Smart Instruments - DIY Open-Source

Background
-----------

Still a blue sky project at the moment - I have long wanted better instrumentation.

Many of the older diesels I've had to work with have woefully inadequate instrumentation - the worst had a Tacho, plus a blue and red light - blue light was charge (normally) went off when motor started.

Red was a simple oil pressure switch and by the time it came on there were usually serious issues - the donald duck light (rhyming slang);-

So whats with the engine?
Its Donald Ducked!
Completely Ducked?
Yeah - It Disney work any more!

I've discussed this with a friend over the years and we did find some crazy expensive systems which monitored everything down to the height setting of the captains chair and the temperature of his left butt cheek...

 

Is it now possible/practical?

Anyway I recently got an email with this link

http://blog.arduino.cc/2012/05/11/inexpensive-6-channels-temperature-scanner/

with a cryptic comment about;-
Arduino
Can get Ethernet!
PoE (=Power over Ethernet)
Use Raspberry PI touchscreen on board plus remote tablet
"So what are we going to do tonight? the same thing we do every night, Pinky—try to take over the world!"

I think it might be the time when this is actually a sensible project.


The Arduino unit is a CHEAP prototyping computer/real world interface but has a myriad of plug in modules including as alluded to above Ethernet and PoE Ethernet.

BUT there are so many more - NMEA, GPS, AIS, WIFI and others might be of use here.

It can also switch and control things - there is a mention of one being used as to develop an autopilot on BDN. as well as an open source Chart Plotter called FreeBoard.

 

Here is an overview of what I am proposing;-

A self contained 'layer' not tied in to other systems -
12v Battery and Solar panel to charge,
Ethernet Hub with 12v PoE
Wired in Cat 5 cable - hub and spoke topology for easy fault finding.
Dashboard will be a PC/Laptop/Toughbook or something called a Raspberry PI maybe with a touch screen.

(On smaller boats Powered USB might be a better option - WiFi / Bluetooth ditto on wood/fiberglass, if you can get power)

Several Arduinos throughout the boat monitoring engines,tanks, and Whatever.


This opens up 2 key possibilities;-

Smart Instrumentation
---------------------
Some companies offer what are termed smart gauges - often it is just a simple display that can perform the job of several gauges amps, temp, fuel etc.
To be truly smart a gauge must have a set of references and logic sufficient to determine if it is within 'normal' tolerances, or IS THERE A PROBLEM.

The gauge would then change colour and become highlighted on the captains console, even an alarm could sound.

The arduino also has room for an SD card so a Data Log could also be available for interogation.

This becomes a prime concern as the number of gauges increase - your smart dash is much better to display just a green 'tick' than 46 virtual guages leaving you to pick out if they are normal or not.


Monitoring and Logging of other systems.
---------------------------------------
because of the low cost and versatility some vessel parameters can now be logged which have never been considered as being viable, including cross referencing several factors at once.

 

Smart Bilge Pump Monitoring

Most of us just have an indicator light for the bilge.

With a smart system there are more possibilities.

If we add a flow switch (below) next to the output through-hull, we can have a positive check of flow. (you know there'll be these going on the raw water inlet too)

The Arduino module can be set to log;-

• when the float switch comes on,
• Check for flow,
• maybe log the voltage on the pump (will it be different if the pump is OK, vs if its blocked intake or jammed impellor)
• there are some cheap level sensors so we could duplicate the float switch
• high level emergency sensor.
• battery voltage?
• heck we can even connect up a rain gauge if we wanted - does our bilge flood from the propellor gland or leaking hatches?
• finally we can add a manual over-ride so the arduino can turn on the pump itself at the command of a remote system.

So the master PC could get a minute by minute 'Alls Well' or an error -
if your marina has WiFi you might get a text at home
You can also install a mobile phone sim if your mooring is a little further away.

It may say that there's water in the bilge, the switch works but the pump is not pumping and there's no water leaving and level is not going down - but you could check the gauges and see it's not going up, and it's stopped raining and maybe decide it will wait till the morning - maybe try to manually flick the bilge power on and off a few times see if the blockage clears - all from e.g. 20 miles away.


I think this is a good illustration of what I mean by smart instrumentation - of course the Engine is whole different animal under this system.

So what would everyone else monitor and how could that be made 'Smart'?
:?:

 

First point of departure is to list all the devices you can find with ethernet interfaces and how they send data.

Second, all serial devices as these can be made into ethernet devices with single port terminal servers. Analog via an A to D converter to ethernet. Etc.

Third, how much time & effort you want to spend.

Fourth, what's the power budget?

I've done all this in another life but with full power computers and effectively unlimited 240V AC power. Toy boat scale has another set of challenges (I regard my 12m sailboat as a toy boat, FWIW).

Short form - most or all of it has been done before, in many places by many people. The software is dead easy once you've got the hardware interfaces and data streams documented.

PDW

 

Thanks for your input,

That's how it's starting to look to me - much less like trailblazing, than a series of overlapping small projects, each of which can be added to the backbone as it grows. Provided the foundation is solid, there is room to grow.


I understand that others may choose to go the CAN-bus or NMEA route. My background is in Network admin and as such I have the skillset and tools for TCP/IP and Cat-5 cabling (as well as a few reels ends of some industrial spec shielded cable - I didn't know what I was going to use it for at the time but it seemed silly to leave it sitting there in the skip/dumpster).

You know the expression - When you have a Hammer, every problem looks like a Nail.

The Arduino with it's programmable Analog and Digital inputs and outputs is the ideal vehicle to bridge "Off the Shelf" sensors and switches, and talk to a central computer with meaningful data values.

Unexpectedly, the part I am finding hard, is not finding a way to do each particular thing, on the contrary there seem to be at least a dozen options for almost everything and the 'chore' is narrowing the choice down.
I think your idea on checking the power load is going to be a help to choose the best option, though robust and water resistant are still the top criteria.

Been looking at the FreeBoard project too - an opensource GPS chartplotter - very interesting, afterall a good navionics package shouldn't cost as much as the boat.

http://www.42.co.nz/freeboard/

 

following

 

I'm playing with a Raspberry Pi ATM for another project but there are some overlaps.

Reason - it can have a linux operating system, GCC compiler, Java runtime engine (I mainly program in Java these days), 4 USB ports, ethernet port, HDMI port and another port or 2 I think but I don't care about. Low power consumption, low price, easily made wifi via a USB device. I have some of my Java code running on it with some issues coming from the Swing GUI code - all of which I can code around with minimal effort.

Then there are 40 i/o pins that are beyond my technical competence to utilise as I'm only a dumb programmer, I leave hardware to others.

The Raspberry Pi gives me remote access via ssh and VNC so I could easily control it or have data displayed on a cheap daylight visible tablet.

I believe OpenCPN has been proven to work on a Pi.

None of this is difficult or even close to it. Give me a set of instruments with an ethernet or serial interface and I can have a datalogging system running in days or less. The GUI display takes the time to get functional and pretty but there are no real technical challenges.

A couple of us are leaning this way for our own personal systems but I'm probably a year off needing one on my boat, alas, and as electronic stuff changes so quickly, I'm not paying it all a lot of attention as yet.

The power budget still concerns me a lot.

PDW

 

http://opencpn.org/ocpn/


Have just had a look there and the AIS implementation is pretty amazing. I bet there are guys out there who are buying used 'blue chip' navionics which cost more and offer less functionality than implementing OpenCPN on new hardware.

Thanks for the heads-up.

 

From the thread measuring-pressure-diffferences;-

http://www.boatdesign.net/forums/on...ls/measuring-pressure-diffferences-47511.html

I was looking at a similar measurement from a slightly different perspective.

We tend to measure engine RPM as an analog for speed but a more telling force is (as RWatson wants to measure, above) Thrust.

I was considering adding an additional bearing on the shaft (inside the boat) with a couple of load cells on the thrust side of the stationary part.

This gives a couple of very meaningful figures, firstly with a twin motor setup you can synchronise the motors so that there is truly a 50:50 split in workload.

You can also compare baseline RPM to Thrust ratios for your prop when new to the current value - gives a propellor 'condition' value.

Thrust would also tend to peak at the point of maximum efficiency and that may be a few hundred rpm different light boat vs fully loaded?

 

Well if you are going to keep up with Donald Duck and them you won't be doing it with an Arduino. Mickey Mouse has NASA grade programming on his autopilot, I can assure you. Hahahaha

You have seen an autopiloted RC plane, but have you seen one LAND by autopilot?

http://www.astrobe.com/default.htm