Is there a H-Bridge driver ic for 40kHz piezo ?

Go To Last Post
24 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi,

Is there a H-Bridge driver ic with full-swing output at 12V that can do 40kHz ?

I want to make a dog/cat/bug/insect repellent with a 20kHz - 40kHz ultrasonic sound. I have 4 piezo tweeters of 4kHz-40kHz. An ATmega seems perfect for this, since I can vary the frequency, add a PIR motion detector, check the battery, and so on. The device will use a battery of 12V.
If I use a H-Bridge, the 12V should be enough, so I don't need a transformer.
The current would be about 40mA (20mA - 100mA).

I can make a short-safe H-bridge for 0.1V - 11.9V output, but I would need 8 transistors (BC547/BC557) per H-Bridge. So I hope there is an easier way.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Why not a dual high-speed power mosfet driver?

The 4428 is a dual inverting & non-inverting driver with a stiff output. Rated to 18V and 1.5A, driving 1000pF in 25nsec…

Search Digi-Key or Mouser for 4428 Driver.

Stan

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Or just a pair of logic mosfets. The piezos are capacitive in nature. Just store 01 then 10 in the two outputs at 40KHz (in the compare handler I suppose).

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

bobgardner, I can easily use a whole port of a ATmega8 for the H-Bridge. So I can write 01 and 10 to the bridge, thanks. But just a pair of logic mosfets is that enough, the H-Bridge is at 12V.

Thanks sbennett, the MIC4428 is just what I need (logic input, very fast).
But I can't solder smd. Is there something like that as a DIL ?

.... searching Ebay ..... Ah, I think I found it. There are a number of dual mosfet drivers which I could use, like the TC428.
Thanks, I think I know enough.

Last Edited: Tue. Jul 31, 2012 - 06:25 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
But I can't solder smd. Is there something like that as a DIL ?

According to the datasheet the MIC4428(CN), ZN, BN & YN are PDIP.

Additionally, you don't know you can solder SMD until you try. I find it easier and faster..

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

krazatchu wrote:
...
According to the datasheet the MIC4428(CN), ZN, BN & YN are PDIP.

Additionally, you don't know you can solder SMD until you try. I find it easier and faster..

I didn't read the datasheet very well then. Thanks.
And beleive me, I tried soldering smd.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I used the UCC27424 which is also a bipolair/cmos combination, and this driver can do peaks of 4A.

The UCC27225 is invering/non-inverted, so I should have bought that one, but I added an inverter (HF transistor).

The problem is that is gets very hot with my piezo tweeter at 22kHz. After 15 seconds I can no longer hold my finger on it. If the temperature rises more there are ticks and cracks in the sound.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You can use an RS232 interface such as MAX232 which has a built in charge pump to drive a piezo with +-20V when powered from a 5V supply.

The MAX232 has two RS232 drivers, feed them with opposite polarity square signal and connect the transducer between the two outputs.

kevin

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

sbennett, thanks for the links. Using 4 mosfet transistors for a full H-bridge would do, but I wanted something with less components and less soldering.

kevin_white, I don't think the MAX232 can do that. I need a lot of energy pumped into a piezo at 22kHz...25kHz for a dog repeller. I still found dog poo at 10 meters from my piezo peeper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

... so you want your piezo pooper to be able to throw further than 10 metres :lol:

Ross McKenzie, Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I read that the mosfet driver ic can be used parallel. So I solved it by soldering a second one on top (piggyback). I also sound the piezo for a short time, with pauses in between to let the driver ic cool down.

I'm going to make a second version with the same driver ic, but I used it to drive a H-bridge with 4 mosfets (that's what it is designed for anyway).

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Silicon chip magazine had a similar thing published msny years ago that used a motorola piezo tweeter. See if you can get the circuit.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This is the article : http://www.siliconchip.com.au/cm...
But it can't be read without paying.

I bought the loudest piezo for 23kHz : The KEMO L010
http://www.kemo-electronic.de/en...

But now I read this: http://cap.ee.ic.ac.uk/~jd204/Di...
and this: http://www.edn.com/design/sensor...
A lot of energy is wasted by filling the capacitance of the piezo. If an inductor is used to supply that current, and a certain curve is used, the effenciency and output increase a lot (I mean: a lot !).
I'm still trying to understand this.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Sounds like a tuned circuit. When you hit resonance, then the output will increase.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If it were single ended, a current sense resistor feeding the ADC would be an easy way to locate your maximum power point...

Being in an H bridge will require differential unless you time your sense and protect your inputs.

Alternatively, measure phase angle and alter frequency to decrease the angle.
A brief example here: http://krazatchu.ca/2008/04/17/u...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Why not try the NJR NJU8752 Class D Amplifier for Piezo Speaker. Link to Digi-Key.

Typical application shown in attached jpg.

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks. This ic optimizes driving the piezo.
But I doubt if it is able to drive my 120dB ultrasonic piezo.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

How do you quantify whether it is or is not capable of driving your piezo?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I wonder what the roll off is for the PAM8610..

I've a bunch of these and they don't get warm no matter what I try, even 15v supply and 4 ohm load.
http://cgi.ebay.ca/ws/eBayISAPI....

The sound quality isn't top notch but what do you expect for $7.30
Somewhere it says not to bridge channels, I have not tried that yet...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Some full-H bridges have a current sense output. Its internal so no fancy current sensing is needed. Example: ST L6203. Sure, its a motor driver,but no reason it cannot drive a capacitive load and good to 100KHz.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thank you for all your ideas.

A piezo at 23kHz is a capacitance. So it behaves totally different than a speaker or motor. A good driver for a loud ultrasonic piezo does not just use a square wave, because of the capacitance.
In my case, I don't care how much current is needed, I just use a bigger case for a larger battery.

The NJU8752 is limited for audible frequencies.
I will look into the PAM8610. The module at Ebay can go up to 50kHz. I think I just buy it to test it.
The PAM8610 could also be good, I will have a closer look.

By the way, I test the sound level with my computer and Audacity. My computer with a hand-picked mic can capture 23kHz at 1/4 of its level (just a guess), so I can alter my circuit to see how it changes.
The 23kHz seems to be the best frequency to repel dogs. So I don't need 40kHz, but perhaps I will make sweeps from 23kHz to 40kHz.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Your piezo likely has a high Q factor, it will be considerably louder at specific frequencies determined by the physical geometry of the device...

So you might not want to drive it at 23kHz as it may not be very loud. Sweeping is a good idea, sweeping with feedback is even better...

Or you could get all analog on it and build the driver such that it self resonates at it's natural frequency...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I read that 23kHz is the most annoying to dogs. So I don't want its resonance frequency.

If you check the graph of the frequency response, the KEMO L010 seems to have a peak at 23kHz :
http://www.kemo-electronic.de/en...