Jump to content
 

Discussion on using JMRI and WIFI


Recommended Posts

At a recent outing with my N gauge layout , I am using JMRI and WiFI using either iPads and Withrottle or Android device and Engine driver. On  several occasions the DCC signal appears to have been lost resulting in control from the IPAD OR Android device being lost. and resulting in the locomotive or scenic feature not responding to the input. At home the system was tried several weeks with no problem, but at a recent outing the set up became problematic that the operators resorted to wired handsets was taken. The discussion I will forward through the JMRI user group, but I am wondering if any member on reading this forum has experienced the same and how was the matter resolved. An observation was made that the using wi-fi throttle there is an option to use web server , however on several occassion the observation was made on the iPads this facility was intermittent . The wifi router was positioned high on a purpose built self above the lighting rig so there was no blocking of the wifi signal between. I am wondering if the configuration on the network bridge on the Computer using windows 7 had been configured correctly or was in conflict with a wifi signal on one of the other layouts. Any guidance would be appreciated .

Stuart.

 

 

 

 

Link to post
Share on other sites

At any show, you're risking the "noise" of dozens of other unknown WiFi devices, some of which may be swamping your own setup.    Diagnosing that isn't easy. 

 

What are you using as your show access point ?  Is it a dedicated device, or are you turning a Win-7 machine into an access point ? 

 

- Nigel

 

 

Link to post
Share on other sites

This is unfortunately a known issue at shows and is becoming more regular as more people use more Wifi - especially WiFi that hasn’t been configured properly and may not be using a dedicated WiFi router.

 

as Nigel says, impossible to diagnose and rectify that specific situation after the event, all you can do is ensure that good design practices are employed in your network to overcome the failures of design in other networks

Link to post
Share on other sites

I'd be tempted to try a Linux device to run JMRI rather than Windows.  I don't know why this is the case, but I find JMRI under Linux on a Raspberry PI is more predictable than JMRI under Windows on a fairly fast PC.   You can copy your entire JMRI setup between devices, so no work is lost in trying this out. 

 

But, bottom line on this is the router, its SSID (if its a default/common one, chances are that some visitors to the show will have devices which are automatically trying to connect to your router, those repeated failed connection attempts are not helping), the choice of channel number (ideally not on the same channel as someone else), the configuration of the client devices (phones/tablets).   Note that some phones/tablets get upset if their WiFi doesn't include the wider internet, and may try to find alternative networks (so you have to stop that happening in the devices, or provide a wider internet upstream access to stop the device "complaints"). 

 

 

Link to post
Share on other sites

WiFi is simply not a guaranteed reliable system and it disgusting that it is ever be sold as such or promoted for connecting static equipment such as desktop PC - and especially video streaming use such as with smart TVs etc.

 

All manufacturers tests and speed / performance claims are done in isolation, with no other systems in range to interfere & no "hidden station" effects.

 

In the real world using the 2.4GHz band, interference, collisions and hidden stations are inevitable and speeds can be a tiny fraction of the supposed throughput. And that assumes there is no other 2.4GHz gear such as video links or baby monitors which desense or block WiFi receivers without actually being visible to WiFi devices.

 

At one time, through the "11g" 54Mbit stage, it was possible to have up to three separate WiFi systems coexist in the same area without interference, if they were set to channels, 1, 6 & 11. Each 54Mbit connection uses 4-5 channel slots in bandwidth.

 

 

Newer, "faster" gear typically uses 8 channel slots of bandwidth - it's pretty much impossible for even two systems to coexist in range of each other without interference.

 

With dozens of access point running simultaneously at an exhibition, it's amazing that any work to any acceptable level!

 

The 5GHz band is much bigger and can support a lot of simultaneous users, but there is still little support for it in consumer gear.

 

 

If you must use 2.4GHz, try turning off the SSID broadcast on the AP or router, so only your configured devices can use it and other people / devices do not see it as a possible connection. There is not much else you can do.

 

[Designing Radio & electronics systems for 40+ years].

 

  • Agree 1
  • Informative/Useful 1
Link to post
Share on other sites

In response to the questions I was using a laptop with windows 7 and using a t-link 450mbps wireless n router. There is no direct internet access just the Wi-fi output from the router . All the iPads and iPhone or android devices password secure. I did observe access to web server on the hand held devices repeated dropped out. And whilst WiThrottle window recognised which hand held device was connected via the Wi-fi .

Link to post
Share on other sites

It might have been the venues access point that was in the same room that you were in. Neware St Boobays was operating next door just using a USB dongle on a Raspberry Pi and we didn't seem to have any Wi-Fi issues, the Pi crashed a couple of times but it was its first day out!

 

I didn't check what channel they were operating on but it might have been down to clashes between the two units. You did mention the system operated faultlessly at home, is this still the case?

Link to post
Share on other sites

I'll just stick my 2p worth in here. A few months ago I wrote a pair of articles for the NMRA BR region magazine about getting the most out of DCC for Freemo (modular) events. Part 2 was around wifi and wireless throttles. 

 

A quick summary of best practices. 

 

1: use 5ghz where possible. Every phone and tablet made in the last 6 years supports it. 

2: use a dedicated router and extend the number of available dhcp addresses. This helps solve the lost lease issue in IOS. Most home routers are set to 10 addresses.

3: get a copy of inssider (windows app) and use it to view and monitor the channel usage and other Wifi nodes in the vicinity. Use this to find the best , as in least used channels. Never let the router automatically set the best channel for you. 

4: switch off ssid broadcast if you can. That reduces the number of "guest" devices attempting to connect. 

5: follow the router instructions around placement and antenna orientation. Don't bury it under the layout. 

6: use a cable from the laptop running JMRI and the router. (better throughput) 

7: make sure the laptop isn't running anything else that tries to call out to the internet. Adobe updates for example. 

 

 

Link to post
Share on other sites

32 minutes ago, Talltim said:

The Freemo situation is the reverse of an exhibition. (Unless you are doing Freemo at an exhibition!) Only one network but many more devices connecting than on any normal layout.

Not necessarily. The last one I provided the DCC and wifi for had no less than 18 wifi points visible through inssider. About 6 of these were related to the school we were in. It took a fair bit of jiggling to set the channels. Even one phone with hot spot activated can cause contention. 

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...