Hi There,

I'm quite new to the forum and new to electronics in general so I'd appreciate any help I can get.

My project in mind would receive ultrasound from a source at (5Mhz-10Mhz) using a piezoelectric transducer, amplify the signal, and send it back through the same transducer in phase with the reflected ultrasound waves in order to amplify the strength of the ultrasound echo.

I understand that using the same transducer to receive and transmit simultaneous would produce impossible feedback just like trying to use a microphone as a speaker at the same time. But I think I can curb around this issue by using the transducer in periods of "listening" and "sending" modes, and by generating a synthesized signal to amplify the reflected wave. The transducer can listen for periods of around 200us, calculate the frequency of the received wave, generate the same wave frequency and send it out for 200us before it goes into "listening mode" again.

I'm not sure if it is possible to achieve these functions using a fully analog design, so I'm considering using a microcontroller to calculate the frequency and generate the synthesized waveform before sending it out on an amplifier circuit. I'm processing ultrasound waves at 5-10 Mhz, so the processing speed would need to be relatively fast to make sure the ultrasound waves stay in phase. I'd like to know whether this approach would be possible, and any risks and issues I might run into. If anyone has suggestions on a different approach to amplify "echoes", so to speak, I'd love to hear from you.

Thanks!
Eric

 

Most ultrasound projects operate at around 40KHz don't they?

Don't think that you will find a piezoelectric transducer to do what you require. Not from much of a distance anyhow.

 

T.Jackson, yes, this project does operate on the higher end of the frequency spectrum. The operational range for this would only be up to 15cm so the depth of penetration is not so much of a major concern. 5MHz piezoelectric transducers are not as common as 40kHz, but hopefully I'll have some luck sourcing it.

Do you have any insight on whether a digital design is better than an analog design, or if this even seems feasible? If I were to use something more common like a 40kHz transducer as a proof of concept, would an ATMega328 at 16MHz suffice (thinking about using an arduino...)?

 

Bit unclear of what you're trying to do and why really. Usually when we engineer we first create a list of requirements in dot-point form.

 

I'm quite new... to electronics in general so I'd appreciate any help I can get.

I'd like to know whether this approach would be possible, and any risks and issues I might run into.

This project will be a huge uphill battle for a noob. It'd be quite challenging for anyone not already familiar with the required technologies. If you persist, I highly recommend going as far as you can with commercially available technology (maybe a used medical ultrasound device?) and reverse engineering / hacking only as needed.

 

Medical diagnostic ultrasound works in the 4MHz to 20MHz range, and does Tx/Rx on the same transducers as you describe. But, it takes a team of engineers quite a bit of time and bucks to develope such machines (most use synthetic aperture and beam steering techniques, along with color flow, 3D, or even 4D). There are many a pitfall along the way.

 

Medical diagnostic ultrasound works in the 4MHz to 20MHz range, and does Tx/Rx on the same transducers as you describe. But, it takes a team of engineers quite a bit of time and bucks to develope such machines (most use synthetic aperture and beam steering techniques, along with color flow, 3D, or even 4D). There are many a pitfall along the way.

Do you know how the Tx/Rx multiplexing is realised? I can imagine that this could be challenging, as it can be for other pulse echo systems like TDRs. I believe that in the latter case, a delay line may be added so that near echoes return at a clear period, when any hangover from the transmission has decayed away. Can similar things be done with ultrasound, or would the delay medium be too bulky or otherwise problematic?

 

The common way is diode biasing, a similar thing is done in radio when transmitting/receiveing on the same antenna.

Yes, delays are used to sync the return echo to the known Tx pulse because the propgation of delay in a given medium, such as human tissue, is a known factor. But, how to determine and calculate accurate delay times and process them in software and form an image is the subject of many a patent.

 

Hey @nomurphy, I'm working on a similar project and I was wondering if you had been able to make some progress.. Did you manage to get a signal?