag_therm]

I did a trial Al chassis braze test piece when  oxy acet torch was used today.
https://app.box.com/s/1bjr3mnt9x80n8bdzmfpfnbm6bywkhns
https://app.box.com/s/rt91tmfrxzpmc4da87ypfxlws09qc93p

This is 1/8, 3.2 mm 6061 with Forney 46111 filler rod.
Photos are as welded before smoothing.
The joint is a single bevel butt brazed from one side.
The joint was half filled then repeatedly dragged under the melt, with a sharpened tool steel bit to scarify the parent metal.
Then filled up to slightly high.
I have experience now that this method  makes a stronger neater joint.
If the filler appears gray-white when cooling, not shiny,, I think it is indication that  at least some of the oxide floated out.

The filler metal has gone through to back as shown, but any attempt to turn the piece over and reheat the back side will just result in the filler running out.

Happy welding day, waiting for my new scope to arrive.

[coppercone2]

keep graphs like this in mind when brazing


thats strength vs distance between surfaces.
.
https://blog.lucasmilhaupt.com/en-us/about/blog/joint-clearance-and-joint-strength

that has values for aluminum braze I think. its a precise process

[beanflying]

keep graphs like this in mind when brazing


thats strength vs distance between surfaces.
.
https://blog.lucasmilhaupt.com/en-us/about/blog/joint-clearance-and-joint-strength

that has values for aluminum braze I think. its a precise process

That chart is clearly labelled as for Stainless Steel and has zero to do with Aluminium. The 'EzyFlo' mentioned there is a Silver Solder. There is two passing mentions of Aluminium on the entire page and again nothing specific as far as prep, method or joints. At least read the content you find when searching Google 

[coppercone2]

at least read the table on the second link where I said it had constants for aluminum

real eager to be a pain in the ass aint ya

[beanflying]

Hardly you posted off topic garbage!

Either because YOU DIDN'T READ the googled page you posted a chart that has NOTHING at all to do with Aluminium Brazing. OR it is completely an attempt at deliberately misleading or you or just you went looking for data to prove your lack of idea on what joints are actually possible with Aluminium filler. Silver Solder behaves very very differently to Aluminum.

'Ideal' thickness mentioned in that second link is a hypothetical theoretical fantasy from someone invested in selling a product. Aluminium filler will not flow into a gap of that width without additional preparation it will pool on top of that width gap and likely cause a porous result.

Lastly apart from mouthing off about Aluminium Brazing in every thread it is discussed in have you actually done much of it?

[coppercone2]

at least I got a number you got nothing. there is no mention of any kind of measurement. whats your deal. you can see from that number that the ball park strenght comes from something small, not as small as silver but its still small. They ain't write quarter inch there did they, they expect 2-3 pieces of printer paper developing the strength of the joint.

point is its a damn small number that makes the strength in the case of all brazing. you can surmise if its more then a few pieces of paper its bad. thats what brazing and soldering is about, a real small amount of material being really strong and making light weight high strength very neat and compact joints.

maybe this is a discussion about anchoring aluminum with brute force or something. there are alot of ways to use it but if you want that legendary performance thats the number you got to work with. you might actually be able to do some structural mathematics if you meet those joint criterion

[beanflying]

What I have is multiple decades of Gas welding on a range of gear and materials for a start.

The point is is that that 'damn small number' is a FRAUD and cannot be achieved. It simply will not flow into a narrow crack or close fitting surface. If you had done any level of Aluminium with Gas or even Brazing in general (not silver soldering) then you would have a clue about joint prep!

Lastly apart from mouthing off about Aluminium Brazing in every thread it is discussed in have you actually done much of it?

So?

[coppercone2]

thats because most of the alloys they sell are optimized for radiator crater fillers. I worked with testing large surface area stuff that was brazed gas tight aluminum around complex geometry

youtube aluminum braze-offs show that there is plenty of alloys that have a semblance of flow to them, but thats why I recommend the solder, it actually flows, and you can get the strenght from it, rather then falsified crater goo, without having to consult industrial manufacturing experts

[coppercone2]

I think they optimize alot of them as not to fuck up a car radiator with a clog.