• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 47 subscribers
  • Views 918 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Piezo driver

Kyle Aven
Intellectual 950 points
Other Parts Discussed in Thread: OPA452, OPA544

Hello,

I need help designing a circuit that will be used to drive a piezo disc. The piezo specs are as follows:

Res. Freq:  2.8kHz

Max Impedance: 200 Ohm

Cap. @1kHz : 24,000 pF

 

We will using a DAC to produce a sine sweep from ~0-20kHz. The output rails we are aiming for are +/-35V (flexible). Our initial thoughts were to use a lower power amplifier to amplify the signal right out of the DAC, then use larger more powerful amplifiers for the final output stage to the piezo. Last time we tried this the output stage was undersized for the task and failed in a short period of time. Any help would be appreciated. I dont have much experience in driving high capacitive loads. My approach has always been the more output current capability the better. I would really like to learn how to determine the proper amplifier for task like this. If someone could share the calculations and most important specs to concentrate on it would be great.

thanks

kyle

  • 0
    TI__Guru* 93105 points

    Hi Kyle,

    Basically, the piezo load represents a capacitive reactance to the driving amplifier. You mention a frequency range ~0 to 20kHz, and the capacitive reactance will be minimum at the highest frequency, 20kHz. Therefore, that will be the point where maximum drive current is required. But to get to that current the slewing rate and capacitive reactance first must be determined:

    Slew rate (SR):     fMAX= 20kHz       VPK= 35Vpk

    SR = 2 x pi x fMAXx VPK = (6.28) (20e3) (35) = 4.4e6V/s = 4.4V/us

    Capacitive reactance (XC):    CL= 24nF

    XC = 1 / ( 2 x pi x fMAX x CL) = 1 / ((6.28) (2e4) (24e-9) = 332 Ohms

    Peak load current (IPK):

    IPK = VPK / XC= 35 / 332 =  0.105A

    Thus, an operational amplifier having an output swing capability of +/-35V, about +/-100mA output current is needed and a slewing rate of at least 4.5V/us. These requirements indicate that a power operational amplifier, or an operational amplifier with an external drive stage would be required.

    The combination of high-voltage output and high current makes this more difficult to fill with an operational amplifier alone. About the closest TI operational amplifier that could handle this application is the OPA542. It would have to be operated with +/-40V supplies and I would suggest using either two of the OPA542 amplifiers working together as shown in data-sheet Figure 7, or adding the two external output transistors as shown in Figure 8, to get the output current to the required level.

    Also, as mentioned the piezo element is a capacitive load and most operational amplifiers cannot drive the associated level of capacitance without become unstable. This is due to the loop's diminished phase margin. Therefore, it will be necessary to provide external compensation. Likely, something similar to that shown in Figure 5. The actual compensation needed will depend on the final circuit configuration.

    Regards, Thomas

    PA - Linear Applications Engineering

     

     

     

  • 0 in reply to Thomas Kuehl
    Intellectual 950 points

    Thank you for your timely reply Thomas. Your post was very informative. Searching for the OPA542 datasheet has not produced any results. Could you have the part number wrong in your post or am i missing something? 

  • 0 in reply to Kyle Aven
    TI__Guru* 93105 points

    Hi Kyle,

    You are correct; I transposed the numbers in the model. It should be OPA452 - not OPA542.

    Thanks!

    Thomas

    PA - Linear Applications Engineering

     

     

  • 0 in reply to Thomas Kuehl
    Intellectual 950 points

    Thomas,

    After looking at the specs you calculated, is there a reason you did not suggest something like the OPA544 which has the higher current capability? Also do gain on these output stages cause problems. I believe max gain would be ~10V/V

  • 0 in reply to Kyle Aven
    TI__Guru* 93105 points

    Hi Kyle,

    The only reason I didn't suggest the OPA544 was because you had mentioned an output voltage of +/-35V. The OPA544 maximum supply is rated at +/-35V and its actual output swing rails will be less depending on the output current required. The graph in the lower right-hand corner of data-sheet, page 5, shows the swing to the rail for different output current levels. An output current at the 100mA level would allow the output to swing closer to the supply rails than the 500mA line shows.

    If you have some leeway on the output swing such as 30VPK, then the OPA544 is a good fit for the application. The closed-loop gain of 10V/V isn't a problem for the operational amplifiers we have discussed.

    Regards, Thomas

    PA - Linear Applications Engineering