Jump to content


Let's create the best all-in-one modular tactical instrument for s


  • Please log in to reply
43 replies to this topic

#1 mcl

mcl

    Member

  • Members
  • PipPipPipPipPip
  • 58 posts

Posted 27 August 2013 - 09:20 PM

Hi all,

 

Let's create the best all-in-one modular tactical instrument for sport boats!

 

Why?
Last year our TacTic RaceMaster broke down. While searching for a replacement it became clear to me that there is no single unit available that gives you everything you need: Magnetic compass, GPS, easy interface and the possibility to add speed through water and wind set later on. Maybe the Novasail comes closest. That's probably the reason why you see a lot of teams sailing around with both a TacTic and a ProStart ($1200 worth of equipment and still not everything you would really want). In the end we bought a ProStart, which does the job OK-ish, but still lacks some features to be the single all-in-one tactical instrument.

 

So...
I became a bit frustrated and went looking on electronic components sites. Ended up ordering some components (60ch GPS, LM300 compass, LCD display, lots of wires, some buttons and a battery pack) and an Arduino microcontroller. Total investment: $100.
Soldered the stuff together and did some programming.

 

And now?
I plan on further finishing this thing. I don't know yet what I will do with the result. My primary goal is to have fun and learn while doing. If as a result some boxes will make it into 'production stage' and can be shared amongst friends, that would be cool. And from there on: let's see what happens. One of the main drivers is that the components are readily available so that everyone who wants to give it a shot is able to put this thing together. Also the source code can be shared, so that if you're really into this nerdy stuff, you can make a contribution or implement your own specific features.

 

What can you do?
If you dig this idea as much as I do, please give me as much feedback as you'd like. One of the critical success factors I believe is the ease of use of the interface. So feedback on that is highly appreciated. Below is a high level overview of what the thing is currently capable of and some ideas for future improvements / add-ons.

 

-------------------------

 

 

Result so far
At this moment the whole Frankenstein thing is loosely held together with tape and is in no way water proof or shock resistant at all. However there are four buttons and a working LCD screen and a battery pack consisting of 6 AA batteries which keep the unit running for well over 14 hours. The whole interface is built up around the four stages of a race: Tune-Up, Pre-Start, Start, Race. The four stages can be cycled through via the buttons and every stage gives you the following options:

 

Tune-Up:
Sail upwind and make some tacks. Tacks are automatically detected. The device logs average course over port and starboard, up and down deviation on these courses, mean wind direction plus deviation, average speed on both tacks and a tack angle. Using two buttons you can initiate the tune-up logging, stop it (and store) or start again if conditions severely changed (big shift, significant drop or increase in wind speed).

 

Pre-Start:
Displays a summary of the data collected during Tune-Up. Allows for 'pinging' of the pin end and committee end. After pinging both ends of the line, based on the average wind direction, the biased end of the line is indicated. However this last option is of course quite arbitrary...

 

Start:
Countdown, distance to line, time to line at current speed and time to line on average upwind speed.

 

Race:
Magnetic and GPS heading (HDG and COG). GPS speed. Deviation of course based on last minute (lift/header), deviation of course based on average port / starboard tack. Also the possibility to set new upwind reference data (quick redo of data gathered during Tune-Up) when conditions severely change. Same for downwind.

 

Future improvements - software:

  • Allow pinging of start line not only by getting two ends of the line, but also by sailing on or parallel to the line.
  • Work out a way to get more relevant data on the downwind leg and detect passing of the upwind / downwind mark.
  • Pinging other marks of the course (upwind, spreader, gate), don't know yet if this will be useful and what to do with the data.
  • Improve tack-prediction accuracy during tune-up by combining COG/HDG with heel sensor of compass.
  • Data logging in a common format (csv, gpx)
  • Calibration by user of compass

 

Short term improvements / add-ons (hardware)

  • Waterproof case and waterproof buttons. Doesn't have to look good for now, but should allow to take the unit out for testing instead of running around with it on the parking lot... ;-) Friend has a machinery factory so this shouldn't be much of a problem.
  • Add wireless wrist mounted small display for tactician so he/she can see the numbers without having to stop hiking
  • Add possibility to add through-hull speed sensor and wind set. Figure out how to display this addidional data in a useful way
  • Get better LCD screen. Current thing has 20x4 text only LCD. Should be bigger and graphical. Key of the whole thing is to get an extremely idiot proof and intuitive user interface. This works best using graphics and/or colours I think.

 

Longer term add-ons (roughly in order of realism):

  • Get a really good casing
  • Proper NMEA output
  • Solar cells
  • Possibility to mount a bigger screen (or more of them) in front of the box so the box is useful for bigger boats as well. Around 10", daylight readable, low power consumption. Something like the PixelQi screens ($300). The current Arduino microcontroller is by far not powerful enough to handle such a hi res screen, so additional computational power should be added using something like a raspberry pi or beagleboard. The latter should be integrated in the screen, so the box itself can be left as is, and the screen is a real add-on.
  • 12V adapter and wiring since the bigger screens won't run on the 6xAA batteries for sure.
  • Since additional computational power has to be added for the bigger screens anyhow, might as well put an open-source navigation package on it.
  • Integrate Full HD camera in the bigger screen to allow filming the on board action and add performance data to the video stream. No more need to fit an additional Go Pro anymore...
  • Maybe some entertainment features on the bigger screens as well (music, video) for after-race beer time in the cockpit.
  • More wireless options (connect to laptop over wifi etc etc)
  • Even more wireless options (two-boat tuning with data wirelessly transmitted)
  • etc
  • etc
  • etc


#2 Vogel515

Vogel515

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 242 posts
  • Location:Boston
  • Interests:Sailing, snowboarding, skiing

Posted 27 August 2013 - 10:46 PM

Pretty awesome timing, I was just working on putting together spreadsheets for decoding gpx paths, interpreting tacks, identifying marks/roundings, and I was thinking it would be cool for windward leewards to be able to ping the line, then enter in the heading and distance to the upwind.

 

... different levels of data smoothing for CoG/VMG/SoG

... tack and mark detection

... ability to add in course direction and distance to aid in VMG

... ability to add in gov't mark locations for around the cans

... automatic polar generation

 

A large amount of the work is done...

 

Anyway, if you want to provide me the schema, I will mock up a GUI and put together something... 



#3 Throatwarbler-Mangrove

Throatwarbler-Mangrove

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 3,147 posts
  • Location:New England

Posted 27 August 2013 - 10:51 PM

If I had a large fortune and wanted to make a small fortune, I'd start a sailing instrument company.

 

Add in range/bearing to mark, upwind and downwind laylines, TWD/lifted/headed. 



#4 us7070

us7070

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 6,704 posts

Posted 27 August 2013 - 11:55 PM

anything other than a compass seems contrary to the spirit of small one-design sport-boat racing.



#5 mcl

mcl

    Member

  • Members
  • PipPipPipPipPip
  • 58 posts

Posted 28 August 2013 - 06:25 AM

Anyway, if you want to provide me the schema, I will mock up a GUI and put together something... 

Great! Tonight I will draw some schematics of the GUI so far and post some pictures. Being able to enter direction and distance of upwind mark as posted by the RC is a good suggestion. Also I'll PM you the code used for tack detection and average course per tack. One of the biggest hurdles I had to take there was: Let's say you sail one minute upwind and your course varies between 358 and 002. What's the average direction then? Normal calculations would say 180 ((358+2)/2). But that doesn't make sense, so I came up with some tricks to get it right.



#6 mcl

mcl

    Member

  • Members
  • PipPipPipPipPip
  • 58 posts

Posted 28 August 2013 - 06:36 AM

anything other than a compass seems contrary to the spirit of small one-design sport-boat racing.

Well you could be right there, but my observation is that about 95% of the competitors of the classes I race in don't live up to this spirit then ;-) We've sailed some of our best races while the instruments were totally toast.

However I believe that having the right tool and an extremely intuitive interface to display the data really can assist in the decision making process of the tactician.

If I had a large fortune and wanted to make a small fortune, I'd start a sailing instrument company.

 

Add in range/bearing to mark, upwind and downwind laylines, TWD/lifted/headed. 

You're right about that. There are way better methods to turn a large fortune in a larger fortune. ;-) However this project for me is mainly about learning and fun. And of course about proving that the collective wisdom of a bunch of great minds on sailing anarchy can come up with requirements about for an all-in-one instrument which are way better than anything out on the market at this moment...

Range/bearing to mark: good idea. About upwind and downwind laylines: I think to do this in an accurate way in the end you would need a windset. See 'future improvements'.

About lifted/headed: that's already in (deviation from average upwind angles for last minute and deviation from average upwind angles compared to tune-up data). Still struggling with a way to present this in an intuitive way. I don't believe in the left/right increasing / decreasing bars of the ProStart for example. It isn't intuitive to me and I have to keep on thinking what it means if the bar moves left while on starboard tack: good or bad?

The best way to indicate lifts / headers I've came up with so far is a vertical bar. If it goes up you're on a lift (regardless of port/starboard). If it goes down you're on a header. Does that make sense?



#7 BalticBandit

BalticBandit

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 12,728 posts

Posted 28 August 2013 - 08:57 AM

RockBox comes close to this already



#8 Shaytoon

Shaytoon

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 101 posts
  • Location:Dallas
  • Interests:I have a very hard time engaging in passive relaxation.

Posted 28 August 2013 - 04:05 PM

a screen you can read with polarized glasses...

 

while we are on the subject of the screen, why not make the thing out of a touch screen, buttons could change as you went to each mode...I still have a hard time remembering how each instrument works on each boat...http://www.femacorp....orTFT/id78.html

 

d



#9 RockBox Chris

RockBox Chris

    Member

  • Members
  • PipPipPipPipPip
  • 30 posts
  • Location:Detroit Michigan USA

Posted 28 August 2013 - 05:14 PM

RockBox comes close to this already

 

We do have a lot of what you are looking for already: Here's what we have now http://www.rockboxsa...blue/functions/ and what's new in our latest firmware: http://www.rockboxsa...ware-downloads/

The latest firmware lets you program the buttons on the RockBox and the remote to whatever you want them to do, has waypoint functions (including the ability to project a waypoint with a distance and heading), and a few other improvements. The screen on the RockBox is perfectly compatible with polarized sunglasses too.

 

We've released three new firmware updates this year already, and we are currently working on the magnetic compass (tuning filters, etc...), better filters to detect tacks (for wind direction and wind shift detection) and a few other things. 

 

For the hardware we went with a remote control over a touchscreen because a remote keeps everyone where they should be on the boat, not where the electronics are mounted, and we also designed it so that external sensors could be added later.

 

Rather than developing something from scratch, we would definitely be interested in working with anyone who would like to get their ideas up and running on the RockBox hardware. Send me a message and we can work on the details. Our main purpose in doing this is to have fun and create something useful.

 

Also, Post #3 is more true than you can imagine! 



#10 mcl

mcl

    Member

  • Members
  • PipPipPipPipPip
  • 58 posts

Posted 28 August 2013 - 05:14 PM

Rockbox looks promising, but as far as I can see has no magnetic compass. Or am I wrong?
Ability to read with polarized glasses is for sure a big must have. Will look into this when shopping for the final screen. However the current 20x4 lcd is readable with polarized glasses but lacks graphics.

#11 Throatwarbler-Mangrove

Throatwarbler-Mangrove

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 3,147 posts
  • Location:New England

Posted 28 August 2013 - 06:48 PM

Let's say you sail one minute upwind and your course varies between 358 and 002. What's the average direction then? Normal calculations would say 180 ((358+2)/2). But that doesn't make sense, so I came up with some tricks to get it right.

Modulo arithmetic: 358+2 modulo 360 = 0.



Ability to read with polarized glasses is for sure a big must have. Will look into this when shopping for the final screen. However the current 20x4 lcd is readable with polarized glasses but lacks graphics.

Look at e-Ink.



#12 Alex W

Alex W

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 519 posts
  • Location:Seattle, WA
  • Interests:sailboats, bikes, metalworking, diy

Posted 28 August 2013 - 06:49 PM

I'm interested to see what screen you come up with.  I've been playing with something similar for the last 6 months but my screen choice (256x64 pixel 3" Newhaven OLED) is useless in the sunlight and hard to read with polarized sunglasses.  If you do this with a Netduino then I'm happy to share out my driver code for the screen.

 

I think Rockbox is onto the right idea of using an IR remote.  I put in two hard buttons, but good waterproof buttons are expensive.  A small IR remote can be used from anywhere in the cockpit, is easy to interface, and cheap.

 

I used a clear polycarb case off of this list, but it isn't great, annoyingly it has a molding dot right in the center of where you'd want to put the screen:

http://www.budind.co...A Boxes/NEMA 4x

 

I'm interfacing with my GPS (for compass direction, wind instruments) over NMEA 0183, which is annoying because it has a much lower data broadcast rate than the same GPS would send out over NMEA 2000.  I haven't decided if it is worth getting a CANBUS interface and trying to figure out NMEA 2000, or just not worry about some aspects of my project.  I was playing with some graphical displays showing wind shifts for instance that require a fairly high sample rate to be useful.

 

This whole project started out because wind instruments were annoyingly expensive and the Davis hardware was cheap.  A couple of hundred dollars and lots of time later I have a good wind instrument and a somewhat useful but not great secondary display.



#13 mcl

mcl

    Member

  • Members
  • PipPipPipPipPip
  • 58 posts

Posted 28 August 2013 - 09:05 PM

Chris,

Our posts just crossed each other. Don't get me wrong: I really like the approach and attitude you have a RockBox: open-minded and involved in the community. I'd definitely like to get in touch with you to share ideas. I'll PM you to exchange email addresses. The stage I'm at with my project is in no way nearing a 'final' product. The input I got so far comes mainly from sailng friends and their day to day frustrations about the capabilities and ease of use what's currently out on the market. For now my goal is to equip some of these guys with a box and see how it works out.

One question: is there already a magnetic compass included in the RockBox? And is it just a matter of firmware update to get it activated?

For sure you're right about post #3! And it's indeed about fun and learning.

RockBox comes close to this already

 

 

 

Rather than developing something from scratch, we would definitely be interested in working with anyone who would like to get their ideas up and running on the RockBox hardware. Send me a message and we can work on the details. Our main purpose in doing this is to have fun and create something useful.

 

Also, Post #3 is more true than you can imagine! 



#14 mcl

mcl

    Member

  • Members
  • PipPipPipPipPip
  • 58 posts

Posted 28 August 2013 - 09:24 PM

Alex,

 

Cool to see that there are other people thinking the same way!

 

Currently I use a 20x4 text LCD screen. It's ok for testing purposes. Apart from this I recently received a ezLCD 302 (http://store.earthlc...21&category=297). It's perfectly daylight readable and I will give it some test with polarized sunglasses soon.

It connects quite nicely to the arduino (I have a mega) over serial. It has it's own on board driver so you can just send text, make selections about fonts and draw some primitives. However it requires a bit more 'intelligence' to get the output nicely on the screen, so until now I only use the simple 20x4 screen while I'm still getting the algoritmes for the actual data processing right.

 

I'll PM you the code I have so far once I have access to my PC, maybe we can swap some ideas...

 

The GPS I'm using is a 66ch LS2031 (https://www.sparkfun.com/products/8975) with a sample rate of 10Hz. It sends it's data over I2C bus. Really fast and low computational stress.

 

About the remote: I'm planning to use 2,4GHz Xbee and integrate a tiny screen in the remote so the tactician doesn't have to come out of the rail to see the numbers (wearing it around the wrist as a watch). Maybe even some buttons to swap pages.

 

Indeed, waterproof buttons seem to be relatively expensive. Maybe use micro momentary switches behind a membrane? About the housing: I'm getting a friend involved who has a machinery factory (and sails) so that problem will be solved. Until then: any waterproof readily available housing will do.

 

Have you looked into the e-vane wind sensors already? Any thoughts on that?

I'm interested to see what screen you come up with.  I've been playing with something similar for the last 6 months but my screen choice (256x64 pixel 3" Newhaven OLED) is useless in the sunlight and hard to read with polarized sunglasses.  If you do this with a Netduino then I'm happy to share out my driver code for the screen.

 

I think Rockbox is onto the right idea of using an IR remote.  I put in two hard buttons, but good waterproof buttons are expensive.  A small IR remote can be used from anywhere in the cockpit, is easy to interface, and cheap.

 

I used a clear polycarb case off of this list, but it isn't great, annoyingly it has a molding dot right in the center of where you'd want to put the screen:

http://www.budind.co...A Boxes/NEMA 4x

 

I'm interfacing with my GPS (for compass direction, wind instruments) over NMEA 0183, which is annoying because it has a much lower data broadcast rate than the same GPS would send out over NMEA 2000.  I haven't decided if it is worth getting a CANBUS interface and trying to figure out NMEA 2000, or just not worry about some aspects of my project.  I was playing with some graphical displays showing wind shifts for instance that require a fairly high sample rate to be useful.

 

This whole project started out because wind instruments were annoyingly expensive and the Davis hardware was cheap.  A couple of hundred dollars and lots of time later I have a good wind instrument and a somewhat useful but not great secondary display.



#15 mcl

mcl

    Member

  • Members
  • PipPipPipPipPip
  • 58 posts

Posted 28 August 2013 - 09:26 PM

Modulo arithmetic: 358+2 modulo 360 = 0.

>> Yeah, indeed something like that ;-)

 

Look at e-Ink.

>> sounds like a winner, however I checked it out half a year ago but at least at that moment it was near to impossible to find drivers to control the screen; do you have any clues on how to hook up an e-ink screen to a microcontroller / arduino?



#16 samc99us

samc99us

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 744 posts

Posted 29 August 2013 - 02:03 PM

I've been thinking about this for some time. I started on the hardware side then worried about the software. I couldn't reasonably expect to put something together that was waterproof for less than $300 a year or so ago. That's where the tip-off point was for me, the tack-tick costs a bit more. The trouble is the boat I race is extremely wet, so testing a setup like the Arduino with 6 AA batteries was doomed for failure. I considered going the corporate route but financially I'm not sure there is a viable market. The Rockbox offers a lot of the functions I would want, but relies heavily on GPS. GPS isn't allowed in the F18's except when distance racing (where we turn some class rules off in the interest of safety, more details here: http://www.catsailor...r=248460&page=2).

 

So, how does this sport boat instrumentation setup fit what I would like to use when distance racing? I'm not sure it does, but a few things that might differentiate you from the crowd:

 

1) Small enough setup to be easily mounted anywhere and yet still readable. In my head that is Ipad Mini screen size at max. Maybe a second/remote unit with 2x screens like the tack tick micro compass. Love your wireless setup for the tactician idea, if I'm stuck on a rail calling tactics 9/10 I use a GPS wrist watch rather than getting off the rail and using the full nav suit down below.

2) Put some dang maps on there-can be disabled for buoy racing but flash memory is cheap!

3) Consider adding the ability to pair with a cell-phone for data. Look at what is allowed in your class rules, but this is something that will make you standout and appeal to the big boat crowd (read: money). Start with just adding wind and current as a function. Consider adding full routing capabilities down the road. Maybe this is all too ambitious, and I know not needed in the typical sport boat buoy race.

4) Use of a 6 DOF mem's sensor to add better filtering, wave state predictions etc. Again maybe too ambitious but I think required to handle the GPS anomalies, plus some of the new Invensense chips get you a magnetometer as well. Not as good as a standalone Honeywell HMR2300, or even the much cheaper single-chip solutions but it's a pretty dang good starting point.

 

If you want some help let me know, I might be able to spare a few hours for some code writing.



#17 Murphness

Murphness

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 713 posts
  • Location:Boston
  • Interests:Sailing, Brewing Beer, Drinking Beer, Skiing and other outsidey things

Posted 29 August 2013 - 02:30 PM

Hardware: http://www.raspberrypi.org/faqs

 

$35 is a pretty good start. Get a readable screen and design a waterproof case and you're good to go for under $100 likely...



#18 Alex W

Alex W

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 519 posts
  • Location:Seattle, WA
  • Interests:sailboats, bikes, metalworking, diy

Posted 29 August 2013 - 03:59 PM

The Beagleboard Black is a much better choice for this project than the Raspberry Pi.  It is about the same CPU power, a couple of dollars more, but has a much better interface for plugging in other electronics.  

 

Both it and the Raspberry Pi won't run for too long on AA batteries.  That is less of a concern for me since my boat has plenty of 12V, but a standalone device that works on dinghies and sport boats would be cool.

 

I haven't seen the e-vane, but a quick look at it is very interesting.  It doesn't include the anemometer though, and I find that to be more useful than wind direction.  The wind direction on mine has been off by 10 degrees for over a month now, and I haven't even felt inclined to download new software to offset that.  The Davis Wind sensor is about the same cost.



#19 TheFlash

TheFlash

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 13,697 posts
  • Location:San Francisco Bay
  • Interests:Rum

Posted 29 August 2013 - 04:11 PM

sometimes even the compass is too much. I stick with a watch...



#20 Raked aft \\

Raked aft \\

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 955 posts
  • Location:The North Coast

Posted 29 August 2013 - 04:50 PM

Not sure if available in any of the above instruments, but I would like to see some sort of graphic displaying trends.

 

  if after a square beat the unit let you know that there was a ? deg left trend through all the smaller oscilations, or oscilations

were even both sides but moved 5 deg either side of average over a ? min period.

 

  On the speed front, seeing a bar graph showing current speed as a pointer ">" which moves up and down over a shaded area, where

different shades represent an average over a certain period of time and both faster and slower than that average.



#21 RockBox Chris

RockBox Chris

    Member

  • Members
  • PipPipPipPipPip
  • 30 posts
  • Location:Detroit Michigan USA

Posted 29 August 2013 - 08:47 PM

Not sure if available in any of the above instruments, but I would like to see some sort of graphic displaying trends.

 

  if after a square beat the unit let you know that there was a ? deg left trend through all the smaller oscilations, or oscilations

were even both sides but moved 5 deg either side of average over a ? min period.

 

  On the speed front, seeing a bar graph showing current speed as a pointer ">" which moves up and down over a shaded area, where

different shades represent an average over a certain period of time and both faster and slower than that average.

 

We've got that. Next to Speed and VMG there are arrows on the RockBox that are based on averages to show you a larger view of the trend, while the numbers show you the instant info.

It also has histogram graphs, you can set the total period for the display and watch COG to see a long term wind shift.



#22 RockBox Chris

RockBox Chris

    Member

  • Members
  • PipPipPipPipPip
  • 30 posts
  • Location:Detroit Michigan USA

Posted 29 August 2013 - 08:52 PM

Chris,

Our posts just crossed each other. Don't get me wrong: I really like the approach and attitude you have a RockBox: open-minded and involved in the community. I'd definitely like to get in touch with you to share ideas. I'll PM you to exchange email addresses. The stage I'm at with my project is in no way nearing a 'final' product. The input I got so far comes mainly from sailng friends and their day to day frustrations about the capabilities and ease of use what's currently out on the market. For now my goal is to equip some of these guys with a box and see how it works out.

One question: is there already a magnetic compass included in the RockBox? And is it just a matter of firmware update to get it activated?

For sure you're right about post #3! And it's indeed about fun and learning.

 

Cool, PM me and I'll send you my email address.

 

Yes, the RockBox already has the hardware for the compass, we're just working on the firmware to run it.



#23 trkarl

trkarl

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 106 posts
  • Location:Fort Collins, CO

Posted 29 August 2013 - 09:10 PM

I think Rockbox is onto the right idea of using an IR remote.  I put in two hard buttons, but good waterproof buttons are expensive.  A small IR remote can be used from anywhere in the cockpit, is easy to interface, and cheap.

 

I seriously doubt that Rockbox uses IR for their remote. IR does not work well in full sunlight. I bet they're using RF, which can control it anywhere on the boat not just line of sight.

 

I'd like to add that this sounds like fun!



#24 RockBox Chris

RockBox Chris

    Member

  • Members
  • PipPipPipPipPip
  • 30 posts
  • Location:Detroit Michigan USA

Posted 29 August 2013 - 09:13 PM

I think Rockbox is onto the right idea of using an IR remote.  I put in two hard buttons, but good waterproof buttons are expensive.  A small IR remote can be used from anywhere in the cockpit, is easy to interface, and cheap.

 

I seriously doubt that Rockbox uses IR for their remote. IR does not work well in full sunlight. I bet they're using RF, which can control it anywhere on the boat not just line of sight.

 

You're right about that, its RF.



#25 andyxs

andyxs

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 375 posts

Posted 30 August 2013 - 07:54 AM

How long have you got? these guys were just looking into making the 20/20 style screens but they got it all working, the biggest problem seems to be the case. 

http://www.ybw.com/f...ght=Instruments



#26 BalticBandit

BalticBandit

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 12,728 posts

Posted 30 August 2013 - 08:45 AM

Well the thing is that pretty much anything that does automated calculations for you is banned in classes that ban GPS.   But most devices these days are moving towards GPS.    So one way some class rules deal with this is by banning any electronics that offer more functionality than you could get with a mechanical magnetic compass.

 

 

Now there's a fair amount of info that the older high end mechanical compasses could do:  You could get compass roses with multiple rings that would let you set median wind settings for both tack as well as TWD or Course TO Mark.  But most of the functionality being discussed here is not legal.

 

Remember that on dinghies, and smaller boats, part of the challenge is in being able to learn how to do "time and distance" calcs in your head.   If I have a startline inside my electronics.  And I've got the polars for my boat, and its GPS location and the windspeed,  I can calculate a "Time to the Line" dynamically as you sail

 

That means I no longer have to use my own intuition as to how much I need to slow down to not be OCS, takes a large part of the skill of starting out of the equation.



#27 Alex W

Alex W

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 519 posts
  • Location:Seattle, WA
  • Interests:sailboats, bikes, metalworking, diy

Posted 01 September 2013 - 04:21 AM

Thanks for confirming that Rockbox uses RF.

 

That thread from YBW is using this LCD:

http://uk.rs-online....splays/5327221/

 

I'll upload and share out my code next time I'm using the computer that has it loaded.  If nothing else it has useful libraries for NMEA 0183, drawing various size fonts to these bitmap displays at high speed, and giving you a nice debug dump of what is going on with your NMEA traffic.  It looks like I got to a similar point that the YBW guys did, except that I built my case and am unhappy with my display, while they are happy with their display but not the case.



#28 nroose

nroose

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 4,559 posts
  • Location:Berkeley

Posted 01 September 2013 - 05:02 AM

If I was to do this, I would start with an iPad Mini with 4g and a decent case. I would write an app. Then I would figure out how to integrate it with the possible wind and speed instruments. I would focus on usability of the app. I have some ideas about graphic visualizations that I would love to have while sailing. I would try to avoid inventing the thing from scratch. I think my biggest challenge would be to stop my kids from playing games on it during the race.

#29 Alex W

Alex W

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 519 posts
  • Location:Seattle, WA
  • Interests:sailboats, bikes, metalworking, diy

Posted 01 September 2013 - 02:00 PM

The iPad mini screen isn't useable in full sunlight.  The iPads that we've tried on the boat (older ones) also overheat when used in full sun and shut down.  I live in Seattle, I can't imagine how quickly that would happen somewhere hotter.

 

The Nexus 7 screen is a little better in full sunlight, but still not good enough.

 

This is interesting hardware however:

http://www.meetearl.com/

 

IP67, e-ink, GPS, good battery life, reasonable price.



#30 mcl

mcl

    Member

  • Members
  • PipPipPipPipPip
  • 58 posts

Posted 03 September 2013 - 07:00 PM

Wooaaa! Couple of days on holiday and already so many replies! Very cool!
I'll use these days off to revise my 'roadmap' based on your feedback and ideas. And of course come back on various individual ideas/suggestions.
About the bigger screen : I already have a beagleboard lying around. I'll look into the other big screen project. For now I am thinking about the PixelQi screen. 300$, 10", low power and sunlight readable.

If I were to summarize the different replies I would say: 1)keep it simple and 2) user interface is key.
I think about getting a friend in who is extremely good in interfaces, but doesn't have a clue about sailing. Might be good for some out of the box outside in perspective... ;-)

#31 Alex W

Alex W

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 519 posts
  • Location:Seattle, WA
  • Interests:sailboats, bikes, metalworking, diy

Posted 04 September 2013 - 12:15 AM

I spent some of this morning looking at better screens for my project.  I think that this is a strong candidate for my personal project.

http://www.newhavend...tabs-p-610.html

 

It is transflective like the PixelQi, 20% of the cost, and what I think is a more useful size.  Monochrome, which is better for my Netduino-based project (memory is an issue with full color displays), but not really a concern with the Beaglebone Black.

 

The annoyance with that one is that it has a parallel-based interface.  The same manufacturer has some with a serial interface that would be easier to work with, but they are a little big for my project.



#32 Throatwarbler-Mangrove

Throatwarbler-Mangrove

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 3,147 posts
  • Location:New England

Posted 04 September 2013 - 02:08 AM

The iPad mini screen isn't useable in full sunlight.  The iPads that we've tried on the boat (older ones) also overheat when used in full sun and shut down.  I live in Seattle, I can't imagine how quickly that would happen somewhere hotter.

 

The Nexus 7 screen is a little better in full sunlight, but still not good enough.

 

This is interesting hardware however:

http://www.meetearl.com/

 

IP67, e-ink, GPS, good battery life, reasonable price.

IP67 is given.  Beware that IP67 tests dust and water intrusion, not corrosion or long-term gasket deterioration or UV deterioration.

 

I like e-Ink as a display for marine instruments.    Low power (important), great contrast, no polarization (I think)...  works great with sunglasses. 

 

There is GPS and there is GPS.  For tactical instruments, you want precision that probably isn't going to come from an embedded GPS device with internal antenna.

 

Tablet form factor is kind of awkward for a tactical instrument, unless you have a  pedestal mount. 

 

That said, this is overkill.  Lots of stuff you don't need (consuming power, real estate, BOM cost).   On the right track, though.



#33 Throatwarbler-Mangrove

Throatwarbler-Mangrove

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 3,147 posts
  • Location:New England

Posted 04 September 2013 - 02:24 AM

I spent some of this morning looking at better screens for my project.  I think that this is a strong candidate for my personal project.

http://www.newhavend...tabs-p-610.html

 

It is transflective like the PixelQi, 20% of the cost, and what I think is a more useful size.  Monochrome, which is better for my Netduino-based project (memory is an issue with full color displays), but not really a concern with the Beaglebone Black.

 

The annoyance with that one is that it has a parallel-based interface.  The same manufacturer has some with a serial interface that would be easier to work with, but they are a little big for my project.

Be sure to try samples in bright sunlight, from all reasonable angles, while wearing sunglasses.   A tactical instrument that you can't read from the rail is useless.

 

Again consider dust, water intrusion, corrosion, UV.  You may be able to come up with a COTS package that will serve the need.  Otherwise, it becomes a big mechanical project.

 

Parallel interface should be at most a minor nuisance, a few dozen lines of code.  Unless you're concerned about pinout constraints on your MCU.  I haven't looked specifically at Arduino, but find it difficult to believe that it is lacking in GPIO ports.



#34 Alex W

Alex W

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 519 posts
  • Location:Seattle, WA
  • Interests:sailboats, bikes, metalworking, diy

Posted 04 September 2013 - 02:56 PM

I agree that the displays require good testing.  I went with OLED at first not expecting it to have problems with polarized sunglasses, but it does.

 

Arduino/Netduino don't have enough pins to do parallel displays unless that is the only device that you'll be using.

 

The Bud enclosure that I'm using is IP67.  It is easy to get IP67 cable gaskets as well.  The display has been the hardest part for me.

 

I haven't had much luck finding small e-ink panels.  Is there one that you can recommend?  I agree that it would be ideal.



#35 Throatwarbler-Mangrove

Throatwarbler-Mangrove

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 3,147 posts
  • Location:New England

Posted 04 September 2013 - 03:34 PM

Google is your friend.  For example, have you checked out these guys?

 

I hear you on the pin-out constraints.



#36 Alex W

Alex W

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 519 posts
  • Location:Seattle, WA
  • Interests:sailboats, bikes, metalworking, diy

Posted 05 September 2013 - 02:15 PM

Fair enough.  I meant one that you'd had personal experience with.  That kickstarter seems to be unavailable (people who joined the KS don't even have their panels yet), and the one that Adafruit sells has a parallel interface and doesn't seem to work well from reports that I can find.

 

No backlight make also make these less useful for this purpose than a transflective LCD.



#37 Throatwarbler-Mangrove

Throatwarbler-Mangrove

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 3,147 posts
  • Location:New England

Posted 05 September 2013 - 11:30 PM

Personal experience?  Sorry.  Can't help. 

 

e-Ink does make front lighted displays.   Of I suppose you could illuminate it from the top with a small LED strip + diffuser.

 

Did you contact e-Ink to see if they'd sell you samples?  Or maybe try a Maker forum?



#38 Alex W

Alex W

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 519 posts
  • Location:Seattle, WA
  • Interests:sailboats, bikes, metalworking, diy

Posted 06 September 2013 - 02:25 PM

The e-ink sample kit is what Adafruit is selling.  It looks like all of these kits are still using the older non-lit displays (like an earlier Kindle).

 

I took a couple cell phone images of what I've built so far:

 

WP_20130904_003.jpg?psid=2

 

That is the default screen.  Wind speed/depth/SOG.  On top is a linear drawing of the wind angle.

 

WP_20130904_004.jpg?psid=2

 

Back side showing both electrical connections.  USB for programming, the wire sticking out of the side carries power and NMEA 0183 for interfacing to the boat.  Enclosure, front buttons, and connections are all NMEA 4x (good) or IP67 (great).  USB is only watertight when the cover is on (hanging from a lanyard there).

 

WP_20130904_005.jpg?psid=2

 

Another view showing size.  Not as small as I'd like, but okay.

 

When I find a new display I'm going to take another stab at the enclosure using an opaque cover that I add a clear display lens to.  The clear cover on this BUD enclosure is too reflective and has that molding dot right in the center.

 

None of it is rocket science, it's just fun stuff to play with.



#39 FatimaRules

FatimaRules

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 797 posts

Posted 09 September 2013 - 04:14 AM

Disclaimer: I know enough about this stuff to be fully aware of how little I know!

I ride bikes a lot, and there a small group of companies using the iPhone as the central processor for both sensors (heart rate, cadence, speed) and small remote displays, interfaced through Bluetooth. Wahoo fitness is a good one to look at.

Iregatta et al already have the majority of the clever stuff covered.

Is the answer the same model? How hard to get Bluetooth speed, depth, compass and wind speed sensors?

#40 ColinG

ColinG

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 1,551 posts
  • Location:Sydney, AUS

Posted 09 September 2013 - 10:58 PM

Have a look at this.

http://www.holdentec...th-nook-display



#41 ColinG

ColinG

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 1,551 posts
  • Location:Sydney, AUS

Posted 10 September 2013 - 02:23 AM

Along the same lines, you could always root a waterfi waterproofed kindle as the display



#42 red_sky_sf

red_sky_sf

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 90 posts

Posted 11 September 2013 - 11:01 PM

The e-ink sample kit is what Adafruit is selling.  It looks like all of these kits are still using the older non-lit displays (like an earlier Kindle).

 

I took a couple cell phone images of what I've built so far:

 

WP_20130904_003.jpg?psid=2

 

That is the default screen.  Wind speed/depth/SOG.  On top is a linear drawing of the wind angle.

 

WP_20130904_004.jpg?psid=2

 

 

 

WP_20130904_005.jpg?psid=2

Good box selection, I use same for an AP I have created.  Consider the teensy++2 as it has lots of I/O for display drive, or go with a serial display using I2C.  For buttons these come in multi colors: http://www.mouser.com/Search/ProductDetail.aspx?R=48-1-RB-N-RD-Bvirtualkey56650000virtualkey850-48-1-RB-N-RD-B   The round buttons seal nicely, hole size is 1/2".



#43 muppet

muppet

    Anarchist

  • Members
  • PipPipPipPipPipPipPip
  • 205 posts

Posted 13 September 2013 - 09:08 AM

Have a look at this.

http://www.holdentec...th-nook-display

Purchased a Nook Glow Light to run iregatta. Just wanted to compare boat speed with polars and have a function to quickly calculate gybe angles

 

Good idea due to the e ink screen being daylight visible, has a night light and very good battery life. I bought a clip on case and stowed it in a zip lock bag and all was ok for the occasional splash it would get.

 

Rooted ok, picks up nmea through wifi with no problem, but, the version of Android is so old that most modern programs including iregatta will not run on it and has a screen saver which is not possible to turn off so you have to keep reminding it that you are there. Needs to have someone write software specifically for it.

 

Now purchased a second hand Gen 5 ipod with retina screen to do the same thing. Not quite as good solution for battery efficient as you have to turn up the brightness. May buy another and just swap them over to recharge.



#44 SailingPerformance

SailingPerformance

    Newbie

  • Members
  • Pip
  • 10 posts
  • Location:Valencia, Spain

Posted 13 September 2013 - 01:46 PM

Wow, awesome project! Does your system have a datalogger system, where you might be able to recover a data log at the end of the day? If so we suggest you try out our RaceReplay software to view the data, assist you in checking all your variables, improve your calibration, and once this is done, get into performance analysis so you can start building all your polars and targets.

Let me know if you are interested in trying it out!

Cyrille

KND SailingPerformance






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users