Increasing sonar beam width

Hi, all I currently I am building a robot that is about a foot and a half wide and a foot tall, and I am using two Devantech SRF04 ultrasonic rangers.
The problem is that at a close distance the sensors cannot cover the whole front of the robot. Has anyone had this problem before, if so how did you fix it. I thought maybe I could connect a two more transducers on to the solder joints that connect the main ones. Would this work? Its just that I don't want to pay fore more circuitry if I don't need to, and I am running short short on IO Pins also. Thanks -Kit
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
kitmor wrote:

So long as you don't make the wires more than a meter long and you can supply enough current for all the transducers it will work fine. You may get slightly odd results, however because a received sound wave will reach one sensor before the other.
--
/-----------( Brendan Gillatt )----------\
<---( http://www.brendangillatt.co.uk )---->
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Brendan Gillatt wrote:

Hmmmm. I wonder if it would really work to parallel up a bunch of ultrasonic emitters -- or did I read the post wrong? I don't know what the SRF04 transducers use, but my guess is a piezoelectric element, which looks rather like a capacitor to the driver circuit. Unless changes are made to the circuit I am doubtful you could connect up two or more transducers and still get the desired output.
Also, what happens to the multiple wavefronts from the separate transducers? The receiver transducer is cut to accept a narrow band of frequencies back, and reject the rest -- helps reduce errors from Doppler effect, for one thing. Multiple emitters are bound to cause irregular reflections that will result in some wild readings, I am thinking.
There are some omni-directional piezo film ultrasonic transducers, but so far I haven't seen one of them used in a robotics application. What folks usually do is use multiple SRF04s (or their ilk), and fire them sequentially. Parallax has a deal on their SRF04 knockoff, the Ping, at 5 for $100. Or put the sensor(s) on a turret.
If anybody has successfuly paralleled ultrasonic transducers I'd like to read up on how well it worked.
-- Gordon
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Well I have some transducers that I can take off of a dead SRF05 detector. So I will try it, I will definite post the results. -Kit
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
kitmor wrote:

Another suggestion: place the transducers far apart. As the transducers are almost a point source (aperture is 8 mm, wavelength = 25 mm), there's no problem to separate them let's say 25 cm or even more, so just move one of the transducers. (of course the measured length should be recalibrated and isn't completely linear).
Stef Mientki
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Thanks for all the ideas. So I started to desolder the transducers but I found that it looks like the transducers are soldered on both side of the board, and on one side the joint is under the transducer. so I can't get it off. Is there a known way to remove them? Thanks again -Kit
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
kitmor wrote:

Use two soldering irons or a large tip to heat both joints at once, and/or use solder wick to remove all solder from the joint first.
But in response to your original query, you can use a reflector at the end of a box, like a periscope:
\ \ \ \ \ | | | | | | | | | | | | | || | | XD | |____|
where the slot at the open end of the box is as wide as the box is long (to preserve beam width), you'll have a fairly effective periscope. Mount it vertically with the open slot facing forwards. The slot must be at least 10mm across, since the wavelength at 40KHz is about 7mm.
Ultrasonic waves reflect off moderately hard surfaces like light off a mirror. Anything harder than cardboard would work well, but for testing, even a cardboard box would suffice.
Clifford Heath.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Clifford Heath wrote:

Nice idea, and because the shapes are small compared to he wavelength, I think a rectangle periscope will work equally well.
-----| | | | | | | | | | | | | | | | || | | XD | |____|
Stef Mientki
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Stef Mientki wrote:

You'll have trouble with end-to-end reflections if you do that. You might get further with sound absorbent material under the transducers, but you want a coherent wave heading *out* the slot (not directly back to the sender) and an angled mirror will be important in achieving that.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Clifford Heath wrote:

That's true if the mirror is much larger (say at least 5 times) the wavelength.
Stef
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Hi there, thought I would give you an update. So I finally got the transducers off with a flat head screwdriver and a very hot iron. So I got it connected up and it worked. though I do not know how, or if they will interfere with each other because I have not hooked them up pointing the same direction. But it worked, and I only killed one transducer (pulled one of it's legs off, and while I was putting it back on I destroyed it completely) Thanks for all the help and ideas -Kit
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Hay, Is there a difference between transmitter and receiver transducers? Also Can I hook them up backwards? I had to get one new transducer (of course they come in sets) but dose it matted which one I hook up where and how. I could not find out which pin is + and which is - from the data sheet. Here is the new ones I got https://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?langId=-1&storeId 001&catalogId001&productId9491 and here is the ones off the sensor http://www.robot-electronics.co.uk/datasheets/t400s16.pdf Thanks -Kit
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Kit wrote:

They can be, just as there are differences between mics and speakers. In non-OEM transducers I've bought, the elements are identified with a T and an R, obviously for transmitter and receiver. Looking through the fine mesh of the transducer I can also see differences in the internal construction.

I've seen them where the metal casing of the transducer is obviously part of one of the terminals. The other terminal is insulated. I would gather it's a good idea to make the insulated terminal +, and the other -, so it can serve as ground. Might help if you mount the transducer in a metal bracket (which I don't recommend anyway, because of the potential for HF ringing.)
If one of the terminals is not connected to the case, and there is no other marking or other polarity identification, then I guess it doesn't matter.
-- Gordon
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.