parallel connecting ultrasonic transducer (transmitter)

[aruna1]

Hi
I'm working on a project that need to send ultrasonic signals using multiple transducers in same time.
all signals are identical.
currently I'm using only one transducer and driving it using a h bridge.

and I want to connect other transducers.

so is it ok to connect all 8 transducers parallely and drive all of them using single H bridge?

or do i have to use separate h bridge for each transducer?

 

[kak111]

You can use them parallel, but add fuse for every transducer.
So that if one wire or transducer is shorted others can still work.

KAK

 

[aruna1]

hi
by using them in parallel sums up the capacitance. each transducer has 2.68nf so total is 21.44nf. will this be a problem for the 40KHz?

and I'm thinking about adding a inductor and resistor in series to make it resonate. I'm doing this with hope of increasing the transmission range and power. simulation shows this make +16V to -16v voltage swing. (with 9V supply to h bridge) transducer is modeled as 2.68nF capacitor and 5.91 inductor is used

please let me know your idea about this? will it increase the range?
and also what will be the maximum pk-pk voltage I can give to these transducers? I dont have a datasheet or model number for them. nothing marked on the transducer. I think they are generic.

https://obrazki.elektroda.pl/28_1306072671.png

 

[kak111]

In parallel connection of piezo´s add a resistor 1k for every piezo´and you
dont need fuses. ( Maybe resistor can be smaller value )
We should know piezo data.

So the load is 125 ohm and 22nF and coil L in series

I think 22nF load is not problem ( add supply capacitors near as possible to H-bridge )

with 720µH resonance freq. is 40kHz

I dont know how much coil give more power, but it takes out higher harmonics.
You must simulate it.

General info................

**broken link removed**

**broken link removed**

LC Resonance Frequency Calculator at WhatCircuits.com


Regards KAK

 

[aruna1]

I have simulated it
following is the image
https://obrazki.elektroda.pl/28_1306072671.png

any way insted of using H bridge
can i use two comparators as h bridge circuit for this?
when one comparators out is low other one's out is high (and vice versa) to drive transducer with +9V and -9V ? ( or +16 V and -16V if we use an inductor)

here i have added this schematic

 

[FvM]

I dont have a datasheet or model number for them. nothing marked on the transducer. I think they are generic.

You'll find short specifications from some DIY catalog distributors, e.g. this
**broken link removed**

If I remember right, the aluminium case ultrasonic transducer has been orginally a Murata product, but the presently available types are most likely clones. Murata has now a reduced size MA40S4S type in a plastic case.

can i use two comparators as h bridge circuit for this?

Refer to the LM393 output characteristic in the datasheet to understand why it doesn't work.

 

[kak111]

You can use some low cost H-bridge driver ic like L298 for example.

https://www.st.com/stonline/books/pdf/docs/1773.pdf

look Figure 7 : For higher currents, outputs can be paralleled.

Inputs are 5V logic level.
Now desing simple 40kHz gated oscillator with cmos-logic gates and that´s it.

KAK

 

[aruna1]

H,

Refer to the LM393 output characteristic in the datasheet to understand why it doesn't work.

I don't know why it will not work. is it because maximum sink current is 16mA or s it because outputs are open collector?

I have seen that in SRF04 they use MAX232 to drive transducer directly.MA232 has output current of maximum 10mA which is lower than LM393.

if it is because open collector outputs, I have used pull up resistors in my schematic. I believe they also serve as resistance in RLC circuit.
Other than that I cant imagine why wouldn't it work. Can you tell me why?
I ran a simulation, and in it it is working with LM393

You can use some low cost H-bridge driver ic like L298 for example

L298 and L293 cant run with 40kHz, .That's why I tried to use LM393 as a H bridge. you

---------- Post added at 01:00 ---------- Previous post was at 00:45 ----------

@FvM

what is the maximum voltage I can use to drive them?

is it 20V peak - peak
or 40V peak - peak (20V each polarity)

 

[FvM]

I don't know why it will not work. is it because maximum sink current is 16mA or s it because outputs are open collector?

Referring to ohms law, you would need a sink current of > 70 mA in your circuit, considering 5V supply and no additional load impedance. You have been talking however of 9V supply.

Open collector is simply unsuitable. You should go for a low impedance impedance push-pull driver. Your previously suggested discrete transistor circuit is basically of this kind, although I'm not sure if cross-conduction may occur in the particular circuit. My personal favourite would be a high current gate driver like TC427 or TC428. For a single ultrasonic transducer, standard CD400x CMOS devices are usually suggested, but you'll need lower impedance/higher output current for the parallel circuit.

There's nothing against using a number of CD 400x ICs, of course, e.g. CD4049 inverters.

what is the maximum voltage I can use to drive them?

The linked "data sheet" mentioned 20 Vrms. Murata in contrast only specifies 20 Vpp square wave for it's newer 10mm MA40S4R/S, corresponding to about 10 Vrms. You should however consider the voltage "amplification" achieved by the resonant circuit. So a 20V supplied H-bridge with series inductor may be already too much.

But it's not clear, how the maximum ratings have been exactly evaluated. As I already mentioned, there's an absolute level set by the piezo depolarization voltage. You would need to know the piezo disc thickness for a reasonable estimation. Below this limit, either breaking or wear in long term operation may occur.

 

[aruna1]

are you suggesting a mosfet h bridge over bjt h bridge?

 

[aruna1]

I was doing some simulations and following is the latest. its single transistor drive of transducer. I'm using 21.44nF capacitor assuming all transducers are connected in parallel. top graph shows waveform of transducer it has about +11v and -11v
bottom graph shows wave form of between emitter and collector.it has about +16V and -6V.

what is your openion about this method over h bridge?

 

[FvM]

are you suggesting a mosfet h bridge over bjt h bridge?

Not particularly. I was suggesting some ICs. A discrete design can use MOS or BJT as well.

what is your openion about this method over h bridge?

It can work. It's a class C output stage, often used for RF PA. The 10 ohm resistor won't be necessarily needed, but the inductor must be able to carry some DC current. The RMS transducer voltage will be always below the supply voltage, there's no resonant voltage transformation available as with a series circuit.

 

[aruna1]

The RMS transducer voltage will be always below the supply voltage, there's no resonant voltage transformation available as with a series circuit.

I didnt understand this part. can you explain it more clearly?
thank you

And if I can supply 18V and use an h bridge is it necessary to use a inductor? coz h bridge can give +18v and -18V

finally what method do you suggest to implement this? I need to get maximum power possible from this transducers (parallel) so burst transmission distance will be maximum. If you were me what type of circuit will you use to achieve this?

I'm using a 11.1V lipo to power this. ,but if it required I can use boost converter to get higher voltage to drive transducer