Applied Science - 2020-09-09
Technical details and how to build an ultrasonic soldering iron. This technique can bond difficult-to-solder metals such as titanium as well as glass and ceramics. Cerasolzer technical info: http://cerasolzer.com/cerasolzer/basic_info_gb.html S-bond technical info: https://www.s-bond.com/solutions-and-service/ultrasonic-soldering/ultrasonic-solder-materials/ Overview of active soldering process: https://sci-hub.tw/10.5772/intechopen.82382 Another good overview: https://www.intechopen.com/books/recent-progress-in-soldering-materials/recent-advances-in-solderability-of-ceramic-and-metallic-materials-with-application-of-active-solder Discussion of Sn - La solders: https://www.hindawi.com/journals/amse/2015/269167/ Sn - Ti phase diagram: https://sci-hub.tw/https://doi.org/10.1007/s11669-010-9663-2 This patent has the key list of ingredients (not just broad ranges) listed at paragraph 45: https://patents.google.com/patent/US20160204303A1/en https://www.patreon.com/AppliedScience
I freaking love this channel. The kinds of stuff that has been done on here is amazing and the fact it's being done in a garage is encouraging.
If I remember correctly, while researching how to desolder surface mount components, I found an article that stated that soldering iron tips will degrade over time in continuous use, and it's important to replace the tip every few months. Perhaps if you had converted a much better soldering iron, you may have been able to retain the tip being replaceable, thought I'm certain that would have made the conversion 10 times more difficult.
I think it could be interesting to test this procedure on bone and stones. They mostly consist of metal cations from the first two groups of the periodic table and the anions are different to those that you have already tested.
Your question is a little worrying, honestly 😂
@Alan Matthews just wiccan things
Very cool. In a rather large tangent to what you tried with the carbon, Robert Murray-Smith did recently do a bit of experimenting with aluminium brazing rods, and managed to get a decent bond between some Grafoil and aluminium. Obviously not a chemical bond, but it may be a relatively straightforward way to mechanically do that.
Super nice! It would be really, REALLY interesting to see this applied to high-vacuum glass-to-metal seals! Would make viewports and electrical feeds a breeze!
You know you're a real engineer when you fire up the homemade vacuum induction furnace to avoid sending an email
It's not so much to avoid the email, it's just usually expensive when a company says contact for pricing so he made his own to save money and because he can
@jakob findlay an email is already too great of a cost
@jakob findlay on the contrary, if you have even the most basic appearance of a business or a promising application, many places will gladly send you engineering samples and have an engineer reach out to you to help you design your process.
haha most dont know that this crappy behavior was standart in the >2000's , usually if a company didn't list price you could not afford it xD
also totally unpractical. Its a method from the 1970's where it was normal to order over physical mail.
@gasfiltered This is very true and sometimes you can be pretty open about the fact that you're only looking for a few pieces for proof of concept. Years ago I, and a few other people, got a handful of JFETs as samples for projects which seemed to arouse enough interest in the DIY community for smaller distributors to begin stocking.
I find this technique extremely interesting. Two things: 1. would love to see more on your process of making the solder, and 2. would like to see tests bonding aluminum foils to copper wires, as that could be handy in thin circuit prototypes. Great video! Thanks!
A while back I was working on a battery charger for a golf cart oddly enough the transformer was made with aluminum wire. I found a special solder that was specified in the manual called Kapalloy9 used with Kapp golden flux work great soldering the copper connectors to the aluminum wire.
Outstanding video with zero fluff, as usual. You brought attention to many people, myself included, of a technology of which there was near-zero awareness!
I've used this stuff. Got it from S-bond (you described their 220 alloy), which wasn't much of a hassle, though they sell it for $5/g. I didn't need much, so that part was fine. It's actually possible to brush it on with a hotplate, so I cheaped out and did that, although you need to clean up the surface afterward--I used a fly cutter on a milling machine.
I was bonding glass to ceramic, so the thing that got me was the high yield/creep strength. One of the easy ways to cope with thermal expansion as you come down from the melting point of the solder is for the solder to creep a bit, but this stuff has a fine matrix of SnTi intermetallics which act as rigid reinforcement and prevent it from creeping easily.
I had a strong alumina piece bonded to a glass piece with CTE 7-8. Everyone will tell you the CTE of alumina is 7-8 also, but that's averaged over either 0-500 of 0-1000 C. Over 20-100 C, the range relevant for this process, it's more like 4-5. So as the assembly cooled, the ceramic pulled the glass apart without debonding the solder; it's got some strength. We learned that we should have stuck with B33 glass (CTE 3.3 ppm/C), which we would have done if not for this red herring about the CTE of alumina.
Glass is stronger in compression, so I think the active solder might have worked with a B33 glass part, but we went for a softer solder, and solved some of its problems for our application instead.
Ben, I genuinely need to thank you so much for this video. I had never seen this technique before, and it's so simple. I wish I had seen this 4 years ago when I was working on a research project that required soldering to doped silicon -- we ended up using this weird proprietary compound called nickelex instead, and it was a disaster and our contact quality sucked, and this was despite the fact that said compound was 1% HF acid. This is so much simpler and safer...
I'd be really really curious to see if you get good ohmic contacts across silicon oxide -- if you could do a quick IV sweep to check I'd be /extremely/ grateful. I'm also really curious if that silicon wafer holds up to a post-solder anneal, and if it improves the contact quality at all. Again, thank you so much for this!
I believe that ultrasonic welding is the norm for silicon?
@Jared Maddox I've never heard of ultrasonic soldering being leveraged within a semiconductor engineering context prior to this -- usually, within that context, affixing anything to silicon via methods other than wirebonding to pads is out of the question. The goal of this project was solid state sensor fabrication outside of a clean room, which restricted the options pretty significantly. I'm sure this was an option that was known at the time, but neither my teammates nor the professor we were working under knew about this possibility ahead of time or stumbled upon it during our research. Dunno if that means we just missed the right sources, if it wasn't as widely adopted at the time, or if it's a research siloing thing.
@hedgeberg : Not ultrasonic soldering, ultrasonic welding. I understand it to be the norm for semiconductors, using gold wires that are ohmicly bonded to contact points on the semiconductor die via ultrasonic actuators of some sort. I have neither training nor experience in this field, so I might be misunderstanding something about the whole subject (for instance, maybe the bonding is actually to pre-deposited solder instead of directly to silicon?), but I've run across mentions of it intermittently for years.
That's amazing! Ben, you come with the most fascinating ideas to check into and show us how you made work, I love it.
I worked for the company which developed much of this ultrasonic joining technology, not requiring solder, which was then taken over by EWI.
If you'd like to pursue this further and have specific technical questions feel free to reach out.
@sudo scapy same question
@Donald Johnson thanks
you could test the electrical properties of different "non-typical" materials under different enviorments with an oscilloscope (ex: how light or heat affect the electrical flow through glass). Nice video btw :)
Fantastic video as always! Could you perhaps look at developing a simpler alloy which could be ultrasonically soldered under a shielding gas?
You're the most amazing and skillful maker on YouTube, I'm just blown away by the things you do and your humble way to present. Great stuff, keep it up!
Wow dude you are my new favorite channel. I can't believe I didn't know about this channel until about a week ago. Amazing content. I am /trying/ to build a lot of the stuff you have built already. Fantastic. I was wondering how to build ultrasonic welders, and now, I know. THANKS!
Electrical passthroughs for high vacuum or cryogenic applications tend to be very expensive. You could probably use this to connect an insulator into piping, then a solid metal wire inside that for sensing. It looks like regular glass and titanium have similar coefficients of thermal expansion (at least at room temperature), so you might be able to make a decent passthrough that stays leaktight down to cryogenic temperatures.
I'm starting to lose track of how many times I've seen something on this channel that I would have thought was flat out unrealistic for anyone to be able to pull off in a home lab/shop. genuinely amazed
The secret ingredient is MONEY
@Justin Koenig Google X probably pays him at least 300k, so yeah, money helps.
@Taygetea As far as I'm aware, he basically designed the HTC Vive hardware, so money and experience is a hell of a combination.
@Kyle Emmerich Well at the moment he's one of the most important researchers in that medical offshoot of X, so that probably pays well.
Well, it helps to have an electron beam microscope in the shop. Oh yeah, built that too!
To add something more relevant to the topic, you might want to look at the reactive brazes that are used for bonding diamonds to metal substrates - the one I came across used a mixture of copper, titanium hydride, and maybe some tin/other minor ingredients. I fired it in an induction furnace, bonding a grit onto a steel plate. The really interesting feature of this system is that the titanium hydride first decomposes and the hydrogen expels air from the powder, before the metals start to melt. Then the Ti and Cu form a braze alloy, and the Ti also reacts with the surface of the diamond as a layer of TiC, so there is a chemical bond to the diamond and a graded interface as well as a mechanical bond. I was surprised to find out later that there are some really high performance brazing systems that can produce remarkably tough and strong bonds for specialist metal joining, no welding required. These are really useful systems. The ultrasonic part of this one is a new aspect for me - the overlap with ultrasonic welding of plastics is just a coincidence.... cheers, Miles
I love watching your experiments and deep dives. Could you do a video on tips for researching papers? Most stuff I find is blocked behind a paywall.
You come up with more really cool applied physics than I see anywhere else. Every video you upload is a must watch.
I keep coming back to this one while waiting for the next bit of content from Ben. Great stuff and always eager for more!
For the graphite, you have to intercalate a relatively smooth area. Otherwise, the flakes will separate.
Chemical supply house: "wont sell to hobbists"
Hobbyists: "Hold my vacuum induction furnace"
Lassi Kinnunen Bingo 😎
Nah he lives in California so he'd probably go for a Nevada corp.
"Hobbyist" is stretching it a bit with Ben though :)
@Tim Williams more like mad scientist. but i love it anyways.
"Hold my Scanning Electron Microscope"
This would probably make things like copper-glass tube bonds sooo much easier. I know it probably wouldn't be as strong as a traditional bond but I wonder how well it would handle vacuum pressures.
I would be very interested in knowing if you can active solder onto crystals, anything really but quarts is an easily available material. Not only could that lead into interesting tests of piezoelectrics and resonant frequencies but could also go further and maybe be used for testing ionization energies of their valence electrons. Not gonna lie, the prospect of being able to remake the original transistors without a massive headache is the first thing my mind went to.
Tees active soldering alloys are very interesting! I like the idea of playing around myself but I don't have the equipment to do small batches! I imagine a tin lead silver gallium alloy would stick quite well
Absolutely love your channel and content. Just as a random suggestion, I work in a laboratory which does cancer research, and I would love to see any ideas you may have on future videos having to do with genetics/genetic engineering/bio-lab subject matter. Thank you for all you do! :)
For a graphite bond, what if you threaded a small hole to provide mechanical strength to the bond? Maybe run a left and right hand tap through the same hole (carefully) to produce a crosshatch pattern and prevent the solder plug from backing out the thread? Or maybe that's overthinking it? You could probably just drill a straight hole and use a needle to scratch up the inside of the bore.
It's not going to hold if you yank on it very hard, but it might be strong enough to withstand modest vibration if not subjected to any other outside forces.
@ 3:30 ish. it might be helpful also to mention that this principle applies to iron/steel as well. A black iron oxide finish is often used to keep forged pieces from rusting, and a blued (oxide) finish is often used to protect firearms and other mechanical parts where a thicker oxide coating is not desired. These protective oxide coatings just typically don't form on their own like they do in copper.
rust bluing does not, itself, provide much rust protection, rust bluing is a porous coating meant to hold oil that does the protecting.
SuperAWaC clockmakers use the bluing to protect the metal. Its not bluing like on guns with a bluing liquid and oil, they literally burn the steel very evenly and it literally turns blue (or straw) and if you do it like that it is not porous and you dont need oil at least as far as i understand.
@Melody I am a machinist. Any form of bluing (converting the iron to magnetite) is porous. They all require oil to get maximum mileage out of them. What you are referring to is just using heat to temper the metal, the oxide layer formed doing that is extremely thin (the thickness of the layer determines the color due to the way it interacts with light) and largely cosmetic.
@Melody Ah, the difference between "Clickspring bluing" and "This old Tony bluing". But I think clickspring do it mainly because the metal needs to be tempered after hardening, the color and protection is just a by-product.
Good commentary
Very cool indeed Ben. I.m going to build this project. I looked at buying an ultrasonic soldering iron and they are way too expensive at around $7000.00 US. Looking forward to the next video.
This is an Amazing process. It was used to bond the SR71 windows (Quartz) to its structure (Titanium Alloy) and it worked fantastically well, keeping it air tight at high temperatures (>350°C) at supersonic flows (> Mach 3.2)
Neat. I've been using a Sunbonder and Cerasolzer to make electrical contacts on tin oxide coatings for the past few years. Now I'm really impressed that the commercial irons don't wear much even after thousands of hours of use. Maybe they use some really hard stainless for the tip?
Just a note that you can solder some difficult metals with conventional solder by using some supplements. For example I have bottle of ZnCl solution, that is used to solder ferric alloys with quite success, it can also be used to solder Ni / Cr alloys. Another mixture is dilute HF + EtOH for soldering aluminium. I haven't studied deeply how these work, but it may give some insight to this problematic as well.
As to what to do next, I personally would go wild with alloys trying to substitute expensive or rare elements. Al instead of Ti comes to mind, Zinc would be interesting. Unfortunately I am not really excited by this technology, otherwise I would do my own experimentation
i don't care what you're planning to do, just do more of it, every single video on this channel is awesome!
As always, an absolutely awesome video. I need to make myself an ultrasonic soldering iron now. Looking forward to the video making the active solder alloy
Thanks, man!
Awesome diy! I'm curious about the connection resistance, have you done any resistance measuring? Thanks
Unbelievable! You're a magician, on every single video of yours that I've watched. And I've watched most of them. Thanks Ben!
Very cool! This really simplifies the soldering process for a lot of metals. No more finding the special flux or technique for each metal!
One thing you have to be very careful with is Chrome and Zinc around Ti, it damages the titanium quickly. When I was in the RNZAF, we had some aircraft parts that were Titanium, and we were not allowed to use regular Chrome plated tools on them, as they left marks like pencil marks, but in a few weks the Titanium would exfoliate in those areas and even fall apart.
Regarding the Graphite, you may have soldered to it, but it has layers very thin layers, so they tear off easily. If you look at a motor 'brush' they are compressed around the copper wire to make connection. Cool fun you are having there.
Nicrome soLders just fine with normal Pb Sn soLders with phosphoric acid as the flux. This is also an excellent flux for stainless steel and copper base alloys. This is why coca cola used to work as a flux - once it was allegedly 18% phosphoric acid.. Martin
PS. Absolutely love the HP analysis thingie
This is awesome! I learned a lot from this video, as always. I had no idea that flux was used to remove the oxide layer. Thanks ben!
Very interesting video, thank you.
Have you experimented with the use of mildly reducing (and non-toxic, if there are any) or inert shielding gases to keep the oxygen out of the melt?
With respect to the attempts to solder to polymers, you might have more success with polymers that can form covalent bonds with e.g. dyestuffs or other chemicals, such as textile-fibre-grade nylon 6 or nylon 6,6, or the cured form of the two-part resin that is used to make fibreglass (assuming that some reactive chemical groups remain after curing), etc. A textile chemistry textbook would help identify such reactive polymers. Overheating during the soldering could damage the polymer.
Wow! This is amazing, I never knew this kind of soldering/joining existed. I'm absolutely fascinated.
The threaded rod that goes into the horn is the same thread type as the ones in normal bicycle axle shafts. Any regular shimano will thread in. Really nice cause you then have access to cheap threaded tube for those (for the quick release axles).
Very interesting. How does the joint resistance compare to traditional soldering? I think this would be great for connecting rechargeable batteries but I wonder if it would add an additional resistance component to a circuit.
I think it would be dope for making art with obsidian, I just like the idea of the contrast those two materials would have
Have you tried using this for glass-to-metal seals for vacuum stuff? Would be REALLY handy!
I wonder if you could solder a canning jar sealing ring onto a glass jar through the ring? Would you have to have contact with the solder and glass or the transducer is powerful enough to penetrate through. Don't know what purpose that would serve but neat to find out.
I'm a beginner (barely in anything) but I understood everything. It's very clear and well illustrated, thank you very much!
NightHawkInLight - 2020-09-09
Really interesting stuff. I'm surprised your ultrasonic iron worked so easily just bolting stuff to the front of the transducer. When I've played with them they seemed really finicky about tuning and I think I burnt out two of those ebay driver boards. I guess being so overpowered means they don't have to be performing optimally to get the job done.
mitchell - 2020-09-09
If you two teamed up and brought in Tech Ingredients, you could rule youtube. Please consider it
Applied Science - 2020-09-09
Thanks! My driver board is finnicky too. I think it helps that there isn't much mass connected to it, and I'm not pushing hard, so it doesn't have much mechanical load. If I build another rev of the iron, I'd use a smaller driver, and a more controllable circuit. I like your new channel logo!
chain 3519 - 2020-09-09
Unrelated, but kind of crazy, I've been watching both of you since I was in middle school. I'm a senior in aerospace engineering in college now. Both of you have had a positive impact on my life
Overland One - 2020-09-10
I just made a similar post above before reading yours here. My company owned two ultrasonic impact grinders and two ultrasonic rotary drilling/milling machines and all of the diamond tooling had to be tuned to a nodal point for the given mass and length of the tool or it would put too much stress on the transducer and the tools would not perform. We machined ceramics and other very hard materials using all diamond tooling. You did not want to burn out a transducer as, even back in the 80's, they cost about $30,000 each. I enjoy watching your videos and always learn something from them.
Benjamin Esposti - 2020-09-20
We've got some ultrasonic press machines at work, used for sealing and joining stuff, primarily plastic.
One time, one machine had a bolt fracture on the booster, and it made such a horrible screech that could be heard outside the building, with the doors closed, over 50 feet away!
Ear protection is a must!
Granted, these machines are about 900W each.
They're indeed picky about the dimensions of the tooling, it has to be tuned, otherwise you risk damaging the horn (transducer) or having poor efficiency.