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.

How to drive a 1MHz Ultrasound Therapy Transducer

Status
Not open for further replies.

FoxyRick

Advanced Member level 4
Joined
Jan 8, 2004
Messages
1,249
Helped
321
Reputation
640
Reaction score
277
Trophy points
1,373
Location
England, UK
Activity points
12,979
1 mhz ultrasound

Hi,

I need some help driving a powerful(ish) piezo-electric ultrasound transducer. Here's the background...

I have an old ultrasound therapy machine that has stopped working (again). The circuit is quite complex, and is controlled by an MCU, and I think it would be better to build a completely new driver for the ultrasound transducer, rather than repair this thing again. It's the size of a tank!

The machine is intended for deep tissue massage - it's great, after a minute you can feel the muscles inside getting warm - it really relieves muscle aches and problems. Mine was intended for use on race horses - it worked well on my back though!

Anyway, the machine drives the transducer with these settings:

Frequency: 1MHz or 3MHz
Power: Pre-set selectable limits, up to 2W/cm²
Pulsed or CW

The power meter used to move (a lot) depending on what type of tissue was being massaged. I don't want to just drive the transducer with anything, as ultrasound can damage tissue if too powerful (don't want that!)

So, I need to know:

1. What sort of drive circuit do these transducers need? (what voltage/waveform/current)

2. How exactly do we measure the power it is giving? (is it just a case of voltage x current / area of contact)

3. Any advice on how to limit the power output?

*Any* advice would be useful, my back is aching and I need my ultrasound! I have asked the manufacturers, who are only interested in selling me a new one (£800) and googled with no useful results.

Thanks,
FoxyRick.
 

ultrasound transducers 1mhz

Hi,

I really don't know muich about these things....but you said 'any advice' :D I don't want to post something that isn't useful to you, but heres my 'two cents'....

Peizo transducers (which is I assume what you have) are 'similar' to speakers, in that they pconvert electrical energy to sound energy....but, unlike speakers, they are not inductive, they are capacitive. The smaller ones, used for little beeps and buzzes are usually driven directly from an MCU, which is a poor way of doing things, they are 'voltage' devices, and don't take that much current. However, for your '2W/cm²' that seems awfully powerful to me! How large is the transducer?

In all the 'larger' unltrasonic transducers I've seen (20Khz - 1Mhz) there is some form of inductor in series with it, driven by a power transistor, which is carefully controlled by a MCU (power/frequency/waveshape). Think the 'inductor/transformer' is used to match impedence, so no power is lost.

Unfortunately, I tihnk your best best would be to first repair your device, then 'probe' it with a scope/multimeter/frequency meter to find out exactly what shape signals are used, and measure voltage p-p, current, impedance etc... Then you can 'clone' it, and make a more reliable version.

It's the size of a tank!

Hmm, don't know why, I've seen similar things on google, same output power, same frequency, and they're the size of a pack of cigarettes. I guess, if its for horses, the transducer must be huge, meaning the '2W/cm²' becomes signifcant. Although, the larger the transducer, the lower the frequnecy it can work at. Is it a 'array' of transducers?

....as ultrasound can damage tissue if too powerful

Yep. But even some 'low level' sounds can do significant damage....depends on frequency, and absorbtion in the body. I read somewhere, a better way of doing this is to use HF radio waves, modulated at the desired frequency. Since flesh absorbs RF quite well, its a good way to get 'other' frequencies into the body, indirectly. At least, thats what I've read in a few medical journals.

Personally, if its driven by an MCU, I doubt very much we are talking about a simple 'sine' wave here. An analogue osc would be much better for that....even if its controlled by a MCU. So...I suspect we are dealing with waves that are a tad more complex, possibly modulated, many a harmonic set.

If you could outline what the main components are it would be helpful. I mean, you could tell a 'pwer driver' by large transistors, and 'control' as a bunch of digial chips, analogue as loads of, well, analogue! (caps, resistors, inductors). Any 'large' components should also be mentioned. I doubt very much that the MCU has gone bad, so you could still use that....its probably just the driver/PSU thats gone.

Hope this helps, always willing to give these things a whirl.

BuriedCode.
 

1 mhz osc circuit massage

Thanks for the response. I always meant to 'scope it while it worked, then wished I had when it stopped!

I've got it partly working now, turned out to be a bad solder joint on the output socket this time, nice simple fix. Last time is was the power regulator to the digital circuits.

The 1MHz is now OK, but the 3MHz oscillator section has a very erratic amplitude - possibly overheating something. Some of the transistors are too hot to touch after a few minutes running.

With regards to the overall design, The digital section (lots of TTL and an old ceramic MPU, it's covered in glue and labels so I can't see what it is) controls the treatment time, pulse-width modulation and power. I'm not sure yet if it sets the power simply by setting the oscillator/amplifier gain, or if it uses feedback to maintain the set level.

There are two sine-wave oscillators, 1MHz and 3MHz, with big (4cm cube) air-spaced variable capacitors on the PCB for tuning. There seem to be more transistors than I would have thought necessary (about 6 per section), that's why I suspect something like feedback power control. The oscillators look to have larger TO-220 preamplifier transistors which are then selected and switched by relay into a TO-3 class-A single transistor driver. There are four multipole relays routing signals around, controlled by the MPU, just to make things tricky to follow.

The main driver runs off a 110V DC supply, and you are right about there being a big coil (3cm diameter, 8cm long air-core, several taps going into relays) and more variable capacitors to match the output.

The output signal to the US head is a 60V peak-to-peak sine wave, when the power is set to 2W/cm² at 1MHz CW. Seems a little low, given the 110V power rails feeding the transistor. I might try to up the gain a little somewhere, after checking, replacing and heatsinking the overheating components.

The US head is aluminium, about 4cm diameter and 10cm long, with a big handle on the side. It joins to the machine by a slim coaxial cable. It has screw-on end-pieces, which appear to have the disc-shaped piezo element in them, about 2.5cm diameter and 2mm thick, with a contact arrangement that pushes the element against the main body of the head.

As to the 'tank' dimensions, my osteopath has an even larger one, looks 19-inch rack mount width and about 20cm high and 40cm deep!

I could really do with tracing out the circuit schematic from the PCB, but that's a job I just don't have time for at the moment. I also need to figure out how/where it measures the 'power'. Seems to come from another winding on the output coil. More investigation needed as it is ultimately the power measurement that I would need to build my own driver. Don't want to zap my cells!

The circuit gives me the impression of being way to complicated. I'm sure I could come up with something simpler, driven by a PIC.

Anyway, that's where I'm up to so far with it. Any circuit/driver/power measuring suggestions welcome!

FoxyRick.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top