Discussion in 'Analog & Mixed-Signal Design' started by alaala73, Aug 25, 2016.

1. ### alaala73 Thread Starter New Member

Aug 25, 2016
3
0
Hi
I need to build some ultrasonic transmitters and receivers, and I found this guide (written by Kerry D. Wong). I built Wong's project and made it work, but before including it in my project I would like to fully understand it.
Could someone please describe to me the transmitter's and the receiver's circuits, how they work, if they are very common circuits or not and so on? (and also if there is a simpler or a better way to do the same thing)

These are the circuits:
transmitter:

Ala

2. ### Nykolas Member

Aug 27, 2013
87
31
First a question: What do you want to do with the thing?

I have seen the transmitter arrangement before. A H bridge where a single transistor would do. The maximum you can get to drive the transmitter is the supply voltage (Common to Vcc less the transistor VCE(sat), with an H-bridge the VCE(sat) is x2). So a single transistor, preferably a FET, would be better.

The preamp has a gain of 73dB. With that amount of gain it should be in a shielded enclosure. Using odd resistor values (80k) is indeed odd. The two 1.2k resistors to create the virtual ground could have a higher value (10k +) and need to have a capacitor (10uF or so) between their junction and ground to avoid oscillation. Also a .01uF between pins 4 and 8 of the IC, close to the pins, is advised.

Enjoy, E

3. ### alaala73 Thread Starter New Member

Aug 25, 2016
3
0

So you mean the first circuit doesn't makes much sense?

For your question, I need to calculate the distance between two robots in a room. I don't have specific needs in max/min distance and precision: when the robots are very close together they don't need the ultrasonic system to know the other robot's position, when they are distant they don't need to know the other's position at all, and in any case the don't need to know it exactly. Obviously I would like to get the maximum precision and max distance... what would be the better way to get them, and what do you think are the better precision/distance values I can get with my transmitters/receivers? Could you suggest me some better circuits? (I have no problems in programming the microcontroller, but I'm not able at all to design circuits)

4. ### alaala73 Thread Starter New Member

Aug 25, 2016
3
0
Sorry, I forgot to tell you witch transmitter/receiver I'm using...
these arte the links to a brief datasheet.
transmitters
I also have these transceivers, but I don't think they are useful for this project (I need only one robot to know the other's position, not vice-versa).

5. ### AnalogKid Distinguished Member

Aug 1, 2013
4,546
1,252
No, it wouldn't.

You have completely missed the raison d'etre for the creation of the H-bridge circuit. True, for each half cycle there are two saturated transistors in the driving string, so a Vcc of 5.0 V appears to be 4.8 V to the device, as opposed to 4.9 V with a single-ended driver. BUT, an H-bridge effectively *doubles* the Vcc that the device sees by reversing both pins. In a single-ended driver, one end of the transducer is at Vcc and the other end is switched between an open circuit (no current flow) and 0.1 V (Vcesat) So there is 4.9 V across the device. For this example, lets proclaim that the current is 4.9 mA. So the total current through the device is 4.9 mA from pin 1 to pin 2, then 0 mA, the 4.9 mA, etc. In an H-bridge, both pins are switched, and not just between a voltage and an open circuit. The current through the device reverses for each half-cycle. So the total current is 4.8 mA from pin 1 to pin 2, then 4.8 mA from pin 2 to pin 1, etc. In effect, the device "sees" the system power as Vcc x 2, and sees twice the current. This *quadruples* the power through the device. Depending on what the circuit is driving, an H-bridge design can be a real bear to get right. But for ***4 times the power*** at the same battery, it's usually worth it.
They could, but they shouldn't.

Just because the virtual ground voltage is intended to be halfway between Vcc and GND, that doesn't mean the currents through the two resistors are equal, and that the virtual ground is in fact at Vcc/2. Any asymmetrical virtual ground currents come through these resistors, shifting the node voltage value. And AC transient currents must be supplied by larger decoupling capacitors at the junction. Lower resistor values provide a much "stiffer" virtual ground by virtue of the lower Thevenin equivalent impedance relative to the other AC and DC impedances in the circuit. Because of the lower impedance, the decoupling capacitors must be larger for any specific decoupling corner frequency, but that is a consequence of good design.

ak

6. ### AnalogKid Distinguished Member

Aug 1, 2013
4,546
1,252
Also, in the transmitter circuit, the base resistor for the lower left bridge transistor (not the far left drive inverter) might be the wrong value. On the right side, the arduino sources current to drive the left bridge pair high and sinks current to drive it low. On the left side, a 500 ohm resistor is the only thing driving the bridge pair high, a much higher driving point impedance than the output impedance of an arduino I/O pin. This is not automatically a bad thing, depending on the peak current through the piezo device relative to the available base current. This is another case where a single-ended driver is not the best circuit for the job.

ak

7. ### ian field Distinguished Member

Oct 27, 2012
4,415
784
It might be worth hunting down some vintage Thorn Consumer Electronics service manuals (Went by many badges) TVs from the ultrasonic remote era. Occasionally they published both transmitter and receiver schematics - each code was a frequency slightly offset from the 40kHz centre. Some of the receivers used a Gilbert cell chip to decode the frequencies. Another source of receiver circuits, is bat detectors. many types have been published in hobby magazines, a lot of them use the same inserts as TVs.

There were also TV remote front panel boards that used an FM frequency discriminator for decoding.