Building driving circuit for ultrasonic transducer - 100khz 60W

Thread Starter

WreXplorer

Joined Jan 28, 2016
6
I am trying to drive a 120khz, 60W ultrasonic piezoelectric transducer. To do so I have an arduino mega2560 producing a 120khz frequency and am feeding that into the circuit to switch a TIP102 transistor. This transistor is feeding power to my transducer from a 12v power source. I intend to connect an inductor in parralel with the Transducer as this will amplify the output but will do so at a later time. For now I am just trying to get any output at all. So far I have been unable to get the transducer to do anything. I am basing my (very simple) driving circuit on this(full image found below), a schematic posted here. My transducer is this. The specs of the transducer are as follows.
  • Frequency 120 KHz
  • Power 60W
  • resonate frequecy: 120±1.5 KHZ
  • capacitance: 4000 ±10% PF
  • resonate impedance :<=25Ω
  • Isolation Resistance :>=100MΩ
Can someone please let me know what I am doing wrong here or what other avenues I should run down to get this working?
Thank you!

 

Thread Starter

WreXplorer

Joined Jan 28, 2016
6
The more reading and testing I do, the more i am suspect of the transistor not properly switching with the high frequency. Any recommendation for a high speed transistor capable of switching 12v at over 5amps. That and thoughts on the capacitor and inductor values.
 

Thread Starter

WreXplorer

Joined Jan 28, 2016
6
Just out of curiousity, what range are you trying to detect, and in what medium?
The detector in the schematic is not something I am using. As I describe above, I am trying to produce an ultrasonic frequency by driving the transducer. Not sensing anything with it.
 

Thread Starter

WreXplorer

Joined Jan 28, 2016
6
I would advise you to use a transformer and a push-pull amplifier FET.
So with just some quick research, I assume you mean I should use a transformer to amplify the signal from the microcontroller? As the Fet would appear to require over 10v to activate and the microcontroller only outputs about 5v. Also, would a fet be fast enough to perform this task at the 120khz frequency?
 

Bordodynov

Joined May 20, 2015
3,177
Of course in addition to the two output transistors require more driver circuit. It is preferred to supply the output transistors used 12 volts. I'll think about it a little bit and offer the scheme. But it will not be until Monday, and if I'm not too busy. For 60W justified complicating the electronic circuit. The prototype clearly has less power.
 

Alec_t

Joined Sep 17, 2013
14,280
A piezo transducer behaves somewhat as a capacitor. As your driver transistor allows only a single polarity intermittent current to flow in the transducer (with or without D1) there is no oscillation as would be required for an ultrasonic output. The transducer's capacitance simply charges up to about the supply voltage.
What is the purpose of D1 in our schematic?
 

Thread Starter

WreXplorer

Joined Jan 28, 2016
6
A piezo transducer behaves somewhat as a capacitor. As your driver transistor allows only a single polarity intermittent current to flow in the transducer (with or without D1) there is no oscillation as would be required for an ultrasonic output. The transducer's capacitance simply charges up to about the supply voltage.
What is the purpose of D1 in our schematic?
D1 is there for the inductor. Inductor causes voltage spikes and I assume having those pass backwards through the circuit is less than ideal. Now for the moment I don't have an inductor but adding it later, as well as the diode will happen. Mainly to protect the transistor.

Is there not a transistor that will support switching 60watts at a high frequency like this? The most I have seen would require a higher base voltage than I am capable of having from my microcontroller. I am becoming quite sure that the transistor is just too slow.
 

Alec_t

Joined Sep 17, 2013
14,280
D1 is there for the inductor.
Then it's wrongly placed. The anode should go to the transistor collector and the cathode should go to the +ve rail.
Is there not a transistor that will support switching 60watts at a high frequency like this?
Plenty of N-channel power MOSFETs can do that, but unless they are 'logic-level' types they would need a level-shifter drive circuit for the gate to ensure they can be fully turned on.
 
Top