Friday, August 16, 2019

Exciting New Barometers for Navigators

Barometric pressure remains crucial data for navigation decisions based on the weather. An accurate measurement of the  absolute value of the pressure is needed to evaluate surface analyses maps and hence the associated forecasts. We also watch pressure trends for judging the timing of the forecasts. Indeed, consistent changes of just a few tenths of a millibar can be our first sign that the pressure is rising. This can mean we are getting too close to a stationary isobar, or the High is moving towards us, or the High is building. In any event we can get very early warnings from slight pressure changes that a potentially serious change in our wind is possible—providing we have a good barometer and a good way to watch it, meaning a way to store and display recent values.

Historically, we needed an expensive device to do both things: be accurate and offer a clean versatile display of past pressures, but that is all different now, thanks to Dracal Technologies in Quebec. They have just announced a new USB barometer that sells for about $50 (US) that puts out the pressure every second in two NMEA 0183 sentences (XDR and MDA). This data can be read by any navigation program, which in turn will do all the data storage and graphing for us.

No power and no programming required. Just plug it in a USB port and tell your navigation software where it is located. Below is a sample screenshot from OpenCPN followed by one from Expedition. Video links at the end illustrate the set ups.

A Dracal barometer input to a serial connection to OpenCPN, viewed in the Dashboard plugin (comes stock with program) and in the Plots plugin, which is a separate plugin download. The OpenCPN program and all its many plugins are free products for Mac or PC—some of the Mac plugins are not quite ready yet for the latest ver 5.0.  There is a history function in the Dashboard, but it does not have enough resolution to match the precision of this device, hence we need the Plots plugin.  Also, we need to use the outdated MDA sentence for Plots; it will not read the XDR.

A Dracal barometer being read in Expedition using a Number Box and also being displayed in the Expedition app called StripChart, which is a powerful tool for presenting any instrument data.

Calibration offsets

A beauty of the Dracal USB Barometer is its simplicity. There is no interaction with it at all; we count on any manipulation of the output to be done by third party software, in this case the navigation programs, OpenCPN and Expedition, or others.

This device uses a modern pressure sensor, and these have become very good over the years. A new one out of the box, assembled according to instructions, will generally read the correct pressure to within ± 1.5 mb.  If the maker has access to an accurate pressure standard, they can then set the device to a higher level of accuracy.  This also applies to the barometer sensors in our cell phones, which will likewise, out of the box, untouched be accurate within 1.5 mb.  Using our free Marine Barometer app you can then find an accurate reference pressure (explained in the app's help file) and then enter an offset to make the pressure accurate to within a couple tenths of a mb.

The same is true with the Dracal barometer. We can fairly expect that it can be off by a few tenths, which we can learn from watching it, and then we can make the needed corrections in the programs we use to display it.  In expedition, this is done very easily with a digital offset accessed from the main menu/calibrations/baro. In OpenCPN, the correction process takes a few steps. We use the plugin called NMEA Converter, where we modify the NMEA sentence itself, as shown in one of the video links below.

At Starpath we have a full barometer calibration station with pressure control and two NIST traceable pressure standards plumbed into the system as shown below.

We used this system to do a full range calibration of the Dracal barometer, and for the sample we have the results are shown below.

The blue circles mark the calibration points we can enter into Expedition.

This is a very good instrument, with pressure correction varying by just 0.2 mb over the full range of pressures.  Over the typical sea level pressure range of 990 to 1040 mb, the instrument correction is -0.4 mb (it reads slightly too high). Knowing that we can insert that correction into our nav program so we always have an accurate pressure value.  If the weather maps do not agree with our pressure, then the maps are wrong!

Expedition is unique in that we can add a full calibration curve and not just an offset at a single pressure.

I should mention that there is also NMEA 2000 device from another company, Yacht Devices, that works in a similar way, namely plug and play accurate baro sensor. This fine barometer has a lot of extra features accessible from a NMEA 2000 gateway and deserves an article of its own. We are doing first a review of NMEA 2000 (Jan 2020) and then we will have a new article on the YDBC-05.

A hearty thanks to Dracal Technologies and Yacht Devices!  We no longer have any excuse for not having accurate pressure at hand and in our nav programs to view at all times.

For a set of reference articles and links on use of barometers in marine navigation, see

Video illustrations

Display Barometer with Full Calibration Curve in Expedition

Display Barometer with an Offset in OpenCPN


Anonymous said...

I hope that other device you're testing is from Yacht Devices. They spec it as follows:
"Pressure data output resolution to a NMEA 2000 Version 3 network is 0.01 hPa (mbar), absolute measurement accuracy is ±1 hPa, relative measurement accuracy is ±0.12 hPa."

Not bad for a $100 device.

David Burch said...

Thanks for the note. Yes, that is right. I am new to NMEA 2000, so still learning the ropes. They also have a gateway that gives the user options on the output as well as allowing built in offsets. Very nice. We hope to get back to this project in about 2 weeks. Also have a challenge of getting the 12V power into our test chamber and getting pressure out, but we will figure that out. It is almost academic, in that they use the state of the art sensor in the device. Still have to do it though. We are the first ones to proclaim that you cannot just quote the specs of a sensor, but have to quote actual measured accuracy of the final (soldered) product. More on all this later.