Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Ultrasonic Anemometer - signal recognition issues

Status
Not open for further replies.
Z

zenonmb

Newbie level 4
Joined
Apr 13, 2010
Messages
7
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,283
Activity points
1,342
Ultrasonic Anemometer

DSP experts,

I am designing (or trying...) an ultrasonic anemometer and need some help to identify the received signal .
Basically I meassure the TOF , so send a train of 8 40Khz pulses in one end and the received signal is compared against a threshold . The issue is that (because of noise) I have not a steady reading and the difference is one cycle (25 usec) .
The problem is that I need 2 secs precision, so I am in trouble .
Is there any easy way to detect received signal, aside fom the amplitude recognition ?
BTW I am using a PIC18F2550 and I can not sample fast enough for ADC ., so that's why I am using the comparator module inside the PIC .
Regards,

Zenon
 

Zenon,
While I don't have any answers for you, I have been thinking about attempting the same project. Have you made any headway with your design?

Regards,
irengineer
 

Irengineer,

I still working on it and seriously considering a cup anemometer......
I bought a parking sensor to get the transducers and used the analog part of it , mainly for driving the transducers and the switching part with the 4052 IC .
It is interesting to note that the parking circuit is very similar to this one .
**broken link removed**

Still with that I am getting noise and readings may change within 1 cycle, so it is really bad. It looks there is no way to avoid FFT o DFT and the problem is that I am not getting a fast enough pic in my country (Argentina) . PIC18f2550 has 100kps and for this app I need at least 500kbps (sampling rate) .
I am sorry I can not help..... at this time
Cheers,

zenon
 

The "one-cycle" uncertainty suggests, that you didn't yet manage to process the received signal in a suitable way. You are obviously using small bandwidth transducers, that can't give sharp received packets. You have to generate an envelope signal from the received waveform by rectifying and filtering it, before you apply it to a comparator. And you either need am AGC circuit or a constant fraction trigger to make the detection independant of amplitude variations, that must be expected.
 

FvM,

Regarding the trasducers not sure about your comment . These came with the parking sensor kit, so I am assuming should be good.
The circuit already has a rectifier and again this circuit was working as a parking aid.
My concern now is if my app (the anemometer) requires same precision as a parking sensor.
Cheers,

Zenon
 

Let's see pictures of the sensors and the schematic ( include the analog part from the pdf file as well )

Did you get a datasheet with the sensors?

If not - we might need you to have a DSO to do some sensor peaking.
 

These came with the parking sensor kit, so I am assuming should be good. The circuit already has a rectifier and again this circuit was working as a parking aid. My concern now is if my app (the anemometer) requires same precision as a parking sensor.
Click to expand...
Good for what? You're complaining about a one cycle (25 us) uncertainty. For a parking sensor, 25 us means 4 mm(!) distance resolution, which would be actually an unneeded precision. So a simple calculation can answer your question, I think.

Either what's the sensor's achievable resolution, the requirements for the signal processing are obviously very different. I think a usual small bandwidth 40 kHz transducer can be used for TOF measurements, but it's not easy. I would expect, that commercial ultrasonic anemometers have higher operation frequencies and wide band transducers.
 

FvM,

Thanks for clarification . I found that 25 usecs = 8.5 mm distance (for 340 m/s air speed)and you are right this irrelevant for the parking sensor .
So the wind meter that i am triying to implement is another "animal" . I need 2 usec precision

VsmGuy,

I bought an inexpensive sensor parking kit like this one, and does not bring any transducer spec .

https://www.google.com/products/cat...TMriCpjKM8m-5Z8B&sa=image&ved=0CBgQ8gIwATgA#p

Regarding the schematic, I do not have any one, but I will take pictures so you can see by your self . I have only traced the important signals in the PCB, like the pulse reception, and CD4052 address pins .
 

It's actually 4.2 mm for 25 us round trip delta.

2 us isn't generally impossible with 40 kHz transducers, but not easy. You have to use at least a more sophisticated envelope detector circuit. If I remember right, you have the problem that small disturbances, e.g. a small drop of water on a transducer, can cause a rather large delay skew.
 

Guys,

Attached is the picture for the parking aid . My camera does not have good resolution, but you can see at the top, the 4 transformers to drive the 4 transducers, aCD4052 on the left and NE5532 (8 pins) on the bottom , that handles the analog signal . Beside it you can see 2 diodes that rectifies the received signal.
I have removed the controller and soldered wires for the received signal and the address pins for the 4052 .

Cheers,

Zenon

Added after 1 minutes:

Forgot the attachment....
 

Hello guys ,

i try to built an ultrasonic anemometer project and i will need resources .I wonder if is possible to help me with this project
 

hello guys,
we have trouble with ultrasonic anemometer using pic , we need circuit of this anemometer and which sensor we should use
could you help us immediately :(
 

What is the problem with anemometer?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top