advice on cheap pressure sensors

[Ben]

Hi all,

I'd much appreciate any advice on my search for a cheap and simple pressure
sensor. I need to make a device which can measure how hard a baby is
squeezing a silicon tube, around 2cm thick. I figure the best way is to
connect the tube via a pneumatic tube to a pressure sensor. So what I need
is a pressure sensor that does not have to be accurate in absolute terms,
but needs to be relatively sensitive, so it can measure changes in grip
strength. I reckon if it went up to 1000 kPa that would be more than
enough. Preferably it would come with an integrated USB interface and
drivers which provided an API. Alternatively, if you know a good one but it
just outputs a voltage, perhaps you could also recommend a good cheap and
cheerful USB A to D converter?

So far the best thing I've found is this, which you can combine with a USB
interface from the same supplier:
http://store.pasco.com/pascostore/showdetl.cfm?&DID=9&Product_ID=51374
&Detail=1
But I think it will only interface using their proprietary software - it's
for classroom use, after all. But this gives you a basic idea of what I'm
after.

Other solutions to the basic problem also gratefully received!

Cheers,

Ben

 

[Tim Wescott]

Hi all,

I'd much appreciate any advice on my search for a cheap and simple pressure
sensor. I need to make a device which can measure how hard a baby is
squeezing a silicon tube, around 2cm thick. I figure the best way is to
connect the tube via a pneumatic tube to a pressure sensor. So what I need
is a pressure sensor that does not have to be accurate in absolute terms,
but needs to be relatively sensitive, so it can measure changes in grip
strength. I reckon if it went up to 1000 kPa that would be more than
enough. Preferably it would come with an integrated USB interface and
drivers which provided an API. Alternatively, if you know a good one but it
just outputs a voltage, perhaps you could also recommend a good cheap and
cheerful USB A to D converter?

So far the best thing I've found is this, which you can combine with a USB
interface from the same supplier:
http://store.pasco.com/pascostore/showdetl.cfm?&DID=9&Product_ID=51374
&Detail=1
But I think it will only interface using their proprietary software - it's
for classroom use, after all. But this gives you a basic idea of what I'm
after.

Other solutions to the basic problem also gratefully received!

Cheers,

Ben

That's an absolute pressure sensor; you want a differential one with the
atmosphere as the reference pressure.

Motorola used to make pressure sensors as a board-level part. I assume
these were spun off into Freescale although they may be On semiconductor
parts now. You'd have to build a little board to interface to an ADC.

Or try these folks:
http://www.phidgets.com/modules.php...ns&file=index&req=viewarticle&artid=22&page=1

 

[[email protected]]

strength. I reckon if it went up to 1000 kPa that would be more than
enough. Preferably it would come with an integrated USB interface and
drivers which provided an API. Alternatively, if you know a good one but it
just outputs a voltage, perhaps you could also recommend a good cheap and
cheerful USB A to D converter?

Freescale's parts are good, I'm sampling them now. They have a version
with integrated hose barb, very convenient. The ADC-1R2 from
Superlogics is a serial-connected ADC which is very suitable for
reading this type of device; I'm not sure if Superlogics has a USB
version, but the ADC-1R2 is easily connected via USB-serial converter.

 

[xarvia]

Hi,

Freescale have a good variety of pressure sensors, they are about
$10-$15US from digi key (www.digikey.com). Digi key also has a category
for pressure sensors with 900 products and a decent parametric search
function:

http://www.digikey.com/scripts/DkSearch/dksus.dll?Criteria?Ref=282967&Site=US&Cat=33489888

As for connecting to a computer, personally I use a Microchip PIC with
RS232, if you want USB and do decide to use a microcontroller for A/D
conversion I'd still use RS232 and buy an off the shelf RS232-USB
converter. Playing with USB in microcontrollers seems more bother than
it's worth.

Good luck!

 

[RESO / Claude GUTH]

You may use my pressure sensor.

Download OA_Pressure_Probe at http://sourceforge.net/projects/openavionics/

 

[Ben]

Hi all,

thanks very much for all your advice. Those Phidgets look perfect for
me... I'm not averse to getting my soldering iron out once in a while
but it looks like I could avoid it completely with those and save a bit
of time. They are dead cheap too.

Cheers,

Ben

 

[[email protected]]

Dr. Ben, an inexpensive solution might be to salvage the pressure
sensor and electronic from an inexpensive, broken, blood pressure
monitor. (Usually the compressor on these little boxes fails well
before its other components.)

Off-the-shelf solutions are available, but tend to be costly.

Harry C.

 

[Frank S.]

Put a piece of black anti static foam (like the kind you put IC's in)
between 2 pieces of metal (I used PCB material)
That is the cheapest way I know about.

Frank

 

[notmyrealname]

Hi all,

I'd much appreciate any advice on my search for a cheap and simple pressure
sensor. I need to make a device which can measure how hard a baby is
squeezing a silicon tube, around 2cm thick. I figure the best way is to
connect the tube via a pneumatic tube to a pressure sensor. So what I need
is a pressure sensor that does not have to be accurate in absolute terms,
but needs to be relatively sensitive, so it can measure changes in grip
strength. I reckon if it went up to 1000 kPa that would be more than
enough. Preferably it would come with an integrated USB interface and
drivers which provided an API. Alternatively, if you know a good one but it
just outputs a voltage, perhaps you could also recommend a good cheap and
cheerful USB A to D converter?

So far the best thing I've found is this, which you can combine with a USB
interface from the same supplier:
http://store.pasco.com/pascostore/showdetl.cfm?&DID=9&Product_ID=51374
&Detail=1
But I think it will only interface using their proprietary software - it's
for classroom use, after all. But this gives you a basic idea of what I'm
after.

Other solutions to the basic problem also gratefully received!

The answer really depends a lot on volume, financial requirements, and
exact specifications. If you are are looking for a one-off system and
willing to spend the money, you can buy an integrated pressure sensor
(Motorola makes some for instance) with voltage output and a data
aquisition board (www.windaq.com or National Instruments). This solution
would be rather expensive (my guess is on the order of $150 or more) and
it may not be terribly accurate, but it would get you up and running fast
and may have adequate accuracy for your purposes.

There are a few things you didn't mention in your post in terms of
requirements:

1) How much resolution do you need? (I presume that "accuracy" must be to
at least +/- 1/2 of the resolution)
2) How much full-scale range do you need?
3) How many samples per second do you need? (presumably this would be on
the order of 1 per ms, and not a major requirement)
4) You mention that you do not care about absolute values, so it sounds
like you don't care about an offset in the result. But do you care about
an offset that changes over long periods of time or over temperature?
5) Do you care about absolute gain (i.e. Does a specific pressure change
have to correlate to a specific output change)?
6) If you do not need an absolute gain, do you need to ensure that the
gain doesn't change over long periods of time or over temperature?
7) Can you calibrate against a known accurate sensor either at production
or at any time in the future?

Accuracy is going to be a function of three things: the pressure sensor,
the A/D converter, and the signal processing between the two. The easiest
solution I can think of for the A/D conversion and computer interface that
doesn't include buying a premade board is to buy an 8-bit microcontroller
that has a built in USB interface and A/D converter. However, make sure
that the microcontroller's A/D converter is adequate for your required
task, otherwise you'll have to buy an external one (if you are trying to
get very accurate/high resolution #'s at relatively slow speed (< 10k
samples a second) you may be best off with an external sigma delta A/D,
but it may be overkill).

While integrated pressure sensors are very easy to use, it may be cheaper
and more accurate to get a dedicated current bridge style sensor. This
will require more sophisticated signal processing, possibly aided by
digital calibration in the microcontroller. I only recommend going this
route if you are forced to by the specifications of the project or if you
are going for very high volume (and it actually saves money).

The signal processing could range from virtually nothing to a carefully
designed current source, filters, and precision operational amplifier
setup, and again will vary completely on the requirements of your project.

good luck,
dan

 

[Roger Dewhurst]

Hi all,

I'd much appreciate any advice on my search for a cheap and simple pressure
sensor. I need to make a device which can measure how hard a baby is
squeezing a silicon tube, around 2cm thick. I figure the best way is to
connect the tube via a pneumatic tube to a pressure sensor. So what I need
is a pressure sensor that does not have to be accurate in absolute terms,
but needs to be relatively sensitive, so it can measure changes in grip
strength. I reckon if it went up to 1000 kPa that would be more than
enough. Preferably it would come with an integrated USB interface and
drivers which provided an API. Alternatively, if you know a good one but it
just outputs a voltage, perhaps you could also recommend a good cheap and
cheerful USB A to D converter?

So far the best thing I've found is this, which you can combine with a USB
interface from the same supplier:
http://store.pasco.com/pascostore/showdetl.cfm?&DID=9&Product_ID=51374
&Detail=1
But I think it will only interface using their proprietary software - it's
for classroom use, after all. But this gives you a basic idea of what I'm
after.

Other solutions to the basic problem also gratefully received!

Cheers,

Ben

--
Dr. Ben Kenward
Department of Psychology
Uppsala University, Sweden
http://www.benkenward.com

What about a salt water filled manometer? The salt water forms one plate of
a capacitor. The other plate is metal foil surrounding the plastic tube
manometer. Pressure causes the salt water to rise in the tube increasing
the capacitance. Put that capacitance into a square wave generating
circuit, 555 for example, and then measure the changes in output frequency.

R

 

[Steve]

$75 gas pressure sensor http://www.vernier.com/probes/gps-bta.html

with $59 GoLink! USB acquisitin unit.

http://www.vernier.com/go/golink.html

 

[Don Foreman]

Hi all,

I'd much appreciate any advice on my search for a cheap and simple pressure
sensor. I need to make a device which can measure how hard a baby is
squeezing a silicon tube, around 2cm thick. I figure the best way is to
connect the tube via a pneumatic tube to a pressure sensor. So what I need
is a pressure sensor that does not have to be accurate in absolute terms,
but needs to be relatively sensitive, so it can measure changes in grip
strength. I reckon if it went up to 1000 kPa that would be more than
enough.

1000 KPa? These are baby gorillas? Bionic babies?

 

[Paul Hovnanian P.E.]

Hi all,

I'd much appreciate any advice on my search for a cheap and simple pressure
sensor. I need to make a device which can measure how hard a baby is
squeezing a silicon tube, around 2cm thick. I figure the best way is to
connect the tube via a pneumatic tube to a pressure sensor. So what I need
is a pressure sensor that does not have to be accurate in absolute terms,
but needs to be relatively sensitive, so it can measure changes in grip
strength. I reckon if it went up to 1000 kPa that would be more than
enough. Preferably it would come with an integrated USB interface and
drivers which provided an API. Alternatively, if you know a good one but it
just outputs a voltage, perhaps you could also recommend a good cheap and
cheerful USB A to D converter?

Are you measuring pressure or force? Pressure will depend on the surface
area acted upon which will depend on the child's grip.

Force could be easily measured with a strain gauge attached to a
suitably designed spring. Something like a metal rod bent into a long,
narrow U shape, 2 cm across. If you can't control the direction that the
baby will squeeze, place two such springs at right angles inside the
tube, measure the strain in each axis and sum their vector quantities in
software.