Page 1 of 2

Fast Pressure sensors?

Posted: Sat Mar 14, 2009 7:33 pm
by rgbphil
Hi,

I've noticed the SCP1000 from sparkfun has a refresh rate of upto 9Hz.
I'm looking for a sensor with at least 32Hz update rate....any ideas?

Also has anyone used these sensors to measure boat speed?
Phil

Posted: Sun Mar 15, 2009 11:33 am
by The_Todd
I don't think that the SCP1000 would measure your boat velocity.

To measure velocity (airspeed) using air pressure you need a differential pressure sensor and a pitot tube. Using the differential air pressure measurement from the pitot tube and bernoulli's equation you can calculate the velocity. Its explained quite well here.

An example of similar equipment I've used in a wind tunnel would be a pitot tube and differential pressure sensor from omega. There are likely much more cost effective alternatives out there than omega equipment though.

-Todd

Posted: Sun Mar 15, 2009 1:14 pm
by TheDirty
Not certain how that would be useful, since it would be airspeed measured and relative to ambient wind, not actual speed across the water.

If you want faster, there are plenty of straight analog absolute pressure sensors on Digikey, Mouser, or any major distributor. Hook it up to an ADC and measure the pressure yourself.

EDIT: Ah, I see, the way it is setup it measures the difference from total and static pressure. It also says, it's not all the accurate below supersonic speeds? That's quite a calculation with temp and ambient pressure, though. Seems like it would be fairly difficult to get really accurate.

Posted: Sun Mar 15, 2009 6:04 pm
by The_Todd
TheDirty wrote:Not certain how that would be useful, since it would be airspeed measured and relative to ambient wind, not actual speed across the water.

If you want faster, there are plenty of straight analog absolute pressure sensors on Digikey, Mouser, or any major distributor. Hook it up to an ADC and measure the pressure yourself.

EDIT: Ah, I see, the way it is setup it measures the difference from total and static pressure. It also says, it's not all the accurate below supersonic speeds? That's quite a calculation with temp and ambient pressure, though. Seems like it would be fairly difficult to get really accurate.
It should be quite accurate assuming there is not a high wind blowing directly in either of the ports, this is the tradeoff. This method should be much more accurate than GPS velocity. There are other alternatives such as a Doppler Velocity Log (DVL) instrument to measure boat speed ($30k+) if you need mm/s accuracy.

Again, you cannot use a single pressure sensor to accurately measure velocity. The nasa site says "If the velocity is very high (supersonic), we've violated the assumptions of Bernoulli's equation and the measurement is wrong", so it is not accurate ABOVE supersonic speeds.

Pitot tubes can be very accurate when properly calibrated, that's why they are used in most fluid flow measurement systems, including wind tunnels, and every airplane (You will see multiple pitot tubes, sometimes temp controlled, on the front of every commercial airplane)

Posted: Sun Mar 15, 2009 6:56 pm
by rgbphil
OK thanks for the tips on pressure sensors not being all that practical for boat speed measurement.

Fortunately that was just one question....a nice to have.

The main question was a fast sampling pressure sensor thats easy to obtain in small quantities.

The purpose for this is to measure the pressure of a paddle blade or rowing oar (dragon boat, kyac, skulls etc) during a stroke. The idea being that a paddler or rower is performing better when there is a higher pressure when they are stroking. We already measure the bending force using strain gauges at the rollick/blade junction grip which gives some measure, I want to correlate this to blade pressure with an eye to improving the efficiency of the athletes technique.

A typical stroke for the above sports is 0.8 to 1.5 seconds, so a 9Hz sample of the blade pressure will give some useful data, however I want to get higher resolution data to line it up with the 32 to 128Hz data I'm already getting from the strain sensors.


Phil

Posted: Mon Mar 16, 2009 5:03 am
by TheDirty
Thanks for the explanation, Todd.

Like I mentioned then, a regular analog pressure sensor connected to a ADC of your own would work.

The forward force would depend on both the pressure and the area of the blade that was in the water. So you would need to know both to have an accurate indicator of how much work someone was doing.

Posted: Mon Mar 16, 2009 3:44 pm
by rgbphil
Yep it looks like I'll have to use a bridge type pressure sensor as you've mentioned and sample it myself with an ADC, not technically hard but a little inconvenient with the circuitry at hand. I was hoping for a temp compensated and calibrated device with all those issues sorted out for me already aka SDC1000-D01

We've already got force readings using strain sensors on the shaft, with forward force vector calculated in conjunction with an accelerometer to measure inclination. I suppose the purpose of the pressure sensor is to correlate this with actual blade pressure. I'm wondering if there is always a linear correlation between blade pressure and force (in which case blade pressure measurement is pointless), or whether some modification of technique can lead to higher blade pressure for the same force (which a coach could experiment with and optimise training for to increase boat speed for same performance level).

In any case blade pressure measurement would help optimise blade design (which would be a function of the shape/area), though blade designers would be a limited market.

Phil
PS: Commercial plug warning: We already have an second gen product on sale now for Dragon Boat paddles, and are close to production lining kyac paddles and rowing oars with instrumentation, so if anyone is into these sports and wants to buy product or invest in our company please PM me and I'll pass on details.

Measuring DB Stroke Rating

Posted: Tue May 26, 2009 9:43 am
by gmoseley
Forgive me for jumping in with a related, but not quite on topic, question. I am a paddler and coach with a dragon boat team in Boston. We are trying to gauge and control much more accurately our stroke rating, that is the number of paddle strokes per minute. Are you aware of any existing device for measuring this in real time - so that the rating could be viewed in the boat as it is paddling and the rating adjusted? Thanks.

Posted: Tue May 26, 2009 9:53 am
by leon_heller
An accelerometer could be used to detect the acceleration of the boat on each stroke.

Leon

Posted: Tue May 26, 2009 10:18 am
by gmoseley
Thanks for the reply, Leon. What we need to know is the number of strokes per minute in real time. Sort of like a tachometer in a car. Any thoughts?

Posted: Tue May 26, 2009 10:24 am
by leon_heller
Use a suitable MCU and display. It's a sizable project, of course.

Leon

Posted: Tue May 26, 2009 10:33 am
by gmoseley
Those sound like the basic components. The tricky bit is sensing and recording the individual strokes. Eventually, as the sport of dragon boat paddling continues to grow, a device like this will be developed and marketed.

Posted: Tue May 26, 2009 10:48 am
by leon_heller
Similar techniques are used in other sports such as golf and swimming.

Leon

Posted: Tue May 26, 2009 11:21 am
by Philba
Detailing out what leon said. The accelerometer will give you a waveform that shows acceleration peaks which coincide with the stokes. You will need to do some filtering but I'm pretty sure you can do it. Then you measure the time between two consecutive peaks and that is your stroke period. Divide your time base (minute?) by that and you have stroke rate in pretty close to real time.

I'd start out building up a simple logging device and make a couple of runs with it. Then look at the wave forms, play with various filters and come up with the peak detector. Then version 2 implements the above stoke stuff and tells you the current rate.

Didn't SFE have some sort of device that does this? I don't see it on their site but think this is what Nathan designed. Kind of pricey... http://www.in2rowing.com/product_info.p ... ucts_id=44

Posted: Tue May 26, 2009 11:40 am
by gmoseley
Thanks for the detailed suggestions, Philba. I am feeling inspired to start working on this.

The link to the In2 Trinity is VERY interesting. It is darned expensive - $900. I am tempted to get one just to take it apart and then create a much simplified, less expensive version.

I'll report back on what I come up with.

Thanks, all.