Author Topic: How achievable are my design goals using Moteino?  (Read 4874 times)

delateurj

  • NewMember
  • *
  • Posts: 14
How achievable are my design goals using Moteino?
« on: November 17, 2014, 11:55:36 PM »
Background:
The system is for judging model airplane pylon races.

The judges in a race basically watch if the planes have crossed a point to determine how many laps have been completed and whether or not the airplane has gone beyond a pylon (marker on the race course) and therefore is not “cutting corners” of the race course.

For each judge station there will a Moteino based device for each judge at near ground level (lets refer to these as “judge ground stations”) that the judge will interact with (i.e. push a button or throw switch).

For each judge station there will be a Moteino mounted on a 10’ pole near the judge station (within 10’).  Lets call that a “judge station repeater”

Three places need to know the action of the judges.  Each of these are up to 600 feet from the judges station.  At each of these there will be moteino on a 10’ pole (lets call this a “control unit repeater”) and a unit near the ground communicating to a light system (lets call this a “ground control unit”).

Two judge stations will have 4 judges, so 4 of the ground moteino units.
Two judge stations have 1 judge with one of the ground judge units.

So in total there will be:
10 judge ground stations (4 judges, 4 judges, 1 judge  and 1 judge)
4  judge station repeaters
3 control unit repeaters
3 ground control units

For a total of 20 moteinos.

The area is generally flat open space but there can be some chain link fencing, shrubbery , chairs, tables and people at ground level.

In general, in response to a judge action a judge ground station will communicate to the judge station repeater (which will be listening to up to 4 judge ground stations) which will communicate to all  3 control unit repeaters that will each communicate to their corresponding ground control unit.

The information passed will be small,  4 bytes should be enough.

Actions will occur approximately every 6 seconds at every judge unit and could be near simultaneous (e.g. if two or more planes in the race cross the start finish/line at same or nearly the same time then the 2 or more judges at that judge station could initiate a communication at nearly the same time)

A race is up to 11 laps and approximately 1 minute long, so depending on the number of cuts there is approximately 100 total messages generated from the ground stations.  Its less than 10 stations times 11 laps because some judge station do not have an action every lap.

Acceptable latency performance would be a delay between judge action and corresponding ground unit receiving the data correctly being less than 200 ms.   Good to great would be less than 100 ms.

Acceptable reliability is that in 99.9% of races all messages get through successfully.   Which means looking for 99.999% reliability of each individual message.

Seems like its doable using Moteino from reading threads and documentation and I would rather use it than Xbee because if nothing else than keeping track of all the Xbee product numbers, differences in API, firmware naming etc. gives me a headache plus this would be about $36/unit cheaper than Arduino FIO + Xbee option I am considering and I just think the Moteino is cool and like the transparency.

So my primary question is: Are my performance and reliability targets reasonable given the setup using 900 mhz RFM69HW?

Thanks,
Joe. 

TomWS

  • Hero Member
  • *****
  • Posts: 1930
Re: How achievable are my design goals using Moteino?
« Reply #1 on: November 18, 2014, 08:24:57 AM »
What a cool project!  From the little I've used Moteino, these are the perfect choice for this project.  They have the range, sendWithRetry(), especially if you increase and possibly randomize the retry timeout, will get your message through, and the built-in encryption makes the message integrity that much more robust. 

Given the nature of your application, I would recommend investing in good quality antenna and properly 'connectorize' the Moteino with an edge mount through-panel SMA connector where the panel serves as a ground counterpoise for the antenna.  This will significantly improve the range and coverage and won't add that much cost.

Have fun!  This sounds like a good one!

Tom

Felix

  • Administrator
  • Hero Member
  • *****
  • Posts: 6866
  • Country: us
    • LowPowerLab
Re: How achievable are my design goals using Moteino?
« Reply #2 on: November 18, 2014, 09:18:09 AM »
Thanks TomWS, I agree. 600ft is not that much, less than 200m. I know others have reached 500m+ in open air with the default 55.5kbps settings. Since payloads are so small, bitrate and bandwidth can be significantly lowered to achieve much longer range if needed.
The default settings are meant to fit most purposes well, but every once in a while a project is an outlier and requires a bit of tweaking, but the radios are certainly capable of reaching 1.5miles or more at very low bitrates, as shown in the sticky range topics of this forum. Here's another post that outlines the default settings range possibilities: http://tomblanch.wordpress.com/2014/03/21/moteino/

Another proposition I received recently (i think by Tom) was to use hardware node addressing which might improve the latency and reliability in intense traffic/collision situations. But don't fix what is not broken, if the default settings work then great, nothing else to do.

And as always there is the disclaimer. The hardware is just cold metal and semiconductor. What matters more is how the firmware controls the metal. I feel strongly that Moteinos could do the job but I always recommend new users to give them a try with a pair before investing in very many. And HW-900mhz is probably just as good as anything for your application.
« Last Edit: November 18, 2014, 09:20:52 AM by Felix »

TomWS

  • Hero Member
  • *****
  • Posts: 1930
Re: How achievable are my design goals using Moteino?
« Reply #3 on: November 18, 2014, 10:20:39 AM »
<...snip>

Another proposition I received recently (i think by Tom) was to use hardware node addressing which might improve the latency and reliability in intense traffic/collision situations. But don't fix what is not broken, if the default settings work then great, nothing else to do.

<snip...>
I can't take credit for the proposal but I do agree with it for some applications.  I don't think this is one of them, however.  I completely agree with your position 'if it ain't broke...' and, in this case, Moteinos, as they are currently implemented are literally perfect for the job. 

It doesn't seem that this application would be power constrained so running off a quad pack of Lithium AAs (or even a 6V Lantern battery  :D  or LiPo if you want rechargeable) would power these nodes very nicely. 

delateurj

  • NewMember
  • *
  • Posts: 14
Re: How achievable are my design goals using Moteino?
« Reply #4 on: November 18, 2014, 10:55:18 AM »
Thanks for the input. 

Hardware address filtering was one item in the forum that caught my eye that could help if needed but will do some testing without.

Thanks for the SMA connector antenna suggestion, however I am hoping with the nearby repeaters raised up and going with the high power version I could just use the whip antenna given short distances (30' from ground to repeater, 600' between repeaters).  Existing xbee based systems have connectors, cables and antennas and this complicates set up and tare down and the course is changed depending on type of racing multiple times per day and all the cables make moving the judges cumbersome.  And while improving transmission reliability, this introduces another point of failure and mechanical in nature that has created problems in the past (race halted for an hour while diagnosing a problem that turned out to be bad antenna cable/connection).  Also, my own personal skills are stronger in programming and overall system design than mechanical/enclosure design.

Is my assumption on whip antenna good for 30'/600' have hope?  I'll have the same challenge with xbee so doesn't really affect design choice in that regard.

Regards,
Joe.

 

TomWS

  • Hero Member
  • *****
  • Posts: 1930
Re: How achievable are my design goals using Moteino?
« Reply #5 on: November 18, 2014, 11:29:03 AM »
Thanks for the input. 

Hardware address filtering was one item in the forum that caught my eye that could help if needed but will do some testing without.

Thanks for the SMA connector antenna suggestion, however I am hoping with the nearby repeaters raised up and going with the high power version I could just use the whip antenna given short distances (30' from ground to repeater, 600' between repeaters).  Existing xbee based systems have connectors, cables and antennas and this complicates set up and tare down and the course is changed depending on type of racing multiple times per day and all the cables make moving the judges cumbersome.  And while improving transmission reliability, this introduces another point of failure and mechanical in nature that has created problems in the past (race halted for an hour while diagnosing a problem that turned out to be bad antenna cable/connection).  Also, my own personal skills are stronger in programming and overall system design than mechanical/enclosure design.

Is my assumption on whip antenna good for 30'/600' have hope?  I'll have the same challenge with xbee so doesn't really affect design choice in that regard.

Regards,
Joe.
If you're not making a PCB to carry the Moteino and just using its board directly, then I understand your concern about cabling and such.  A wire whip should work well enough.

In most of my Motes I make a PCB to carry the Moteino as a module along with my peripheral circuitry.  In cases where I care about range, mechanical/weatherproof integrity, and have the room, I will add the PCB traces for an edge Mount, long neck, SMA connector.  This actually provides good mechanical mounting for the PCB and, by just making a 4x4 inch galvanized steel plate, provides a nice counterpoise for a commercial antenna, all very solid and I only need to drill a 1/4" hole for the connector.  The panel mounts behind my plastic weatherproof enclosure and is held in place with the nuts on the connector shank.

Have fun. 

PS: I really don't think you'll have an issue with SW address filtering.

ggallant

  • NewMember
  • *
  • Posts: 23
Re: How achievable are my design goals using Moteino?
« Reply #6 on: November 18, 2014, 02:36:25 PM »
Tom,

Could you supply a picture of the antenna assembly? I am slowing getting started with RFM69's and radio transmission is black magic to me.

George

delateurj

  • NewMember
  • *
  • Posts: 14
Re: How achievable are my design goals using Moteino?
« Reply #7 on: November 18, 2014, 03:45:40 PM »
Ordered 4 with the 2 I already have I can set up an initial test with 2 sets of 2 ground stations and 2 repeaters.

Tom...I will be creating a board for the ground stations since a number of connections.  Repeaters are simpler and will want to mount inside pvc pipe that will be used for extension.  Thanks for your suggestions, I'll see what my initial tests with wire antenna show.

Regards,
Joe.

TomWS

  • Hero Member
  • *****
  • Posts: 1930
Re: How achievable are my design goals using Moteino?
« Reply #8 on: November 18, 2014, 05:24:43 PM »
Tom,

Could you supply a picture of the antenna assembly? I am slowing getting started with RFM69's and radio transmission is black magic to me.

George
I'm waiting for my WiFi-Moteino Gateway PCBs that use this method with Moteinos, but somewhere around here I have another PCB that uses these connectors.  I'll have to dig for them though.  I'll post as soon as I have something.
Tom

TomWS

  • Hero Member
  • *****
  • Posts: 1930
Re: How achievable are my design goals using Moteino?
« Reply #9 on: November 18, 2014, 05:30:09 PM »
<...snip>Repeaters are simpler and will want to mount inside pvc pipe that will be used for extension.  Thanks for your suggestions, I'll see what my initial tests with wire antenna show.

Regards,
Joe.
I think it's so cool that I can fit the Moteino, as is, inside a PVC pipe!  I've already got three designs that use this method outside.  As soon as the bloomin' slow boat from China arrives, I'll post them...  Probably under 'Projects'.

You're welcome on the suggestions, I'm interested in seeing your results.

Tom

TomWS

  • Hero Member
  • *****
  • Posts: 1930
Re: How achievable are my design goals using Moteino?
« Reply #10 on: November 18, 2014, 05:54:21 PM »
Tom,

Could you supply a picture of the antenna assembly? I am slowing getting started with RFM69's and radio transmission is black magic to me.

George
I'm waiting for my WiFi-Moteino Gateway PCBs that use this method with Moteinos, but somewhere around here I have another PCB that uses these connectors.  I'll have to dig for them though.  I'll post as soon as I have something.
Tom
I've attached a photo that shows the concept.  I cobbled this together from an obsolete PCB and hardware.  The sad looking device in the background is an inground moisture sensor that spent 6 years in the hot Texas sun but demonstrates how this setup could be used.

Let me know if you have any questions.

Tom

ggallant

  • NewMember
  • *
  • Posts: 23
Re: How achievable are my design goals using Moteino?
« Reply #11 on: November 22, 2014, 09:29:05 PM »
Thanks for the pictures. I will reread the forum posts on antennaes.

delateurj

  • NewMember
  • *
  • Posts: 14
Re: How achievable are my design goals using Moteino?
« Reply #12 on: December 18, 2014, 08:26:35 PM »
Wasn't sure if the Moteino forum was best or the project forum was best, but decided to start a thread in the project forum to provide updates on this project: https://lowpowerlab.com/forum/index.php/topic,796.0.html

Regards,
Joe.