Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How can I fine tune an ultrasonic frequency?

  1. Mar 7, 2015 #1
    I'm building an ultrasonic drill, and I've already collected or machined all of the components. Since I'm very inexperienced in electronics, I'm planning to simply buy an ultrasonic driver (power source) rather than build one myself. The one I'm interested in is here:

    http://www.ebay.com/itm/110V-100W-28KHz-ultrasonic-cleaner-power-driver-board-/120972831045

    Once completed, the ultrasonic drill with its 28 KHz piezoelectric transducer will NOT resonate at exactly 28 KHz. It may be off by a little, and the transducer itself is rated +/- 0.5 KHz. Since impedance can increase by two orders of magnitude if the drill is driven by just +/- 0.25 KHz away from the optimal, I really need to fine tune the frequency. Someone on another forum told me that I can measure impedance with a multimeter (again, I'm very inexperienced), and they also told me that if the power source is driven by an oscillator then it probably has something he called a "trim pot" that I can apparently tap into. Does the product above look like it has a trim pot? If so, what is the name of the device that will let me tune the frequency between say, 27-29 KHz?
     
  2. jcsd
  3. Mar 7, 2015 #2

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    These things are very cheap. The frequency accuracy need only be better than 1%, which shouldn't be too much of problem for a simple oscillator. You may be worrying too much, even though the pictures don't seem to show a trim pot or trim Capacitor.
    Are the eBay items much cheaper than what you can get from other sources? You could try contacting a manufacturer to get a spec for what you are considering buying. That would tell you the 'guaranteed' power output. It may not matter what power you are getting, as long as it is in the right ball park.
     
  4. Mar 7, 2015 #3

    Baluncore

    User Avatar
    Science Advisor

    You may not need to tune the frequency of the system. If you put an AC matching network between the driver and transducer you will be able to correct the impedance mismatch at the actual frequency of the driver. That network could be as simple as a capacitor in parallel with the transducer or a series inductor inline. By placing a low value resistor in series with the transducer it should be possible to use a dual beam oscilloscope to measure the voltage across the resistor, (which gives the current), while at the same time looking at the voltage across the driver. When those two signals are in phase it is tuned to zero reactance. There may be some advantage in matching the in-phase, resistive component, with a transformer, but I expect an L and/or C network will get it close enough to efficiently peak the power transfered.
     
  5. Mar 8, 2015 #4

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    A good idea, in principle, but not everyone has a 'scope or a drawer full of useful components. :smile: A frequency trimmer would be a real advantage in most situations. But, I guess the OP would still need to be able to measure output power some some way. Would a DMM work (even qualitatively) at 28kHz? A simple detector circuit would be needed, if not.
    But why would the output need to be maximised? A cleaning bath is not a critical piece of kit - I believe the actual dimensions of the bath can be at least as important as the power supplied to the head.
     
  6. Mar 8, 2015 #5

    Baluncore

    User Avatar
    Science Advisor

    If driver power output is maximised then the VA Reactive will be minimised. The impedance should be matched to reduce the circulating current stress on the driver output stage.

    While drilling, the impedance of the ultrasonic transducer will be changed by the mass and length of the attached drill rod.
     
  7. Mar 9, 2015 #6
    Thanks for all the input so far. Like Baluncore said, stress (on the horn) can be large, especially for a stepped design and lowering the impedance to my knowledge can help reduce that stress. I've been doing some more searching, and found some items that may make my life a bit easier. I've been in contact with this company from China, though our communication is limited by their English ability.

    Ultrasonic welding transducer: http://www.ultra-piezo.com/2010/0726/39.html
    Ultrasonic generator PCB: http://www.ultra-piezo.com/Products/ultrasonic-generator/ultrasonic-cleaning-generator/8.html
    Frequency meter PCB: http://www.ebay.com/itm/DIY-Kit-Sec...ncy-Meter-Inductance-Capacitance/271717023286

    The frequency meter is cheap, and I hope that I can just figure out a way to attach it to the PCB generator. This generator also comes with a trimmer Sophie.

    One problem though: The company said that their PCB generator is designed for ultrasonic cleaning instead of ultrasonic drilling. However, their "welding generators" that can also be used for drilling are much more expensive because they're boxed and look very polished. Can't a 28KHz 500W driver, for example, power a 28KHz 500W transducer regardless of its purpose?
     
    Last edited: Mar 9, 2015
  8. Mar 9, 2015 #7

    Baluncore

    User Avatar
    Science Advisor

    What material are you drilling? What is hole diameter and depth? What material is used for the drill rod? How do you cool the cutting face? Do you need to spin a “D” shaped drill rod or will the holes be so shallow that a solid round or tubular annular cutter can be used without coolant?

    Where will all that 500 watt = 500 joule/sec go? It may destroy your drill rod and/or smash the work-piece. Find out how much power your set-up needs to do the job by looking at the power rating of commercially available drills.
     
  9. Mar 9, 2015 #8
    I'm drilling through two 0.5 mm thick wafers of silicon and glass (1.0 mm total) that are bonded together using a 1000V anodic bonding tool that I built last year. I plan on making 0.8 mm or 1.0 mm diameter holes using hollow drill bits found here and using silver solder to hold them in place as described in the second link:

    http://www.hplapidary.com/p/1306/highland-park-ultrasonic-drill-tips-8mm-stainless-steel-quanity-10
    http://www.free-form.ch/tools/manual-sonic-en.pdf

    I was reading through an ultrasonic machining handbook, and the materials used have to be aluminum alloy, titanium alloy, or hardened steel (I know the specific type of each). I have a horn engineering handbook and calculations for the stresses and amplitudes as the horn gradually or suddenly tapers. A 2X gain in amplitude or less is generally not too harsh on the horn. The drill bit on the other hand will slowly degrade and will have to be replaced as it wears out.

    Compared to commercial ultrasonic drills, it seems I was a bit low earlier when I linked to a 100W driver. The cheapest ones operate at 200W, and mid range models start around 400W. One company only sells 1000W or higher. I was just giving 500W as an example, but I'd probably settle on something in the range of 200-300W. As for coolant/abrasive, I'm going to buy a commercial slurry made specifically for ultrasonic drilling.
     
  10. Mar 9, 2015 #9

    Baluncore

    User Avatar
    Science Advisor

    Yes, the remaining problems will be environmental and impedance matching the wide end of the horn to the transducer.

    You may need to take steps to keep the transducer clean in your application.

    Transducers for cleaning usually attach to the outside of a stainless steel container. They may be backed by a tuned mass that will appear like a solid block, (a short circuit at the operating frequency), to give them something to work against.
     
  11. Mar 9, 2015 #10
    Thanks for the help, that's what I was hoping to hear. On a funny side note, I just got off the phone with a company based in CA that sells ultrasonic drills. They had time to talk, so I pitched my idea to them and mentioned that I even had cheap access to a 900W driver. After laughing for a minute, he told me that at 900W, my silicon sample would be obliterated. He recommended 200-400W, more towards the lower end. I'll be keeping the transducer very clean, by the way, and it will be far away from the slurry.
     
  12. Mar 10, 2015 #11
    Last question for you guys. The transducer is listed as 220V. However, my power source is 110V because my outlet is the same and I don't have a variac. Considering that the transducer is just a chunk of metal and ceramic that vibrates at a certain frequency (28 KHz), then as long as my driver is putting out the same power and frequency everything should be fine, right?
     
  13. Mar 10, 2015 #12

    Baluncore

    User Avatar
    Science Advisor

    NO. The 50Hz or 60Hz mains supply voltage to the driver's power supply is one thing. The 28kHz driver signal level to the transducer is another.

    A variac would work on the AC mains supply of 50Hz or 60Hz. The driver will be designed for a particular AC mains voltage. You must use the correct AC mains voltage, which will now be either of the international standards of 115VAC, (110 to 120), or 230VAC, (220 to 240). You cannot normally use 110VAC for a 220VAC device or vice-versa. You would almost certainly need a transformer.

    The 28kHz AC signal amplitude cannot be transformed with a 60Hz transformer, It would need a special iron powder or ferrite core. If you halve a voltage you must double the current to get the same power. The electrical impedance is the ratio of voltage to current, so ( V/2 ) / ( 2*I ) reduces the impedance by a factor of 4, it is a square law.

    We need more information because there is confusion here. Exactly where and what are the frequencies and AC/DC voltages you are referring to?
     
  14. Mar 10, 2015 #13
    So here's a breakdown of the devices and their voltage/frequency/power:

    Power source (driver): Rated 110VAC, 28KHz, 300W
    Transducer: Rated 220VAC, 28KHz, 300W

    I was planning to plug the 110V PCB driver into a 110V outlet, which is compatible. The 110V driver at 28 KHz and 300W then powers the transducer, which is rated at 220V, 28KHz, and 300W. Is this the information you're looking for?

    I'm just confused because normally I see the driver voltage listed, but not the transducer voltage listed as in this case:

    http://www.ebay.com/itm/60W-28K-Ult...onic-cleaner-power-driver-board-/110991762210

    The product I saw on Alibaba is exactly the voltage, frequency, and material I'd like, but they're one of the few sources that actually lists a voltage with their transducer.
     
  15. Mar 10, 2015 #14

    Baluncore

    User Avatar
    Science Advisor

    If your driver electronics needs an input of 110VAC 50/60 Hz mains power to operate, and it will get 110VAC mains power, then that is OK. That is the end of the low frequency power discussion.

    Re: Ultrasonic generator PCB: http://www.ultra-piezo.com/Products/ultrasonic-generator/ultrasonic-cleaning-generator/8.html
    The second last picture appears to show the PCB with “110V OR 220V” power input.
    But does the “110V Ultrasonic Generator PCB” device run on mains power of 110VAC 50/60 Hz, does it generate a 110VAC amplitude 28kHz ultrasonic signal, or maybe both. How can you tell? Is there a data sheet with full specifications for the product? I am still very suspicious of your assumption that it has a 110V ultrasonic output.
     
  16. Mar 10, 2015 #15
    I tried getting in touch with the company but no luck. I suppose the only way is to measure the output. If it's 110V, then I guess I can get a step up transformer rated for 400W (a bit higher than I need). In this case, the flow diagram would go like: 110VAC outlet --> 110VAC/300W/28KHz driver --> 110VAC to 220VAC transformer --> 220VAC/300W/28KHz transducer.
     
  17. Mar 11, 2015 #16

    Baluncore

    User Avatar
    Science Advisor

    A transformer at 28kHz can be made but should be avoided if possible.
    My guess is that you will need only 50 watt to drill the holes.
     
  18. Mar 11, 2015 #17

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

  19. Mar 11, 2015 #18

    Baluncore

    User Avatar
    Science Advisor

    I find the biggest problem with buying “Chinese Junk Brand” modules is that the technical specifications are unavailable.

    It is usually quicker to buy one and measure it, than it is to wait for accurate detailed data, from a shop-front that does not understand the technical details of what they are selling.

    Building your own driver means that you can choose the exact specifications. If you will produce multiple units then total control of design is good, but if you only need one it is often quicker to buy the Chinese unit and then “make” it to do the job.
     
  20. Nov 3, 2015 #19
    Welcome to visit
    If you have any problems about ultrasonic frequency adujsting
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: How can I fine tune an ultrasonic frequency?
Loading...