• Forums
• Posts
Latest Posts
Active Posts
Recently Visited
Search Results
• Page Extras
• Forum Themes

### Questions aboout piezos and theory around them

Page: 12345 > Showing page 1 of 5
Author
bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
0

# Questions aboout piezos and theory around them

I have been reading a lot of material about piezos. Specifically, I am playing with some piezo film. As many hours as I have read, I am not grasping a few ideas.

My goal is to resonate the piezo at a very high frequency. At first, I tried to hook the piezo to a pin of the pic and pulse DC into. I got nothing. I tried every frequency and duty cycle.

Then, I used a 1:10 transformer and put the piezo on it. The transformer would put out 50Hz AC. The piezo buzzed like crazy! When I would lower the AC voltage to it, it would become more quiet.

I wanted to get the film to put out 1mHz. So, using ccp I put out a perfect 1mHz Pulse at a 50% duty cycle...and nothing. DC produces nothing. At 1 mhz, the transformer appears to not put out any AC voltage.

I have examined circuits. Many of them dont even have AC transformers. They have inductors. So, I am a little confused as hot to go about doing this.

Can anyone give suggestions?

-I dont know if I need to send a 1mhz ac current. If so, my inversion from dc to ac is not working at 1mHz, but does at 50hz.

-to get the most power from the vibration of a piezo, do I send it more volatage, increase the frequency, or simply hit the resonate frequency?

-With resonate frequency, can I use octaves and get the same affect? Do higher octaves make more power? Can anyone tell me how to use an inductor?

-Is there a VERY SIMPLE circuit in which I can connect the piezo to ground and to a pic pin and vary pulses to find the resonate frequency?

Sorry for all the questions. I have read so much and it is either not enough or so technical that I cant grasp it. So, I figured I would ask people who have used it here. Thanks

### 83 Replies Related Threads

idefix
Senior Member
• Total Posts : 141
• Reward points : 0
• Joined: 2009/10/29 09:58:14
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/23 09:56:08 (permalink)
0
For the transformer,  bear in mind 1Mhz is some 12,000 - 20,000 x its intended operating frequency,  losses that are insignificant at 50-60Hz will come into play.  The inter-winding capacitance of the transformer's primary may provide a lower impedance to your 1Mhz
signal than the primary's inductance, so no magnetic field results.  Then, the soft iron core of the ransformer may not react in time to
such a fast excitation;  high frequency cores are often made from sintered, small particle,  ferrites rather than from plain soft iron.
I kind of expected that piezo crystals, rather than film, would be a better choice for a resonator because the flexible film substrate is intended more for sensing applications,  the floppy film would make for a lower, low energy resonance. I found a helpful document while googling "ultrasound scanner transducer piezo film"  :

bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/25 16:48:42 (permalink)
0
Ok, I have read that document. I have examined a unit that does what I am trying to do in a nutshell. Since I know know that there are specifically transformers that are made for high frequency, I can see that is what they have in their device.

They also have an inductor, but I am not sure what for.

Here is what I am actually trying to do. I would like to vibrate the film at a high rate to vaporize a water drop on it. I dont know if this would be best achieved by applying a low resonate frequency AC to it. Or, if I should a high octave frequency AC to it.

It seems like a high frequency will cause a faster negative pressure behind the water to evaporate it.

Any suggestions welcome. Thanks!
kenn
Super Member
• Total Posts : 226
• Reward points : 0
• Joined: 2008/01/19 09:23:48
• Location: Toronto ON
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/25 18:35:00 (permalink)
0
You need to do some basic work using pieces pf piezo materials, a variable oscillator, and maybe a small amplifier circuit. This will quickly give you some insight as to what various pieces of piezo material will oscillate at. I suspect that 1 MHz is far too high a frequency; you may have more luck investigating supersonic frequencies (30 kHz to 100 kHz).  Maybe even buy some surplus supersonic piezo transducers.

what, me assemble?
idefix
Senior Member
• Total Posts : 141
• Reward points : 0
• Joined: 2009/10/29 09:58:14
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/25 18:48:50 (permalink)
bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 08:16:46 (permalink)
0
From what I have seen, they range from 1-3.3 mHz.

I found a couple very nice scopes on craigslist here in Kansas. I wonder if they typically have sine outputs to test with?

I need to find a high frequency transformer. I dont know what else they call these, but I cant find anything on ebay. Maybe spark fun has them. 50Hz is the highest they sell in town ;) I have one in a broken piece of gear I have but it is a \$2500 piece I am afraid to take apart.

idefix
Senior Member
• Total Posts : 141
• Reward points : 0
• Joined: 2009/10/29 09:58:14
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 09:24:03 (permalink)
0
Those links I spent time finding .. What was the bit in the middle abut a ferrite toroid ?
bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 09:46:21 (permalink)
0
Oh, I am sorry. I did read all of those. I dont know if you accidentally put the wrong page for steminc on there, because it is a blank page. Maybe you were just linking me to their home page?

http://www.electro-tech-o...ircuit-humidifier.html

I read all 11 pages a few days ago, but by the end I think I was more confused because so many people put in different ideas. I will read it again. I usually have to do that ;)

I like the page you send me that shows how to build them. Now I know some of the terms I was confused on.

I found a nice scope. They were asking \$300 but I talked them down to \$100 ;)
idefix
Senior Member
• Total Posts : 141
• Reward points : 0
• Joined: 2009/10/29 09:58:14
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 11:55:24 (permalink)
0
Tell about the scope !  I keep hearing people get nice scopes for short money but I can't do better than \$500 or so for a 454.
Somewhere in that big thread, someone wrote about using a ferrite and 10Base-T wire which seems a sensible way to go. I realised a short while ago you possibly don't mean 'resonate' resonate ( thanks, Whoopi ) but you are trying to drive a piezo into vibration with a fairly high amplitude ~1MHz square wave; in which case, current limit your PIC output
with ( 5v / 25mA ) somthing like 220 - 330 ohms and try a few turns of hookup wire around  a ferrite ring and scope the output of, say, 10 x primary turns for a secondary.
Fiddle with the ferrite size and No. turns and see if results improve. I think I'd use solid wire over stranded but I've forgotten what Litz winding was all about.
The Stemic site home page has a sidebar with  'piezo mising ' or somesuch listed, the P/N has changed over the one listed in one of the circuit diagrams. Stemic don't seem to offer any circuit examples.
When all else fails, buy a Glade Misting Air Freshener .. <<  \$2500 !

bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 12:08:32 (permalink)
0
Cool. I am headed to radioshack to get a sample package of inductors. The oscilloscope I am buying is \$100 and is a

## Tektronix 465B

idefix
Senior Member
• Total Posts : 141
• Reward points : 0
• Joined: 2009/10/29 09:58:14
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 12:24:37 (permalink)
0
Scope envy.
bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 12:27:34 (permalink)
0
Oh, and can you tell me if I am on the right track.

From what I have read,

-A piezo has a resonant frequency
-This frequency can be achieved by using an inductor and capacitor
-A person can change the inductor size, as long as the capacitor is changed according to the LC calculation
-The magnitude can be boosted by applying more voltage

OR

-A piezo can have a high frequency AC transformer drive it...as long as the frequency is close to resonance
idefix
Senior Member
• Total Posts : 141
• Reward points : 0
• Joined: 2009/10/29 09:58:14
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 14:14:57 (permalink)
0
I guess from reading the characteristics of Steminc's 'Mister Piezo' , where they state an impedance at resonance of 2-3 ohms then they are intended to be operated at the series resonant frequency where impedance is lowest. From the discussion threads it looks as if it's OK to drive them open-loop at close to that frequency.  I'm thinking, you drive your HF transformer with the secondary winding unloaded, connected to the probe and common of the oscilloscope and verify
you get a relatively steady high output as you drive the primary from your PIC between, say , 500KHz and 1.5MHz ( Square wave 50% PWM, varied repetition rate ).  Then, connect the piezo across the secondary with the scope probes still in place and sweep the frequency again. The output voltage should dip when you hit series resonance and should peak some tens of KHz higher at parallel resonance. I'd be less hopeful of seeing resonance with a piezo film but I don't know.

PS the B+H Precision signal generator goes up to 2.0MHz, with your bargaining skills you should be able to get one for about a buck fifty !

bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 14:26:57 (permalink)
0
Well, Ill see if one of those are for sale. There is one that I found that only goes to 20kHz. Check this out:

bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 14:35:24 (permalink)
0
Thanks for the explanation. That makes a lot more sense. I have to get the high frequency transformer shipped. :(

I dont believe those are in the mister units at the store. Maybe I need to go get one?

I did go get a few toroids at radioshack.  I wish I had a signal generator so I didnt have to guess. Here is what I can get in town for \$50:

SCHLUMBERGER MODEL # SG-5218 SINE SQUARE AUDIO GENERATOR
idefix
Senior Member
• Total Posts : 141
• Reward points : 0
• Joined: 2009/10/29 09:58:14
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 15:30:03 (permalink)
0
Could you put the PWM code in a timing loop,  ramping the frequency by 10KHz every 100mSec or perhaps under pushbutton control ?  My other scope / sig Gen is a PIC18f ..
kenn
Super Member
• Total Posts : 226
• Reward points : 0
• Joined: 2008/01/19 09:23:48
• Location: Toronto ON
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 15:52:15 (permalink)
0
Scope envy - I still have a Hitachi V-212 i bought new like 18 years ago. Still minty, but a little dusty too. As is its owner. Ahem.

From about the same period, I also bought & built some Heathkit test gear, including an oscillator. They're handy... anyone you could borrow one from?

You could probably whip up a basic adjustable oscillator using an op-amp or two, or a 555 timer IC. or even program one into a small PIC.

I've wound some little toroid transformers that worked nicely at around 1 MHz. You can get some nice toroid transformers (and cores) out of trashed PC power supplies, trashed receivers and VCRs, etc.

what, me assemble?
bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 15:52:56 (permalink)
0
Yes. All that I wrote was on a 16f chip. What I did was ramp PR2 from 0-256 and then increase ccpr1l by 1 and repeat. Each increment is a little slower than what you said. But, the led glows according to the period and duty cycle. I think this covers every period at every duty cycle. Since the ac transformer (activated through and npn from the pic pin) is 60Hz, you can hear it slowing phase out as the frequency gets up there. Same is true for the piezo. I will try to post a video. I wish I had that high freq transformer. I am tempted to remove it from my unit...

`  ; *******************************************************************   ; * NOTE: The PIC16F690 requires the AC162061 header for debugging  *   ; *******************************************************************     #include <p16F690.inc>       __config (_INTRC_OSC_NOCLKOUT & _WDT_OFF & _PWRTE_OFF & _MCLRE_OFF & _CP_OFF & _BOR_OFF & _IESO_OFF & _FCMEN_OFF)         cblock 0x20  cnt1 cnt2      endc                org 0    Int      errorlevel  -302       banksel     OSCCON            ;enable 8mhz internal osc      movlw       B'01110000'         movwf       OSCCON       banksel        TRISC              ; disable the o/p driver for the PWM module      movlw        b'11111111'      movwf        TRISC        banksel        PR2                ; using TMR2 as the timebase for PWM & setting its period to max to give max PWM resolution      movlw        b'00000001'      movwf         PR2              banksel        CCP1CON            ; load CCP control register with appropriate values - 00001100 = PWM mode      movlw       b'00001100'      movwf         CCP1CON        banksel        CCPR1L            ; load msb of duty cycle      movlw        b'10000000'        ; 8 msb of duty cycle             (00000000 in this example)     movwf        CCPR1L        banksel        CCP1CON            ;load lsb of duty cycle      bcf         CCP1CON,4        ;5:4 = dcb = 2 lsb of duty cycle (01 in this example)     bcf         CCP1CON,5        banksel        PIR1            ;clear tmr2 interrupt flag      bcf            PIR1,TMR2IF             banksel        T2CON             ;0:1 are the prescaler      clrf         T2CON            ;3-6 are postscaler, not using     bsf            T2CON,0      bsf         T2CON,2            ;enable timer 2 with bit 2     Wait      btfss        PIR1,TMR2IF      goto        Wait        banksel       TRISC        ;enable the o/p driver in TRISC      bcf         TRISC,5    ChangeCycle     incfsz        cnt1,f     goto        ChangeCycle     incfsz        cnt2,f     goto        ChangeCycle      banksel        PR2     incf        PR2,f      clrf        TMR2      banksel        CCPR1L     incf        CCPR1L,f      goto        ChangeCycle               ;-----------------CALCULATIONS------------------------------- ;PR2 = (clock speed /( desired freq x prescaler x 4) ) - 1  ;PR2 = (8,000,000/(1,000,000 x 1 x 4)) - 1 = 2 - 1 = 1 = 00000001   ;Duty cycle = percent / desired freq    = .50 / 1,000,000 = .00000005  ;CCPR1L: CCP1CON<5:4> = (Duty cycle x clock speed) / 4 = (.0000005 x 8,000,000) / 4 = 1 = 00000000:01     ;***********************************************************        end            `

bonedoc
Super Member
• Total Posts : 1017
• Reward points : 0
• Joined: 2009/11/14 21:58:31
• Location: 0
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/26 15:56:16 (permalink)
0
Ok, I am a little confused by one thing. Why does every circuit I have seen have a toroid in it and NO high frequency transformer. If I order this high fre unit, I should not need a toroid, correct? I can just out output a high freq pwm from the pic to the transformer like I do with 60 hz, correct?
idefix
Senior Member
• Total Posts : 141
• Reward points : 0
• Joined: 2009/10/29 09:58:14
• Status: offline
Re:Questions aboout piezos and theory around them 2010/10/27 08:17:08 (permalink)
0
How do you define this "HF Transformer" ?  The ferrite is the core part of a transformer and you thread windings onto it to make make up the turns and winding ratio you decide you need. I'm not sure what else you expect in the item you describe. Perhaps Kenn can add if there's a preferred size for the ferrite for the frequency of interest. Some 'HF' transformers e.g. between IF stages in a radio may have ferrite dust cores, I don't think they'd work well for the power level you need.
Page: 12345 > Showing page 1 of 5