Ultrasonic Cleaner to... Ultrasonic KNIFE?!

A little something I've always wanted to try... can I turn an inexpensive ultrasonic cleaner into an ultrasonic knife?
----------
----------
Music: "Goat" - Wayne Jones

[keithwilliams]

I've dealt with ultrasonics and building systems for custom NDT applications. You do a great job talking through a lot of major items that are critical to make a system function. A couple of quick suggestions: Try using JB Weld as your epoxy, the metal particles in it tend to be a great transport medium for the ultrasound even if you don't completely get a tight metal on metal interface. Next add some lube (yes KY is pretty decent to try at first) between the transducer and your 'horn', this will act as a couplant and much more of the ultrasonic energy will be coupled into the system. You can then try other oils/greases until you find something that works well for you. The only issue with KY is that it is water based and it will 1) dry out fairly quickly and 2) can obviously impact 'rusting' or oxidation. Finally, you have a o-scope and seem fairly adept with electronics, if you have a function generator, you may be able to sweep the transducer and find its true resonate frequency in the system and then adjust the drive frequency accordingly. Feel free to PM me, if you want to go into any of this in more detail.

[PracticalEngineeringChannel]

Thanks for picking a photo showing my good side ;) Very awesome project.

[josephj]

Hi,
I have some info on ultrasonics for you. I haven't watched you're videos very long so I don't know your background. I'm guessing mechanical. My background is electrical so I'll tell you what I know.

1. These horns (aka sonotrodes) are analogous to power transformers. That helped me with the concept.
2. You may achieve better performance by silver-soldering the cutter into the horn. The adhesive may be a bit of a damper.
3. The screw in the middle functions as not only hardware to hold the assemble together, but also as a mechanical pre-load. It applies a necessary compression on the transducer. Relieving that pressure and then shorting the terminals with your shitdiggers will cause it to build and release electrical potential respectively. I believe they do this so that you bias the piezoelectric disc at full negative potential. That would mean you only need to apply a positive voltage to have the transducer reach maximum displacement. It polarizes the piezoelectric so that you don't have to drive it with "true" or "full" AC (including a negative half cycle), but instead with a chopped up dc source. Don't quote me on that. (I was told the terminal in between the two discs in positive.)
3. Piezoelectric transducers have extremely high quality factors (aka Q factor), which means they retain energy from one cycle to the next very efficiently when in resonance. This also means that it has an extremely small range of operating frequencies. i.e. you may have to stay within 0.1% or less of resonant frequency.
4. When operating out of resonance the transducer will become capacitive or inductive depending on which way you drift. This results in a low power factor and higher transducer impedance. That means no current, thus no power.
5. The 1:9 amplification you got in simulation lines up with the math I've seen on it. Because the pressure wave has to maintain the same energy across the cross-sectional area of the horn as it propagates through the horn, if you REDUCE the area by a factor of 9, you INCREASE the cross-sectional power density by 9 so that energy of the wave is preserved. It looks like you've done a diameter step ration of about 3:1, and since cross-sectional area is related to the square of the diameter the simulation hit it right on. This is only true for the resonant mode, as your computer generated figure shows. A transducer displacement of 10 microns can be amplified up to 90 microns with this horn.
6. Piezoelectrics have harmonics, but with increasingly higher input impedance. You could drive it at a harmonic frequency. The power would be lower, but not nearly as low as an un-tuned one.
7. When using a half-wave sonotrode, the area right in the middle (length-wise) is a node. This is where the standing wave is at zero. At this point the horn experiences little-to-no displacement. That makes it a great place to mount it to another assembly at (with dampening) if need be.
8. This machine is a big ole impedance nightmare. You have electrical impedance on the left and mechanical impedance on the right with a stack of piezoelectric discs in the middle to act as translator. The mechanical impedance changes with horn structure, cutting tool geometry, and pressure applied to the tip. It's a ball-busting balancing act.


I made a untrasonic drill/corer for my capstone project. I have my report on it if you'd like it. I can direct you to the most useful parts. Just contact me for that or if you have further questions.

+AvE You are also welcome to this info.

[nathanskinner]

There are not many forms of media in this world that I don't mind watching over and over again. Your videos tony, excellent. Your filming skills are amazing, your character charming, and you knowledge vast. Thank you for gifting us with your videos.

[VSO_Gun_Channel]

Hey man, those units were probably made of titanium or SS because Aluminum is no joy for ultrasonic applications, as you demonstrated with the foil. That horn will probably become brittle. For instance, In the silencer word, we dont use ultrasonics on can components made of Al because they microchannel, becoming brittle, when you re-pressure the thing on the end of the firearm you create an IED. Time will tell.

[moeriou]

Could you take an ultrasonic cutter and make it into a powerful ultrasonic cleaner?

[Nighthawkinlight]

I've been working on this same project for a year. Filmed a whole video and scrapped it because of mediocre results. Glad to see it work for someone

[loupitoufl]

Great project, I use to design ultrasonic probes for digital imagery so I'd like to share a few things :
1. Aluminum for the horn isn't a great material because the acoustic impedance is low (compare to titanium). Most acoustic load (the sum of the electrodes + mass pass the electrodes) wants to be light (in order to maximize the vibration in the c-axis of the piezo crystal, usually PZT) and stiff (as AvE would put it) to maximize the difference between said mass and air/water. This is the condition of resonance. So Aluminum not so good, ideally Titanium, Magnesium, Tungsten, ...
2. The resonance frequency is most likely off now with the new horn compared to the reservoir but there is a simple way to find out : attach the end of your horn to a plate and disperse salt/ sand on it and turn it on, see if you have constructing interference - or modes (Chladni plate)
3. You should be able to tune your system by adapting the signal generator on the board to your new horn/load. Any chance you could share a picture of the board ?
4. What software did you use for your sims ? Solidworks I presume ?

[BrobaFat]

I've been watching you for years but had somehow never seen this video, I'm actually an engineer at a company that makes ultrasonic cutting machines! We use titanium for the horns and run them at 20khz, but tuning them is a crap shoot and really more of an art than a science. Our blades are carbide braised onto a tool steel nut that threads into the horn. We grind the knife edges down to ~10 microns, scary sharp even without ultrasonics! Everything involving the ultrasonics is run DRY, no oil, epoxy etc. It's all solid metal to metal contact from the crystals down to the blade, very finely machined. We also use cutting discs, peeling knives as well as blades. It's really freaky touching the blade while the ultrasonics are on, it feels slick to the touch, like it's covered in oil, your fingers glide right off it.


We can also cut metal on our machines! They do a fantastic job cutting aluminum honeycomb core. Super clean edges with no tear out.

Our machines sell for anywhere from a quarter million to several million tho.

Also the ultrasonics themselves never exceed 120 watts, a well tuned system is extremely efficient, to the point that a badly tuned knife may actually use more or less power but suck at cutting

[TPLeatherworks]

I found your channel yesterday and I haven’t been getting ANY work done since. It’s becoming a problem. Please delete your content.

[TheDAMASCUSSMITH]

Hey there.
After I stopped laughing my azz off from the audacity of a layman to attempt this endeavor without the benefit of the amazing Branson A200A frequency analyzer, I am seriously impressed at your attempt.
I was an Acoustical Tooling Engineer at Branson Ultrasonics.
I've designed and built the exact tool you have tried.
Mine was used to cut snickers bars. We cut all kinds of things.
Your two major problems are, a, cutter connections need to be solid. such as silver solder. b, you need to remember that the horn system resonates in both directions. you dampened the system by putting it in the vise. all mounting needs to be attached to the antinodes only. Also a catenoidal horn style would work more efficiently.
When mounting a thin light weight tool to a horn, just use its weight in your calculations, not the length.

FYI, my new co-workers got a great laugh when I disassembled my first 20khz pizo stack. Way bigger stack, 20khz and 2000 watts. Way bigger shock!

[Equiluxe]

In 1971 I was an apprentice horologist (watch maker) it was my job to repair the cheap alarm clocks. There was an ultrasonic cleaner used to clean all the clock gears etc. On one occasion I put a load of clock parts in the machine and started to put another one back together of course I then went on to start stripping another one and the boss called lunch so I downed tools and left leaving the cleaner on, the shop got locked up and the boss decided to make a delivery or something on the way back, well when he got back apart from me waiting outside along with some customers there was this ultrasonic cleaner still buzzing away inside, when I took the clock parts out there was holes eaten right through the gears, they looked like lace in places strange thing was the teeth were still all intact and the clock went back together again and kept reasonable time. The cleaning fluid was Toluene, that remove all grease and oil, we used the dirty stuff to clean car parts etc.

[HandToolRescue]

Does an ultrasonic fork make it easier to stab things?

[OrlFeyne]

Hey Tony!

Came to your channel through Wintergatan. And now... I can't stop watching. Great combination of comedy/entertainment and knowledge infusion. I feel like I'm back watching Schoolhouse Rock videos as a kid. ... Well, with less cutting lubricant/coolant. ... A little less. Now to buy some merch. The addition of a couple Ts to my drawers should start a fun conversation with my wife about why she doesn't give me the ATM pin anymore...

[thepagan]

As a professional Mechanical Engineer and a keen DIY'er it is great to see similar minded people that enjoy tinkering around in their workshops. Great video, great job.

[Tommy_Mac]

"Not only will your parts come out as dirty as before, they'll be wet, too". Exactly what I was hoping for.

[arduinoversusevil]

Dammit This Old! I called dibs on this 'un!
Er wait, yeah you go ahead and get it all worked out. MUUUuuuhahahahahaha.

[DoRC]

I use my ultrasonic for one thing and that's cleaning carbs. I've lost a bunch of weight since I stopped eating them dirty.

[insaneduderone]

Hey for anybody that wants to know... Just about any 2 part epoxy like JB Weld, or any other can be easily be picked apart or taken off by putting the heat of a flame from a regular lighter for a few seconds to it. Let the flame touch the epoxy just long enough to get it hot, then 10 seconds to cool and pick it away with your finger nail or whatever else you want to use. Don't let something epoxyed scare you from taking it apart. Pass this tip around to everyone that likes to take things apart.

[LeagueOfGaming]

Your analogy for resonance with pushing a kid off a swing is actually perfect with translating to mechanics