# Trying to build an ultrasonic drill

1. Mar 6, 2015

### K_Mitchell

I'm attempting to build an ultrasonic drill to make holes in silicon wafers, since we don't have the budget to buy one in our lab. From what (I think) I know, ultrasonic drills can either work by the "jackhammer" motion of a free floating mass which I don't know how to make, or by the transfer of the ultrasonic frequency to an abrasive slurry. I want to make the latter type.

From what I gathered online, I think I'll need the following components:

1) Piezoelectric transducer. I found a 70W transducer that works at 28 KHz for about $40. 2) Horn. This transfers the vibrations from the transducer to the drill bit. I'm planning to have a "step horn" milled for me at our machine shop out of aluminum. This type amplifies the vibrations. Its length is critical to reduce the impedence, and should be roughly half a wavelength long as a general rule of thumb. 3) Drill bit. For the 1 mm holes I'm planning to drill, the best "bit" appears to be a hollow metal cylinder that I can solder to the small end of the horn. 4) Ultrasonic driver. My electronics knowledge is very limited, so I'm planning to just buy one off Ebay for$30. I found a 60W, 28 KHz driver that works for 110V.

What I'm worried about most is matching the resonant frequencies. I'm trying to follow this instructional the best I can: http://www.imajeenyus.com/electronics/20110514_power_ultrasonic_driver/index.shtml. Apparently, transducers that are rated at 28 KHz might be off by like 0.5 KHz, and this small difference can have a huge effect on the impedance as you can see from the plots on this website. The author apparently has a way to measure the impedance and tune the frequency so that the impedance is minimized. If I buy the 28 KHz driver (the author built their own), what's the cheapest thing I can buy that will let me tune this up and down, and measure the impedance?

Also, when I look up the wavelength of a 28 KHz wave (http://www.csgnetwork.com/freqwavelengthcalc.html), I see numbers like 5000 meters for half a wavelength. However, the author's horn is only 95 mm long. Clearly, there's something about waves I don't understand and I would appreciate any answers for this discrepancy.

2. Mar 6, 2015

### K_Mitchell

Do you guys think this topic is better off in the electrical engineering forums? It's a combination of mechanical, electrical, and physics so I wasn't quite sure where to post it.

3. Mar 6, 2015

### 256bits

Try calculating the wavelength for sound, rather than light.

4. Mar 6, 2015

### K_Mitchell

Did that and found the number was still an order of magnitude off. Then I realized this value was through air rather than aluminum. When the speed of sound through aluminum was used, I got the correct value. Thanks! I also found an excellent paper on horn engineering with easy to perform calculations.

Still need help with finding what I can use for fine-tuning the frequency and measuring the impedance.