Good to see some other members from the UK!

Unfortunately, it would seem that we can't use the mesh in the UK any more, as the frequency it needs is no longer available.

I have been discussing it here: http://www.broadband-hamnet.org/hsmm-mesh-forums/view-postlist/forum-1-general/topic-966-mesh-in-the-uk.html Jack.

Sorry for my slow reply.

I don't know if there are any frequencies in the 2.3 GHz band that are still usable. It seems that the available frequencies that are left do not provide enough bandwidth for a mesh node to operate. The frequencies that are no longer available are 2350 – 2390 MHz. This includes the high speed data section.

I have looked at the 5.7 GHz band as well, but there is no allocation for high speed data in there. http://www.md0mdi.com/rsgb-band-plan-6cm/

Also, the Ubiquiti devices that operate on 5.7GHz are much more expensive than the Linksys routers, and most of them are only available from America with postage that costs nearly as much as the router!

I did read something about D-STAR potentially offering a similar type of function, but there is not a lot of D-STAR users/repeaters here due to the cost of the equipment, manufacturer lock in and closed source codec. I don't know if DMR/Mototrbo is capable of performing these functions, but there are even less Amateur/Ham users of that here.

Jack.

Remove the amp and test it again.  Amplifiers generally add more in the way of spurious noise than actual distance benefit to these systems.  You don't mention how the remote nodes are configured (antenna gain for instance).  It is likely that the remote node is hearing the local node but not vice-versa.  It could also be that one side is in a higher noise environment.  You might try shifting both sites to horizontal polarization (you can get up to 20dB in unwanted signal rejection with this technique).  I have bench-marked the Linksys routers at several miles with 18dBi antennas at each end (15 miles may be stretch though... perhaps consider a Ubiquiti Bullet M2 or a Rocket M2 if you're already using a Ubiquiti sector antenna).

Andre, K6AH

we have a remote node running v 1.0.0 software on a wrt54gs router with a relysis amp (1 watt) and ~ 16 dBi sector antenna.  this node can be seen from as far away as 15 miles with a decent LQ (~90%) but all we can see is the IP address. this is from other nodes running the same version software. we were able to connect from ~100ft but nothing beyond that. the same is true for other nodes as seen by the remote node.!

any idea what is going on?

thanks.. N7RTA

Please keep in mind per BBHN->ticket:55

It only occurs under a specific set of circumstances. Part of the dependency is also on what the internal chips are on the device (laptop,camera, desktop, etc) being used.  It takes a very specific set of circumstances for this to occur. Unless you have a proven "doesn't work" you don't have a "it does work on this version" result because you may not have all the conditions that are needed for this issue to occur.

As for differences in firmware files: negligible, the differences are around telling the router "it's safe to load this firmware" not about any function.

Hi Mike:

Serial numbers for my v1.0 affected routers are:

CGN10D345726 and CGN10D707820.

Those being the same serial number sequence as your units.... which loaded v1.1.2 successfully using (presumably) the "firmware for a non-mesh wrt54gs......

That makes me wonder if there is a difference between using the "upgrade" firmware (listed as "upgrade an existing mesh node") vs. installing the model-specific firmware (listed as "firmware for a non-mesh wrt54gs")  used to install Hamnet from scratch on a stock WRT54GS.

Bill WA8APB

Hi guys,

I've only recently discovered HSMM & have just bought my first router, a WRT54GS v1.0

I've flashed it with v1.12 firmware & it connects via LAN just fine.  I can't vouch for any of the wireless functionality as there's no-one else within miles of me on the mesh-map.  :(

According to the Wikipedia entry for the WRT54G family, the GS v1.0 had serial numbers beginning CGN0 & CGN1... mine begins CGN1... maybe the bug you mentioned was fixed after the CGN0 models?

How do the s/ns begin on your two GS routers, Bill?

73,

Mike

M0MLM

I missed cave city, are you going to set up at the greater louisville hamfest?

Thank you for confirmation on the WRT54GS, this model of the hardware seems to be the only model not affected by the issue with your 2.0 and 2.1 report and a previous V4 report. Your report that a WRT54GS V1 doesn't work is also interesting and let's is know it's not the entire WRT54GS line that works.

A fix will trigger a new version number, one should not expect us to slip fixes under a version that has already been released.

Per BBHN->ticket:55 this has been traced to a flaw in the on board Linksys switch module that shows itself under certain circumstances, Circumstances that just happen to be present in the 1.1.x releases to date.  We do not have the ability to bug fix the switch itself because it is a closed module that just so happens to be inside the same plastic case of the router.

This problem is related to previous post about "One PC discovered... other not".

I upgraded three WRT54GS routers with no problem (Two V2.0, one V2.1).  Upgrade process proceeded normally, and everything worked when upgrade was complete. 

I then upgraded two WRT54GS V1.0 units, and experienced identical DHCP problems with both of them.  The upgrade process appeared to go normally; however, I found that when I setup the router with node name and password, and then rebooted, the DHCP function stopped working.  The DHCP screen showed no current DHCP leases.  Both routers refuise to talk to anything plugged into the LAN ports... to include the PC used for the  upgrade (using  USB to Ethernet adapter), a tablet (also using a USB to Ethernet adaptor), an ATA (reflashed BasicTalk Grandstream), and an IP Camera.  The PC used for node control posted an error message stating "connection has little or no connectivity". 

I re-flashed one of the problem routers with the V2  upgrade firmware, communicating with it through another node... and that process appeared to work OK; however, result was the same.  As soon as I rebooted following entry of node name and passowrd, the LAN ports stopped working. I could still log into the node from other nodes, but not from any client connected to the LAN ports.

I tried to configure DHCP Address Reservations manually for one of the V1.0 routers by plugging in the Name, IP address, MAC address, etc for an ATA and a Camera... and the router accepted the manual DHCP reservations; however, router posted an error message saying "Configuration saved. However, problem with port setup."  Clients could still not communicate with router via the LAN ports.

It appears the V. 1.1.2 firmware is not yet compatible with the WRT54GS v1.0 routers.

Bill WA8APB

First I suggest a good book on networking like Cisco books but there many others.

There are good books on all your questions! Just google your questions.

Im told one can do anything on a mesh network that you can do on any TCP/IP network!

One can sec up severs like web,file and ftp.

Agood place to start is with the book "wireless networks for developing countries" a free web book that comes in a PDF file.

It will depend on a lot of factors.

For example: 

Local noise floor at each end.  If one end is remote and has no noise then only the onboard receiver is the limit however if the remote node has lots of interference ( even from a single wifi device on the same property)  this end would become the limiting factor for a link.

Path the signal is taking: I once had a 20db SNR however we had a tree in the path so we would see the station and it would disappear for 10 seconds, return for 1 second and repeat (you can't see this on the simple interface ) Similar synptoms could occur from a bouncing signal.  If the either end is moving in the wind this can affect the aim as well as the antennas blow out of line in the wind.

Hardware at each end -  diversity receive has a better chance to work than a single chain device. I have heard a report of a 20db dish with a bullet could be heard by a NanoStatiom 12 miles away, but the bullet couldn't hear the NanoStation.  Put a NanoStatiom in place of the dish (a 10db reduction in antenna gain but an adding of a second polarity) and both sides could hear each other.

the fact you are able to see the node is a good starter however. It may be as simple as a fee degrees of aiming required. I've heard stories where units were aimed towards a peak signal but it turned out to be a peak off one off the side lobe of antenna. And I've heard of cases where the peak was a bounce off of some surface and the antenna was getting a strong signal but because it was a bounce it would be unstable.

Another fact that comes up is frensel zone.  I've experianced this myself for the first time recently where 4 feet of extra height actually caused me to loose the signal.  It can be counter intuitive but it does happen.

So in the end there is no "this level is always guranteed to link" signal level.

amazon now has a 18dbi yagi for less than $25.00 compleat with cable and connector for Linksys mesh node.

amazon now has a 18dbi yagi for less than $25.00 compleat with cable and connector for Linksys mesh node.

Good evening everyone.

Glad to see the weekend is finally here. I went out wednesday with my Ubiquity 5G25 on a small portable military tower with an inverter and laptop. I setup my equipment on a near by mountain top and started to scan. I knew a fellow ham had a node up on a local tower and I wanted to see from a mountain top to the tower some 14 miles away if I could connect.  Both nodes on the default SSID I could see his node  with a signal level of  -87 at best. I flip flopped from horizontal to vertical to get my best signal. the two of us never connected.

What is the minimal signal needed to connect and make contact ?

Mike