Ultrasonic Distance measure circuit question

[engineerj2010]

HI, THIS IS MY FIRST POST. I AM ACTUALLY A STUDENT AT A UK COLLEGE AND I AM STUDYING ELECTRONICS. I HAVE ALREADY PASSED MY COURSE WITH FLYING COLOURS WITHOUT SITTING MY FINAL. I AM NOW WORKING ON A PERSONAL PROJECT AT HOME.

IM JUST HOPING THAT SOME OF YOU GUYS COULD LOOK AT THYE ABOVE CIRCUIT AND TELL ME IF THERE ARE ANY COMPONENTS THAT I COULD GET RID OF TO MAKE THE CIRCUIT SMALLER?

THANKS ALOT IN ADVANCE

 

[Mike_In_Plano]

IC1B is operating at it's very limit on the phase gain curve. The open loop gain is 100 at 40kHz, while your resistors set the called gain for 100.
The second stage has open loop gain to spare with the closed loop gain set at 10. I'd balance the gain between the two stages, that is, make each have 31.6 or so. This will give you the desired gain (1000) without getting so close to the op amp's limits (100).

I built one of these in high school, and one of the big issues was that return strength varied greatly from target to target and over distance. Thus a means of changing your gain on the fly is a great asset. One way is to use a logging detector. Analog devices sales these. Another technique is simply vary the gain using a digi-pot or trans impedance amp.

Finally, if you can get a director of sorts, it will help enormously. I would get return signals from everything in the room, floor, ceiling, walls, people (people were interesting). A little dish would help greatly.

Mike

 

[engineerj2010]

Thanks for the reply,

however could u please explain it to me a little more obviously, lol sorry but I am kinda a novice! are there any parts of the circuit that can be scrapped? also can you explain it more in depth so i can understand!

Thanks x

 

[Mike_In_Plano]

Gotcha,

It sounds like your starting out and may have a fairly big project for a beginner. Why? Because there are concepts and circuits that you have not worked with previously. I'd start by getting a scope and a couple of 555 circuits to get started.

Attach the output of your ultrasonic sensor to the scope and use a 555 circuit to drive the ultrasonic transmitter. Now, play awound with the 555 to get some feel for what frequency the system responds best to.

Next, use the second 555 to gate the first. Set the second 555 for about 900us on and 100ms off, and you should get a sound wave that 's about 1 ft long.

If your circuit is working about the way mine did, you should be able to set the two transducers side by side and see a pulse followed by fainter echoes for objects that are 1-4 feet away. The wave will likely be wobbly and somewhat frustrating to look at.

If you are satisfied, go to the next step and add some gain. The op amps you were using are worth a gain of about 33 to 35 each. So, I'd add in U1, but change R3 and R5 to 330K each.

Now, if you watch the signal, it should be huge at U1 pin1. At this time, you should bring another scope probe over to the second 555 so that you can trigger the scope when the pulse goes out. Set the scope for about 10ms per division, and you should be able to make out wobbles from the floor (in the first division), a blob from the ceiling (in the second devision) and multiple blobs from walls etc...

If the sensor and transmitter aren't somehow focused, they'll tend to pick up things from all sorts of unexpected directions. It all just goes out and reflections come back. If you take it outside, things are easier to make out. If yours turns out like mine, you should be able to pick up people from about 40-50 feet out. The signals will be faint by then though.

This method of plodding along is how I handle most "new" technology. Once you've done it, you have a fair Idea what it takes, but the first time through, you need to take ownership of the technology. An old boss of mine had the expression "make mistakes fast." The idea was to get you hands on the technology as fast as possible, observe what it does, act on the observations, and when you screw up, do it over.

To that method, I add stride. Once something is working, and only once its working, you add a few miserly improvements. A few, not many, because each change takes time to iron out.

All that said, could you remove parts and cost. Oh yes! First and foremost, get rid of the extra transducer! You can drive it to transmit the burst, then "squelch" it stop it from ringing, and then use it as the receiver. The squelch device is an analog switch that shorts across the transducer for a short while to stop it from ringing after transmission. It's quite likely you could use the same chip to squelch it as you use to drive it. Refer to the CD4066, CD4051, CD4052, CD4053 for some starter ideas.

Next, I'd try to get rid of U8. the processor will much more likely than not have a PWM generator that come out to one of the pins. It can generate the "tone." It will be 0-5volts, so it will need something other than the CD4069 (IC4) to use as a driver.

Then again, you could do away with the microprocessor and displays go back to a more anceint technology. Fish finders used to use a spinning lamp. The finder would transmit as the lamp went past the zero distance point. Then, the lamp would flash whenever an echo came back. Thus you could see all the various causes of an echo at once. Hopefully, some of your flashes were fish.

 

[sophiecentaur]

I thoroughly endorse all that.
There's nothing more disheartening than a complicated bit of kit that you built and doesn't work. You have no idea where to start in order to get it going.

 

[engineerj2010]

after looking at it, i think it may be a bit hard for me!

Any ideas on an easier ultrasonic distance meter kit that i could build?

thanks

 

[skeptic2]

http://www.acroname.com/robotics/info/articles/sonar/sonar.html

http://archive.chipcenter.com/circuitcellar/july01/c0701rr1b.htm

http://www.robot-electronics.co.uk/htm/srf1.shtml

http://www.micromint.com/app_notes/ti01_data.htm

http://www.robotshop.ca/home/products/robot-parts/sensors/ultrasonic-range-finders/index.html

http://www.best-microcontroller-projects.com/pic-sonar.html

http://www.active-robots.com/products/sensors/sensors-devantech.shtml

 

[engineerj2010]

in the original circuit, where is the actual power supplies? i see all the +9 +5 and GND. But if i copied this circuit onto a schematic and printed a pcb, how would i connect them?

thanks

 

[engineerj2010]

also, in this circuit: http://www.robot-electronics.co.uk/htm/srf1.shtml

how would i connect several led 7 segs to show the result?