Urbana, Illinois, Now a goTenna Mesh Ambassador City!

[MikeL]

Chatted with a prospective node host this afternoon and it looks like UMESH 9 will also be going up in the next couple of days during this spate of nice weather. A great way to celebrate my 1 year anniversary in the mesh community!

UMESH 9 will add some redundancy in an area that needs it.

I think the last 10 days may have dimmed out at least one node. Short days and a persistent really thick cloud cover may have taken their toll. I’ll be checking on things, as well as putting up the 2 new nodes between now and Sunday.

[MikeL]

It was a frustrating day in some respects, but we made a lot of progress in the end. Got one node up successfully (in the end). Ha dto delay

First, went to set-up UMESH 9. The node host took it up and tied it down on the chimney. He remarked on the great view towards the north and took a few pics, so may have some of that to post soon. Usually I like to spend a little more time with a new host and make sure the node is all squared away, but I was a little pressed for time because of scheduling two nodes in one day. I had trouble connecting home and decided to see if I could solve for things before the second appt of the day. A check was somewhat inconclusive but pointed toward the home relay and suggested loss of power during the recent gloomy period. I got it back in relay mode eventually, but it still seemed flaky at that point, so I vowed to come back later and sort it out as best possible.

A brief time later, I arrived at the site of UMESH 8, talked with my climber and we jointly decided the ladder just wasn’t going to be long enough. The node host has gutter cleaners coming, so we came up with a plan to ask them to set UMESH 8 in place. It remians pending, but will be done soon.

Still, wasn’t a complete bust, as after a long chat about what the mesh was my climber suggested another potential node site in a great location that he will check on and let me know. I’m not calling it UMESH 10 yet, but we’re very close. With 8 nodes up, we’re now 25% of the way to our goal of 32 nodes to mesh Urbana. With 3 more (one site in hand, one possibility, plus one more) we’ll be 1/3 of the way to our goal.

After supper, I headed back out to determine whether or not the home relay was up. It sure was and a drive around the mesh showed that the newly minted UMESH 9 seemed to be functioning well, I noted with some relief.

[MikeL]

After some analyzing and testing, I narrowed the potential problems area that was causing a lack of connection from north to south down to two nodes and checked them both today. Happy to see a nice full battery pack and a live goTenna Mesh relay in UMESH 3. Found a full battery at UMESH 2, but it apparently lost power in the depths of the our long nights and needed a reboot. Then things were working well in the brief time I had for testing, but will do more testing when the local snow event subsides over the next 48 hours that is just beginning. For now, I will keep the coverage map the same, but will update if I find expanded service to the southwest which is possible.

Was just thinking, there was one difference between the two relays that might have made the difference and kept the one on the air when the other one lost power. UMESH 3 had a 6 W panel, while UMESH 2 had a 2 W and a 3.5 W panel paralleled together for a total of 5.5 W. Did the 1/2 watt difference make a difference? Looks like it, but would require more testing to confirm. However, for coping with the short winter day lengths around Latitude 40 North, this does suggest that 6 W might be as low as you’d want to go to avoid those pesky service interruptions when relying on solar panels.

[MikeL]

It was a gloomy day when we put UMESH 9 up, but these are still nice pics of its place up top of everything. The host of this new relay node was just sizing it up for how to tie it down when he snapped this pair of pics I’m sharing with their permission.

Pic courtesy of UMESH 9

As you can see, solar power is catching on in the Midwest. Of the 8 solar-powered nodes on the air now, two of our hosts have decent sized solar arrays. There are some real possibilities for sharing data between solar arrays with goTenna Mesh.

Pic courtesy of UMESH 9

These pics remind me I need to finish the wood before final assembly in the next group of nodes built to this design. We have nodes looking for homes right now, so if you’re in Urbana, let me know through a Direct Message here or by texting me on UMESH at 92290810565183 and we’ll see what we can arrange, despite the weather. The vital linking relay node I fixed yesterday is now a hump under 6"+ of snow and it still gets the signal through. Right now I’m looking forward to the sun coming out and clearing the solar panels off.

And thanks again to the great folks at goTenna who are supporting the build-out of UMESH by providing the essential ingredient for 21st century off-grid communications, the goTenna Mesh.

[MikeL]

I had a chance to go out and do some network evaluation today, mostly to check on the effects of the 8" of snow we received over the weekend. When the snow first came down, it’s effects were very minor, mostly involving needing to resend a message where it was typical to get a confirmation returned after a several hop connection.

Things aren’t so good two days later. Apparently, the melting of the snow compacted it, which had a bigger effect on RF signals getting out consistently. I was able to ping on second try in places while needing to retry the send message 5 times or more before I got a confirmation on some links. Others were mostly unaffected. I suspect the way the signal goes and the position and thickness of the snow on the waterproof container makes a difference.

Be patient with things. We’re hoping the melting gets the snow off the solar panels on the relay nodes before battery life becomes an issue.

[MikeL]

I had a chance to check to see how we fared through this miserable week for solar energy locally. I’ll just say there seems to be significant network degradation. Right now we’re facing Winter Storm Warning for the next 24 hours, 4 to 6 inches of snow with winds to 40 mph.

This provides a great opportunity for someone with rooftop access somewhere in the twin cities to offer a helping hand to your fellow meshing citizens. If you can safely brush snow off a node, all the better! We’ll persist.

[MikeL]

Over the last couple of days, there are indications that the UMESH network is not completely down. I was downtown at a meeting and reached home on Monday evening via 1 node. I wasn’t sure whether it was the Meshmobile that might have accomplished this, since it was sitting just outside. But it wasn’t particularly well situated to reach home, so I suspected UMESH 4 on the IMC may have been involved. Always hard to tell with mesh, but it was working.

Yesterday evening, a node host reached me from downtown with a 2 node link. That proves at least one node other than the Meshmobile is working and I suspect it is UMESH 4. Yeah! The signal may have hit the Meshmobile and not the home relay node, can’t tell right now. So it’s possible that another relay node or two may still be functional.

This does tend to again point to the advantages of installing a GTM on your vehicle. It’s good backup in an emergency.

That said, the weather isn’t looking good for significant work to reboot the downed nodes for several days ahead. The 1" or so of rain, freezing rain, etc isn’t planned to stop until we get back into the freezing category later today over what’s left of a frozen crust of snow.

[MeshTheWorld]

Mike, snow and ice accumulation are very real problems for those operating solar powered relay nodes in regions prone to below freezing temperatures.

I noticed that in some of the photos you shared earlier, the solar panels look like they are horizontal on top of the dry-box containing your battery. If you keep the solar panel inclined at a 70 degree angle above horizontal, that should help reduce the accumulation while maximizing the power output during December in Illinois. It will also help clear the panel faster once temperatures go above freezing again.

I’ve been contemplating going with a higher capacity battery for my stationary relay nodes to account for an extended duration of poor solar panel performance, considering the reality of snow & ice accumulation and the fact that it can (sometimes) take weeks to melt and restore the normal power output of the panel. What type, size, and brand (if known) are you using? Have you standardized all your setups, as far as the battery goes?

[MikeL]

This has been a more typical winter than recently, so we’re still learning about how to best deal with it.

Most of the nodes here do sit with their panels at an angle facing south. A few just happened to work out better with a horizontal orientation so I am experimenting with that. One reason is that this leaves the GTM inside with a good upright 360 degree view as it keeps the panel on top from shielding the signal. I’ll know how this performed versus the standard ones when weather get better. I do know that a tilted panel will often clear faster, but not always. With the kind of snow we had last week, it managed to form a dome even over the tilted panels that was about 6" tall, which is why the system reboot is needed (with a couple of possible exclusions.)

This is the battery pack I’ve been using with good success:
https://www.banggood.com/8000mAh-Ultrathin-Solar-Battery-Charger-Power-Bank-For-iPhone-iPad-Tablets-Smart-Phone-p-1217860.html?rmmds=myorder&ID=228&cur_warehouse=CN

It says its capacity is 20,000 mAh on the back, they advertise it as being 8,000 mAh, but I found the battery inside is rated at 7,500 mAh. Whatever. It will run a GTM for about 6 days if fully charged. You can charge it and power a device at the same time. When it senses a load, it turns on. All the nodes here have these and they’ve proven to be reliable.

[serendipitysal]

Wow…what an amazing achievement . One word Legend . What an inspiration Power to the Mesh community

Sending love and light from the sunny south coast.

[MikeL]

Thanks for the appreciation. I enjoy building community infrastructure. Been doing that a long time. I will say that I’m optimistic of the enormous potential of the mesh, largely because it’s a shared effort, not mostly all about my vision even when I am the initiator of the network locally.

[MikeL]

The host of UMESH 9 got the relay node there restarted and initiated some testing this morning. It’s about 10 degrees outside ! That’s a heat wave after the -17 below zero temp that things bottomed out on yesterday.

We tried some back and forth and he was able to receive via 3 nodes a message I sent. I wasn’t able to get any messages from him confirmed. Still this indicates at least 3 nodes are on the air still: the Meshmobile; UMESH 4 downtown at the IMC; and UMESH 9. At least those are my best guesses as of right now of what appears to be working.

There may be more. While network performance is significantly degraded, it’s not completely down, a very good sign considering recent weather here in the Polar Vortex Zone. I’ll do some survey work tomorrow to try to pin down what’s working and what’s not. It’s supposed to push into the lower 50s by the end of the weekend, with rain, but we may get a chance to get things rebooted.

We’re also testing a clip device to hold the power button down so the unit will restart and relay messages in the event of a power loss. I’m not sure it will work, but we’ll give it a try.

Finally, I installed some extra flashing amber warning lights on the Meshmobile to help make things safer when I’m out doing survey work in order to improve safety.

[MikeL]

Went out to survey where UMESH still has signal. It’s not good, but could be worse. UMESH 4 at the IMC seems to be up and that’s huge, because of it’s height AGL and central location. It’s installed flat in the case, but uses the tilted panel design of the other early nodes.

That and the wind exposure must’ve swept enough snow off to keep things properly charged.

Before I left, I threw the node that’s ready for UMESH 8 (whenever the weather allows work there) with it’s GTM “Powerclipped” inside up on the patio roof. It’s a bright, sunshiny day and the melting has started!

But not nice enough to crawl up the roof here to reboot UMESH 1 yet. So the temporarily spare node isn’t very high up, but seems to be working OK. And we know UMESH 9 is up. Include the Meshmobile and we have 4 relays up right now. I hope that we are able to get to most to check and reset them by the end of what is supposed to be a warm weekend. I also want to verify that the emergency stand-in node with the “Powerclipped” GTM is working as a relay.

[MikeL]

OK, UMESH is mostly back in service. These nodes are confirmed up.
UMESH 1
UMESH 2
UMESH 3
UMESH 4
UMESH 9
Meshmobile

UMESH 5, 6, and 7 still require intervention, we hope in the next few days.

If you need a node to assist you in getting on the mesh in Urbana, send me a message at 9229 0810 5651 83 and we will see what can be done to help. Parts are on hand to build 4 more, plus we have one on hand that’s awaiting a chance to get lifted onto UMESH 8.

When the mesh came back up, I was happy to see a new user checking in to say hi, and it was even an old friend who’s now joined the mesh.

[MikeL]

Thought I’d check in, although relatively little to report. The flashers are on the truck, so start and stop survey work should be safer. Weather has been pretty marginal, but today wasn’t bad…except for the mud. I continue to make new contacts and reinforce old ones, so when the weather does break, there will be several nodes going up. This should result in a nice enhancement of mesh coverage in several directions.

Also neat to see an old friend who recently joined the network experimenting with some antenna enhancement in the form of a parabolic reflector. Seemed to work fairly well, but with a narrow aperture. I had no luck pinging him back, which almost always works when nothing else does, and had to repeat a couple of messages, which did get confirmed. He may have been able to get into the network easier than it’s able to get back to him. Not sure exactly where he’s at these days, other than he’d made it sounds fringish to our current coverage in UMESH, so maybe he’s working on a solution? Maybe even better, he’ll post a pic or two? He’s an old hand at this as he was involved in some of the early mesh networking experiments here starting back around the turn of the century. It’s good to see people finding UMESH useful and interesting.

[dyoung]

Hi Mike,

I’m near the corner of Race and Pennsylvania, so about six blocks south of the UMESH unit near Race and Washington. I have one of my goTennas in front of a parabolic reflector that I put in a second-floor window. It is in relay mode. My second goTenna is ordinarily in my office on the same floor.

I banged together my reflector using some plywood, nails, a cardboard box, and aluminum flashing:

You can see the nails on the inside of the parabola—there are nails on the outside of the parabola, too. The nails are essentially all that give shape to the flashing. The flashing wants to droop, so I put some folds in to give it a bit of rigidity. You can see in the photo that it’s just barely parabolic at the top.

I took a guess about how to orient the goTenna at the focal point of the dish. If I guessed wrong, please let me know!

It’s hard to say how well the reflector is working, but I do seem to get messages through more consistently with it than I did without it. It would be nice if I could get more quantitative information about links from the goTenna.

Dave

[MikeL]

Dave,
Thanks for the report and the pic. This could come in useful for others, too, as it is simple and easy to make; if you’re going to be at the edge of a mesh network and need a little more oomph for a reliable connection, then the fact that it may somewhat reduce the ability to relay off the backside of the parabola is a worthwhile sacrifice.

As for stats out of the goTenna, it’s just the flashing LEDs telling you it’s relaying. No servers or other central control, no room for it in the bandwidth, so we like to think of that as a feature rather than a bug.

Now, some of the SDK options allow some basic numbers as is allowed within the FCC mesh framework, but you’d be far more comfortable just wading into those than I would be advising you on that.

When we were at UMESH 5 rebooting it, your message came through directly, no hops needed. Here at home, I can reach you via either 2 or 3 nodes. UMESH 5 is in between us and you between you and downtown, so you should see better reception because of that, too.

In both cases, the battery that couldn’t keep up with gloomy Midwest conditions proved to be the 4000 mAh Voltaic battery packs. They’re good reliable packs (nothing wrong with the gear, I just spec-ed too small), but need to be one size larger to deal with the local conditions here. Now it seems that the nodes that didn’t last back in Dec-Jan snow were more to do with having those smaller battery packs. I think I feel better about the basic design being adequate with the 7500 mAh packs I’m using as the standard ones. A few other designs tweaks and we may have it as far as snow/gloom.

That’s now as best I can see why we had the partial coverage with part of the network OK and the other part going down.

While UMESH 9 (thanks for the help today, BTW!) and I were out rebooting today, we got UMESH 5 on the air, where Dave caught up with us. We also got long suffering UMESH 6 up and running. The only node that is possibly down is UMESH 7 and it has the big battery so may be fine when I get the chance to check on it.

Users should see coverage pretty much as depicted on the map.

Thanks for being patient with things and we hope that what we’ve learned will limit the possibility of such an extended down time as some must have suffered. We also welcome any technical assistance someone may be able to offer with small low voltage solar design. I mostly bolt things together, but am inclined towards more experimental approaches if off the shelf fails me. I would value any insights offered, although I’m getting a few basics OK I want to make sure I’ve got a good design that is inexpensive and reproducible.

Also remember that UMESH 8 is PENDING getting it in place. Keep your fingers crossed as we hope to have it and some other nodes up as soon as conditions permit. If you are interested in being a node host, send me a DM here or text me at the usual 92290810565183

People seem to be making use of the mesh as a community resource. The interactive map above is just now nudging over 2,000 views!
And that’s not all, it’s just the places where people are officially hosting a node as UMESH infrastructure. You also can see that new nodes are popping up locally on the main goTenna Mesh Community Map. Some are mobile, some are fixed and always on, all are always relaying for each other.

And thanks to Dave again for his interesting and useful contribution!

[MikeL]

I had a chance to do some survey work late this afternoon and found a significant expansion to the SE now that UMESH 6 is staying on the air. It is a well-located node and the land falls away in this direction allowing for long LOS connections. There’s few people over most of the newly added coverage area, but it is an example of how one well located node can expand your coverage area.

On the edges of coverage I was having some trouble getting good pings along with the confirmed messages, but the coverage is there if you’re willing to work a little for it at times. I set a new record connection length at about 5 miles from Curtis Road SE of Urbana to UMESH 1.That’s the way it is when you’re reaching home on 5 or 6 hops, it’s pushing the envelope, but rather remarkable considering the low power and omnidirectional antennas.

I believe that UMESH 7 is currently down and will try to get that addressed soon. UMESH 9 also seemed quiescent this time around, but that’s an easy one once the most recent snow melts off.

[dyoung]

goTenna and BioLite Campstove, two great Brooklyn products that go great together?

The BioLite Campstove burns biomass without smoke using preheated air and draft provided by a fan. Its thermoelectric generator powers the fan and a USB port and charges an internal battery. The USB port supplies about 2 Watts—the latest model supplies more.

The other day I used my BioLite Campstove to simultaneously boil water for coffee and charge a goTenna. Some nut shells, wood pellets, and American sycamore twigs powered my coffee-making and battery-charging:

Here I have the kettle on and the goTenna plugged in:

It took no more than 4 minutes for the goTenna battery to charge from 83% to 100%.

David

[MikeL]

Dave,
That definitely wins the Unusual Power Source of the Month Badge