This forum uses cookies
This forum makes use of cookies to store your login information if you are registered, and your last visit if you are not. Cookies are small text documents stored on your computer; the cookies set by this forum can only be used on this website and pose no security risk. Cookies on this forum also track the specific topics you have read and when you last read them. Please confirm whether you accept or reject these cookies being set.

A cookie will be stored in your browser regardless of choice to prevent you being asked this question again. You will be able to change your cookie settings at any time using the link in the footer.

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Wind Sensors
#1
Hi all,

I've been searching a lot for a wind sensor that I can use with OP. It appears that the Tindie decoder for the Peet Bros sensor isn't available anymore (although I'm not sure I want a Peet Bros sensor).

What is everyone else using? Anyone tried an ultrasonic sensor? Are sensors like those that come from Davis robust enough to be used on the masthead?
Reply
#2
The davis sensors do work on the masthead and seem reliable but very expensive ($129)

I have written an arduino sketch (part of pypilot github repository) that can read these. I also provide boards as well, with a bmp280 pressure sensor included, the output nmea0183. I have a version with and without LCD display:
https://pypilot.org/store/index.php?rt=p...ry&path=73

I have been 3d printing anemometers, but I am still in the experimental stage.
Reply
#3
Thanks Sean. Pretty much everything I've looked at is over $100, so $130, while a little spendy, isn't surprising. There are some that seem to be available from China that range from very cheap to very industrial - probably too heavy for a masthead, but a wind vane and anemometer from them is still around $99.

Thanks for offering the very reasonably priced interfaces. It's starting to look like Davis might be the way for me to go, although I'm lusting after one of those cool Calypso wireless ultrasonic sensors (only 499 euros!).
Reply
#4
So Sean...
Since you are 3D printing instruments anyway, have you considered trying to create an ultrasonic sensor?
Reply
#5
(2018-10-25, 12:46 AM)abarrow Wrote: Hi all,

I've been searching a lot for a wind sensor that I can use with OP. It appears that the Tindie decoder for the Peet Bros sensor isn't available anymore (although I'm not sure I want a Peet Bros sensor).

What is everyone else using? Anyone tried an ultrasonic sensor? Are sensors like those that come from Davis robust enough to be used on the masthead?

Hi abarrow,

have you seen this one?
https://www.segeln-forum.de/board1-rund-...ensor-diy/

It uses the Peet Bros principle, a rotating magnet, metal shield and two hall sensors. The plastic parts can be 3d printed.
He uses a ESP8266 for the processing and wlan to send the wind data to OP. I have got one of his prototypes for just
25 EUR.

One very clever idea is to reuse the existing cable for the anchor light. The white led will not light below seven volts.
If you put six volts to the anchor light it powers the wind sensor with 6-12V -> 3.3V regulator placed inside the anchor light 
housing. If you enable the anchor light the leds get 12 V and start lighting. The wind sensor keeps running.

I have built a windvane around four years ago. It uses a BLE112 bluetooth module from Bluegiga. I have connected a
four quadrant magnetic sensor that measures the angular position of a diametral magnet. The angle is sent via BLE
advertising packet to the RPi. This means there is no permanent connection needed and the power consumption is
very low. A CR2032 battery lasts three to four months at one reading per second. I am using node-red to receive the
data and convert it to a nmea string. It then goes via UDP to the OP port and can be displayed using the standard
dashboard for the wind vane.

https://www.neptun-forum.de/viewtopic.php?f=7&t=5301


Cheers,
Chris
Reply
#6
Very nice, Chris! Using the anchor light wiring is very creative. I know there has been some discussion on other boards about the relative merits of using reed switches or potentiometers for wind direction measurement.

Thanks so much for the pointer!

Andy
Reply
#7
(2018-10-25, 09:50 AM)BellaX Wrote: Hi abarrow,


It uses the Peet Bros principle, a rotating magnet, metal shield and two hall sensors. The plastic parts can be 3d printed.
He uses a ESP8266 for the processing and wlan to send the wind data to OP. I have got one of his prototypes for just
25 EUR.
I think the rotating magnet with shield concept is ok if you don't care about high resolution of wind direction. It is probably ok for a weather station to get 8 or 16 directions, but I am not sure it is good for making boat polars or steering autopilot to wind. Can anyone comment on this?

I'm not sure about a standard wireless link for wind data. wifi protocol can have significant lag, and this will affect the control loop in an autopilot. It's really impossible to control since it is in the linux kernel, and your VNC packets may slow down the weather measurement. It will work with degraded performance I imagine people will do it.

Otherwise, wired or some other type of wireless transmitter, or maybe even if you have a dedicated wifi receiver (ESP8266) just for the wind sensors it would be better.
Quote:One very clever idea is to reuse the existing cable for the anchor light. The white led will not light below seven volts.
If you put six volts to the anchor light it powers the wind sensor with 6-12V -> 3.3V regulator placed inside the anchor light 
housing. If you enable the anchor light the leds get 12 V and start lighting. The wind sensor keeps running.
Very nice idea.

Quote:I have built a windvane around four years ago. It uses a BLE112 bluetooth module from Bluegiga. I have connected a
four quadrant magnetic sensor that measures the angular position of a diametral magnet. The angle is sent via BLE
four quadrant? How accurate is wind direction in degrees?

For sailing and polars, we want resolution of 1-2 degrees although 0.5 or better is desired. The davis anemometer with potentiometer vane gives 0.1 degree resolution, but there is a 7 degree deadband which I adjust to face forward (since boat cannot sail directly into wind)
Reply
#8
I seem to recall reading about a guy who modified a Peet wind direction sensor by adding additional slips of metal inside to give it higher resolution. You have a very good point about needing higher resolution on a boat than on a home weather station, like what the potentiometer - based systems provide.

What about going with a magnet, as provided on the Peet sensor, but instead of a reed switch, hall effect sensors were used? I can understand how there would be some concern about the longevity of a potentiometer.

EDIT: Google and ye shall find! https://wunderweatherstation.wordpress.c...direction/

Okay, one last post on this, and I'll try to drag the discussion back to OpenPlotter!

More research turned up some nice, low friction rotary encoders that are good for 100 million revolutions, and in some cases include dual ball bearings. It seems to me that these could be used not only for wind direction sensing, but also speed. https://www.bourns.com/pdfs/EMS22P.pdf
Reply
#9
Hi,

the 4-quadrant sensor is an AS5050 from Austriamicrosystems. The resolution ist 10 bit.
If this is not good enough you can replace it by a AS5045 with 12 bit.

I am using an axis made of brass running in a sapphire. It is very sensitive. I was wondering
the windvane was continue moving without wind, it then behaves like a compass and shows
north! Smile

The DIY Peet Bros sensor counts the revolution time to determine the wind speed and direction.
The metal sheet is used to seperate the sensors only. The resolution decreases with rising
windspeed.

The overall resolution limit is the ESP Microcontroller. I think it can be improved by counting
every second or third revolution depending on the wind speed.

Chris
Reply
#10
(2018-11-02, 10:01 PM)BellaX Wrote: Hi,

the 4-quadrant sensor is an AS5050 from Austriamicrosystems. The resolution ist 10 bit.
If this is not good enough you can replace it by a AS5045 with 12 bit.
10 bit is sufficient for wind direction.
Quote:I am using an axis made of brass running in a sapphire. It is very sensitive. I was wondering
the windvane was continue moving without wind, it then behaves like a compass and shows
north! Smile
I think it's a problem if you use a magnetic sensor because the magnet turns the vane into a compass.
Quote:The DIY Peet Bros sensor counts the revolution time to determine the wind speed and direction.
The metal sheet is used to seperate the sensors only. The resolution decreases with rising
windspeed.
Unless the uC can't measure timing at the higher wind speeds, the resolution should increase with wind speed not decrease.

This design using a magnetic shield is interesting, but at low wind speed the wind direction is not going to be good, because the cups must be spinning to measure wind direction. In light airs, and especially the top of a moving mast, the apparent wind moves a huge amount, and these puffs and accelerations of wind will affect the wind direction measurement especially at low wind speed. Maybe it can be improved using 4 or 8 reed switches instead of only two, but not sure if I like this idea.

It's probably a good design for a fixed weather station that doesn't need to measure small changes in turbulent wind direction, but for a sailboat following the wind in very light wind, this is important.

I need to do a comparison to be sure, and I am still curious about this concept as I have not used it.. I am only very skeptical.


Quote:The overall resolution limit is the ESP Microcontroller. I think it can be improved by counting
every second or third revolution depending on the wind speed.

Chris

How is the ESP microcontroller limited? At 8mhz or so, it can count down to 1/8th of a microsecond. This is way more resolution than needed even in 60 knots of wind.
Reply


Forum Jump:


Users browsing this thread: 3 Guest(s)