# Ultrasonic Wave Generator (from 50 - 60 Hz to 20KHz and 2KW of Power)

#### AgustinAvila24

Joined Jun 18, 2017
8
Hello my name is Agustin Avila and I'm trying to make an ultrasonic wave generator for an ultrasonic fatigue machine. I already have a power supply like this but its not mine so I would like to make one but don't know where to start. If someone has an schematic or something it would be of much help.
PD: I'm a mechanical engineering student so I'm not so familiarized with electronics but I'm eager to learn for not be just an user of things.
I upload some pictures of what I have

#### Papabravo

Joined Feb 24, 2006
19,250
I understand power supplies, but I'm missing the connection between the power supply and the ultrasonic transducer.
What does the label "Vreal=vx3+33V" mean?
What is it that consume 2 kW of power?

#### AgustinAvila24

Joined Jun 18, 2017
8
I understand power supplies, but I'm missing the connection between the power supply and the ultrasonic transducer.
What does the label "Vreal=vx3+33V" mean?
What is it that consume 2 kW of power?
hello and thanks for the answer
One of my professors came with the one in the first photos but its not from my laboratory and there are some other issues related to the ownership of the instrument so in my laboratory we decide to go with one made by ourself
i'm trying to make a ultrasonic wave generator like one i read in "Gigacycle Fatigue in Mechanical Practice" by Paul C. Parias and Claude Bathias.
In the book says they use
" ultrasonic generator 900BA is made by Branson Ultrasonic Corporation" (page 78)
I want to make something similar to that

#### AgustinAvila24

Joined Jun 18, 2017
8
I understand power supplies, but I'm missing the connection between the power supply and the ultrasonic transducer.
What does the label "Vreal=vx3+33V" mean?
What is it that consume 2 kW of power?

#### Papabravo

Joined Feb 24, 2006
19,250
Your block diagram shows an Analog to Digital and a digital to analog device. These devices are used to send an receive analog data; they are not designed to transfer power. What I want to know is how the transducer receives and converts power into mechanical energy. Is the power involved AC or DC? What voltage and current are involved? Please try to answer my previous question; what does the label on the front panel of the power supply "Vreal=vx3+33V" mean?

#### AgustinAvila24

Joined Jun 18, 2017
8
Hello and thanks again for answer: The label in the front panel i don't know what it means all i know about the ultrasonic generator that i have here is that the power is of 300w, the working frequency is 20Khz, it has a counter that measures the cycles of the ultrasonic wave, and thats all, its not much but its what the manufacturer told me (and i could 't get more info) and well the whole thing works like this:
the ultrasonic generator generates an ultrasonic wave with a certain voltage that goes into an arrange of piezoelectric crystals (like those in ultrasonic cleaning)and the electric signal is transformed into a mechanical strain that goes into the ultrasonic horn (sonotrode) that it's an mechanical amplificator that according to the geometry amplifies the mechanical wave and that its the input in the specimen that is beeing tested also the specimen has an special geometry that amplifies more the strains in the middle section.

#### Alec_t

Joined Sep 17, 2013
13,147
If the transducer is rated for 300W why do you want a 2kW generator? Twiddle a knob wrongly and you'd destroy what looks like an expensive transducer.
Given that the transducer resonant frequency may depend on the mechanical acoustic impedance it's driving (check with the manufacturer), a self-resonant (auto-tuning) drive circuit may be needed.

Last edited:

#### AgustinAvila24

Joined Jun 18, 2017
8
If the transducer is rated for 300W why do you want a 1kW generator? Twiddle a knob wrongly and you'd destroy what looks like an expensive transducer.
Given that the transducer resonant frequency may depend on the mechanical acoustic impedance it's driving (check with the manufacturer), a self-resonant (auto-tuning) drive circuit may be needed.
hello and thanks for answer, yes its true that i have to design first the transducer i'm working on it currently but it will also depend in what are my boundary conditions, the current machine it's not meant to reach the failure of the specimens so i want to design one that will.

#### Externet

Joined Nov 29, 2005
1,922
Saludos, Agustín. Allow my spoon in your plate...

If I had to build an 50 Hz to 20KHz oscillator capable of 2KW as titled, I would use an audio power amplifier capable of 2 KW driven by a function generator or ´direct digital synthesizer' DDS.
A random find ----> https://acrobotic.com/featured/kit-00002
And a random amplifier ----> http://www.ebay.com/itm/1X2000W-2kw...DOM-Module-Subwoofer-/111559468923?rmvSB=true

Using them as building blocks, together with a transformer for output impedance matching/voltage step up.

Unless am peeing out of the can and unrelated to your application.

#### Papabravo

Joined Feb 24, 2006
19,250
I can see that we are getting nowhere at a very rapid rate. You still have not described the parameters involved in powering the transducer. I have 20 kHz and 300 Watts, but no idea what voltage and what current we are talking about. Maybe we could get a scope picture of the driving waveform. Is it a square wave or a sinewave or something else?

#### AgustinAvila24

Joined Jun 18, 2017
8
I can see that we are getting nowhere at a very rapid rate. You still have not described the parameters involved in powering the transducer. I have 20 kHz and 300 Watts, but no idea what voltage and what current we are talking about. Maybe we could get a scope picture of the driving waveform. Is it a square wave or a sinewave or something else?
Thanks for answer me and yes i think that i'm lack of information and i'm not being clear but because of the lack of information on my side.
Thanks anyway.

#### Alec_t

Joined Sep 17, 2013
13,147
If I had to build an 50 Hz to 20KHz oscillator capable of 2KW
That's how I read it at first, but I think it's a language issue and the OP means the oscillator is a ~fixed 20kHz but is powered from 50/60Hz mains.

#### Papabravo

Joined Feb 24, 2006
19,250
That's how I read it at first, but I think it's a language issue and the OP means the oscillator is a ~fixed 20kHz but is powered from 50/60Hz mains.
That is certainly possible, but has it escaped everybody's notice that 20 kHz is not exactly ultrasonic. To me ultrasonic means beyond the range of human hearing. The original CD sampling rate of 44.1 kHz was chosen precisely because half that rate, or 22.05 kHz, is beyond the range of human hearing. A power supply powered by the mains is a good idea for a 2kW power level, but we still have not established what kind of power supply it is. I understand the need for the TS to design his own equipment, but in order to design something you must first understand the requirements, and so far we are no closer to that understanding. The notion of a typical audio amplifier driving this transducer has a certain abstract appeal, but I don't think it would work all that well, but I don't really know any more than the rest of you. Perhaps a link to the datasheet for the transducer.

Last edited:

#### Externet

Joined Nov 29, 2005
1,922
Agustín : I can fully translate accurately any language difficulties for the members to get the right meanings. Puedes decirlo en español.

#### Papabravo

Joined Feb 24, 2006
19,250
Agustín : I can fully translate accurately any language difficulties for the members to get the right meanings. Puedes decirlo en español.
I don't think this is a language problem. This is a ME student with no knowledge of electronics trying to infer a set of requirements for a device he needs to construct. He needs help in formulating those requirements.

Here are two easy questions:
1. Do you have the Converter?
2. Are you trying to duplicate the Branson Ultrasonics 900BA?

#### Simsmolin

Joined Feb 22, 2018
1
Hello! you are still interested in this topic

#### ebp

Joined Feb 8, 2018
2,332
Is this device to deliver 2 kW mechanical power to the item under test or is it to deliver 2 kW to the transducer? There might be a VERY big difference.

As has been pointed out, ultrasonic transducers are generally resonant devices (piezoelectric). Whether you call 20 kHz "ultrasonic" or not is irrelevant. There are certainly transducers that will operate "audio" spectrum to a few tens of kHz. With a resonant transducer, I suspect the the mechanical loading might spoil the Q substantially. It is clear that his objective is to put substantial high-frequency energy into "things" to test their mechanical properties.

Without knowing the transducer characteristics it is impossible to come up with a suitable circuit. Externet's suggestion of a big audio amp could work. Clearly the breadboard photos are switched mode circuits, which is likely to be vastly more efficient if compatible. The approach with a resonant transducer is much different from a non-resonant (broadband). The big question is transducer impedance in the bandwidth of interest. It might require kilovolts.

Branson is a long-established company that has (or used to have) a wide range of ultrasonic products - cleaners, welders (the drawing looks like a welder, to me), non-destructive test equipment (where "ultrasonic" transducers run in the 5 to 20 megahertz range).

This could be a big and complicated project, involving circuitry quite capable of causing serious injury or death.

#### ebp

Joined Feb 8, 2018
2,332
Note to self: pay attention to dates in threads

#### AgustinAvila24

Joined Jun 18, 2017
8

#### AgustinAvila24

Joined Jun 18, 2017
8
Is this device to deliver 2 kW mechanical power to the item under test or is it to deliver 2 kW to the transducer? There might be a VERY big difference.

As has been pointed out, ultrasonic transducers are generally resonant devices (piezoelectric). Whether you call 20 kHz "ultrasonic" or not is irrelevant. There are certainly transducers that will operate "audio" spectrum to a few tens of kHz. With a resonant transducer, I suspect the the mechanical loading might spoil the Q substantially. It is clear that his objective is to put substantial high-frequency energy into "things" to test their mechanical properties.

Without knowing the transducer characteristics it is impossible to come up with a suitable circuit. Externet's suggestion of a big audio amp could work. Clearly the breadboard photos are switched mode circuits, which is likely to be vastly more efficient if compatible. The approach with a resonant transducer is much different from a non-resonant (broadband). The big question is transducer impedance in the bandwidth of interest. It might require kilovolts.

Branson is a long-established company that has (or used to have) a wide range of ultrasonic products - cleaners, welders (the drawing looks like a welder, to me), non-destructive test equipment (where "ultrasonic" transducers run in the 5 to 20 megahertz range).

This could be a big and complicated project, involving circuitry quite capable of causing serious injury or death.
Hello, first of all thanks for getting interest me n this topic.
1. Sorry if it took me a while to answer the replys
2. Actually I had design the transducer, sontrode, and specimen speaking of the mechanical part (everythíng related to wave equations and frequencys of resonance).
3. In the beginning I was trying to replicate a Branson power source for ultrasonic cleaning, so I started look for some schematics but I couldn't find something useful, maybe because I have little knowledge of circuits and stuff. Then I found this equipment at my faculty and I saw the opportunity of making reverse engineering but that was another dead end.
So to sum up, I put the electrical part in stand by, I solved the mechanical parts and i am again dealing with this. My options now are: or I buy something similar to the equipments that I read about, or keep searching for some schematic well explained in order for me to understand it jajaja. I think that it would be the first option.
4. The power source has to be able to give that power. And as for the transducer the mechanical power converted from the electrical is a matter of the piezoelectric characteristics of the cristal I choose.
5. Yes the language it could be the mayor problem when I state my inquiries. It's very stressing for me speak my mind in my native language so it gets a lot more in another language jajajjaa