Sail pressure measurement

[John Perry]

Measuring the distribution of pressure over the windward and leeward surfaces of sails is one of many potential projects I have had in mind although I havent yet been able to make practical progress due to time spent going sailing/rowing and carrying out routine boat maintenance etc.

A basic difficulty in measuring the pressure on a sail is that the pressures that need to be measured are quite low - too low to be adequately resolved by most industrial pressure sensors. Pressure measurement on aerofoils in low speed wind tunnels was traditionally carried out using liquid manometers, the sensitivity of which can be improved by incling the tube so that it is near horizontal and/or by measuring the position of the meniscus between two immisible fluids rather than between one fluid and air. Liquid manometers when set up in an on-shore laboratory can be sensitive and accurate but I can see real difficulties in using a sensitive liquid manometer on board a sailing boat!

At one of the recent AYRS Thorpe meetings I made a short presentation of an idea for an electronic sensor with a flap of shim metal, say about 0.05 to 0.1mm thick stainless steel, that is deflected by an air pressure differential, the deflection being measured by strain gauges mounted on a narrow ''neck' about which the flap hinges. I did some finite element analysis that showed that it should be possible to measure strains in the region of 1Mpa for a pressure differential of 1Pa - futher optimisation of the geometry might even improve on this. I had it in mind to make up an instrumented hollow (i.e. tubular) sail batten that could be fitted into a fully battened sail with sensors measuring the pressure differential between the windward and leeward sides of the sail and the space inside the tubular batten, the pressure within this space acting as a reference pressure which, if the batten is open at the aft end, would be essentially the wake pressure, so I think similar to free stream pressure. The data from all the sensors along the batten could be monitored by a microcontroller in a pouch sewn onto the sail, then transmited by wireless data transfer to a computer below decks. The idea seemed quite promising but so far I havent made practical progress, for reasons given above.

Meanwhile, I wondered about commercially available pressure sensors for low pressures, these are generally piezo resitive sensors. I happen to have such a sensor, an MPL 3115A2, built into a small 'breakout board' from Adafruit Inc. I think one of the main applications for such devices is for use as an altimeter to stabilise the height of drones. I have had this device lying around for some time and this afternoon I got round to hooking it up to an Arduino and logging the readings from it. I would assume that sail pressure measurements would be carried out in fairly light winds, at least to begin with, so the requirement would be to measure pressures up to a few tens of Pascals with a resolution of 1Pa or better. I have to say that from my quick trial I dont think the MPL 3115A2 is good enough for this. I attach a graph (see end of this post) straight from the 'Serial Plotter' facility in the Arduino IDE, this graph showing the readings taken over a few minutes, each reading being relative to a mean pressure determined from a series of initial readings. For part of the graph I blew cool air from a hairdryer (with the heating element off) onto the sensor to simulate a gentle breeze. The sensor did respond to this but the response was hardly more than the variation and drift that occurred between readings with the sensor sitting in still air on a desk indoors.

A couple of weeks ago I happened to meet AYRS member Richard Fish who is also interested in measuring pressure distribution on sails. I wonder if anyone else reading this has such an interest, it would be interesting to compare notes and maybe do something collaborative. I think Richard said that he was thinking about using Bosch piezo resitive sensors, so I wonder how well these will work. Adafruit make breakout boards for various types of pressure sensor including Bosch ones and I summarise some of the manufacturers specs below. It is a bit difficult to make comparisons since different manufacturers specify accuracy and sensitivity in different ways, probably the best way to compare is to get samples (they are only a few pounds each from Pimeroni) and hook them up to an Arduino. On the face of it the DPS310 looks worth a try.

DPS310 (Infineon Technologies)
Sensor precision 0.2Pa
Relative accuracy +/- 6Pa
Pressure temperature sensitivity 0.5Pa/K

BMP 388 (Bosch)
Absolute accuracy +/- 40 Pa
Relative accuracy +/-8 Pa
Noise 0.03 Pa
Long term stability +/- 33 Pa

BMP 380 (Bosch)
Absolute accuracy +/- 50 Pa
Relative accuracy +/- 6 Pa
Noise 0.03 Pa
Long term stability +/- 33 Pa

BMP 280 (Bosch)
Absolute accuracy +/- 100 Pa
Relative accuracy +/- 12 Pa
Resolution of data 1 Pa

MPL3115A2 (NXP Semiconductors)
Absolute accuracy +/- 400 Pa
Pressure resolution 1.5 Pa
Long term drift +/- 100 Pa

In addition to the above sensors, long time AYRS member Graham Ward recently sent me details of another piezo resistive sensor - see here -
https://s3-eu-west-1.amazonaws.com/cdn. ... +28.06.pdf
These sensors have actually been used for aerofoil pressure measurement but I think for aircraft work where the pressure differentials are higher than for sailing boats. I have emailed this company for more info.

John

20-02-11 MPL 3115A2 Pressure readings.png

(89.84 KiB) Not downloaded yet

20-02-11 MPL 3115A2 Pressure readings.png

(89.84 KiB) Not downloaded yet

[AlexQ23]

I don't know if it can help, but Dwyer makes some Magnehilic differential sensors very accurate rated as low as 10 mm of water. One tube can be set in the extrados for example and the other one inside the mast. Hope you the best!

[John Perry]

Thanks Alex - I have taken a look at the website for the Dwyer gauges. The most sensitive ones have full scale deflection of 100Pa (about 10mm wg) which is the stagnation pressure for an air velocity of 12.9 m/s - I still think of wind as a Beaufort scale, this is Force 6. So that is about the sensitivity that would be needed, we dont really want to be doing sail pressure experiments with winds above Force 6. However, using this type of instrument to map a pressure distribution over a whole sail, or even along the length of a single sail batten would be difficult and expensive I think. Perhaps lots of gauges set up adjacent to each other and photograph them all to get simultaneous readings - I think that is how it used to be done with wind tunnels.

I don't think this is an easy project however you go about it. I read somewhere on the internet that North sails are trying to do it, or maybe thinking about trying to do it. Its fairly low priority for me - I can think of yacht research projects that might get interesting results more easily. In the meantime wool tufts or similar on sails can tell a lot about airflow.

The pressure on sails is really quite low - much less than day to day variations in barometric pressure. Thinking a bit further, I suppose this is only to be expected - it is the barometric pressure variations that drive the winds accross the oceans and when we place a sail to locally deviate those winds we are recovering a small part of the pressure difference that caused the wind in the first place.

[AlexQ23]

Hi John,
I'm probably not aware as you are but it's exactly what's I imagine. Atmospheric pressure are measured in mm of Hg and 1 mm of Hg is equal to 13,6 mm of water. So I tough that this gauge will do the job. But actually you still have to connect it with the sail, and that's an other matter. I think that in the '70 or so they perform such tests on yacht Papoose but accuracy on result where doubtful.
Just an idea... Will it be possible to use a kind of tell-tales that are connected true the sail on both intra and extrados. With a certain length, thickness and width they will have a certain mass with a well known CG. By using a picture you can surely be able to measure the angle they make and extrapolate the pressure?

[John Perry]

Hi Alex - I am not quite sure what you mean by 'tell tales connected through the sail' perhaps you have in mind somthing like a hinged flap mounted in a small hole through the sail so that the flap moves when there is a pressure differential between one side of the sail and the other? If that is what you mean I can see that you might be able to have lots of them spread accross the sail, then take a photograph from which you could work out the angle of each flap to get a pressure distribution. This is actually quite similar to an idea that I presented at one of the recent AYRS meetngs at Thorpe, Surrey UK. although I was thinking of measuing the deflection of the flap elecronically with strain gauges rather than by photography.

I attach below the results of some preliminary Finite Element Analysis that I did to get an idea of the sensitivity that would be possible measuring a pressure differential in this way. This may be a bit hard to understand since it is not a proper article, just some pictures and notes. The pictures at the bottom of the page show a hollow full length sail batten which would be instrumented with sensors on each side along its length. The aft end of the batten is open so the space inside the batten is at the 'wake' pressure behind the sail, this acts as a reference pressure. Each sensor consists of a flap of metal shim stock (i.e. thin metal sheet, maybe 0.025 to 0.1 mm thickness stainless steel) sandwiched between metal rings with a narrow 'neck' holding the flap and acting as a flexure hinge. There is only a small gap between the edge of the flap and the supporting ring so 'leakage' through the sensor should be small. Strain gauges mounted on the neck supporting the flap would sense the pressure differential accross the flap.

I suppose I could write this up as an article for Catalyst but I dont really want to do that without having got a bit further with it, I think the next stage could be to look into getting some shim stock cut to shape, possibly by laser cutting. However, I do have lots of other projects I want to get on with so not sure when I might get round to this again.

[John Perry]

18-12-16 Pressure transducer.pdf

(298.11 KiB) Downloaded 1353 times

18-12-16 Pressure transducer.pdf

(298.11 KiB) Downloaded 1353 times

I realise that I forgot to add the attachment to my previous email on this topic, so that attachment is added to this message - see link to .pdf file above. The coloured pictures are FEA results showing the predicted stress in the circular shim - the six holes arround the periphery would allow the shim to be clamped between two metal rings. There is a fine cut in the shim, possibly produced by laser, this creates a flap that will deflect when there is a pressure difference accross the sensor. It will also deflect under gravity when the sensor is tilted - it would be necessary to make a correction for that. The two grey pictures at the end give an idea how this king of sensor might be incorporated into a hollow full length sail batten.

Since my last message on this subject I have given some thought to using piezoelectric film rather than resistive strain gauges to measure the deflection.

Sugestions welcome.