Here's my sense story and why I returned the unit

I tell you, I never learned as much about WiFi as I did when I started overloading my network with stuff. Probably 15-20 separate sedentary devices in every corner of the house, needing three access points to keep everything on board,etc. I don’t have Google mesh WiFi, but I often consider it.

Yeah it’s a pain to have to do for a single device if it means keeping Sense online, but chances are high the Sense failure is a symptom, not the disease.

2 Likes

This…

I got lucky. I already had a 4 access point house, and the Sense ended co-located in the service closet, at the far outside of our garage, with my main router, which is also sports a WiFi base station. The Sense and our cars (4 internet cars ?!) are the only things on that basestation.

I’ve said before, I put off buying Sense because it was slightly more expensive and because I didn’t believe it could deliver as promised. There are a dozen or so monitors that will track your usage, but Sense was the only one I found who was willing to say they could identify devices. When i saw a sale price that dropped it down to level with the host of what looked like inferior products, I snapped it up.

I LOVE it. Even before the first device was discovered it alerted me to a burst bladder in my well house pressure tank. - a burned up pump for me means digging a new hole so putting if off as many years as possible is a big deal.

People love to complain about what Sense doesn’t do, but I think if we would be quicker to stop and think about how amazing it is with what it does do we’d be better off. I haven’t had mine a year yet, but I’ve already seen improvements along the way. I appreciate and respect that the company reps are available and activity working through issue and I love being in the beta program. I admit, I have my frustrations with it, but overall I’ve found it more than worth while and I’m enjoying the ride as the product gains ground.

7 Likes

Sorry things didn’t work out @dnh1263, we do hope you end up with a solution that works better for you down the road! Please keep an eye on us in the months and years to come. We’re committed to finding new ways to make this technology work for more people, and are grateful for everyone’s patience and input along the way.

Thanks for giving it a shot, and very sorry for the frustration along the way!

1 Like

It seems to vary by the usage levels through-out the day but I’m between 40-60kWh/day right now and Sense fluctuates from 8-24Kbps. Here is a snapshot of the last 24-hours – you can see when it is quieter from about 11pm-11am:


Signal level is pretty solid although there are some transmission retries, which I would attribute to just being on the 2G band.

When it does appear to lose internet access I have seen it burst up into the Mbps to backfill data. Had a 10-15min internet outage a couple days ago and it averaged almost 8Mbps for a minute or two (system displays 5min averages so most accuracy I can get).

Jeff.

Interesting. I suspect the difference in traffic volume is due to the amount of load switching that is going on, be it manual or automatically due to thermostats etc. It’s encouraging that Sense have managed to get the data volume down to this level; it seems to indicate that they are only transmitting changes, or at least using dynamic data rates in terms of what’s transmitted.

The question is whether this implies they are generating the necessary frequency domain data to characterise device switch on / off in the Sense unit rather than transmitting the raw time domain data and running the fft’s in the back end. It does imply the former, which means characterisation is not bandwidth limited, a good thing.

The catch-up performance that you’ve seen is reassuring and is leading me to think that the short data gaps that I and other have seen are not due to reboots due to suspect power (too long) or upstream bandwidth (required data rate is pleasingly low). That’s pointing to mischief in the back end, isn’t it, @BradAtSense?

This is a great discussion, thanks all!

It does seem that way, @dave_n_s, and they’re trying to scout it down even as we speak. It is frustratingly inconsistent and intermittent (as all the bad bugs are!) but everyone’s input has been useful. Now we’re just trying to figure out what the pattern is, so we can get this backfilled data working again. I’ll be sure to post an update here as soon as we know more!

I have a number of always connected devices like a thermostat, hue, smartthings, as well as Sense. All the devices are very stable and never disconnect from my wifi.

I had a similar problem with my 2 zone Sonos. I worked with tech support of and on over 4 years… As much as I wanted to love Sonos, I just couldn’t keep it connected. I decided to abandon it and went with Bluesound which used a wired connection.

I always thought it was weird that all my other wifi devices would be so stable except Sonos.

The fact is, all low power WiFi devices have connectivity issues. The problem is because these lower power devices can’t transmit as strong of a signal as they can receive. Sense is mostly transmitting rather than receiving. Your cell phone, computers, etc, are usually receiving. That’s all there is to it. There are two options:

  1. Manufacturer can find a way to send a stronger signal. There are a lot of legalities involved with this due to FCC regulations
  2. Adjust the antenna to find a better signal (I’ve had to do this once or twice)
  3. You can move your wifi access point closer to the device (I have 4 of them in my modest size house)

I have a BBQ grill with wifi that has the same problem. I ended up putting an access point near it and it solved all my issues. It’s probably using a similar wifi chip as the sense. It’s an unfortunate side effect of wireless being what it is.

@phillijw,
What kind of grill ? I have a Lynx Smartgrill and had the same issue - it’s way in the back of the yard, with lots of EM noise. A directional antenna did the trick for me. But my Sense is in a service closet that has a WiFi basestation 4 feet away and always stays connected, despite all the EM noise from several ZIgbee devices and 60Hz harmonics from the main breaker.

I have a GMG. It’s right next to the corner of the house. I ended up putting an access point about 20 feet away from it and that AP can broadcast to the entire back yard for my phone or laptop.

1 Like

A compounding problem is that the internet company support have no idea how to handle these problems, or they’re willfully complicit. When we were having trouble, they suggested… Faster internet speeds! Instead I did some learning, bought a second access point, and our problems have generally resolved.

/don’t trust your provider’s advice, seek the advice of someone with more experience and less financial incentive to screw you over.

It seems to me that a WIFI signal reading display in the settings /about sense would solve many questions if the signal was present and strong enough for proper operation. It might be nice to display the units MAC address also so one could identify it in the AP connection list. A little feedback is a good thing.

2 Likes

Brad, I have been installed about a month and so far detected gas water heater power vent, Microwave, Main kitchen fridge,Spa room wall heater, Coffee maker,Kegarator Fridge,
I can see my Mitsubishi Split heat pump / AC unit but it isn’t being detected. I noticed there is a 35W load every 15 seconds or so.(A nice little square wave running on the top of everything) I believe it is the rotation servo on my AXIS Q6045 MK II IP PTZ security camera. Is it possible that that activity could be masking the signature of other devices including my heat pump? I can disable it to test if you think it would help.

Don’t know if you are still receiving. But would you be interested in selling your unit? It is not for everyone, but I know a local non-profit that would like to give it a try but does not want to invest the $300.

1 Like

Great suggestion dave_n_s - could I take this a step further? Regarding the need for decent WiFi signal, perhaps Sense could build in a WiFi signal strength and bandwidth check into the installation process? The app already directs the user how to install the Sense step-by-step, so this could be integrated seamlessly early in that installation process. If the signal strength or bandwidth are lower than the Sense minimum recommended levels, Sense could offer a pre-baked list of tips for improving and then give the user the option to continue or pause while they troubleshoot the WiFi.

What say you, @RyanAtSense? Good improvement suggestion?

1 Like

My 10 cents here - note I work in IT, so that does color my view a bit. Sense is an interesting experiment. I don’t set my expectations higher than that. If I needed RIGHT NOW to know which appliances/units consumed what, there are more direct methods - very expensive, very time consuming, but I can get the data I’m looking for by adding a simple “kilawatt” infront of the most important devices and manually collect the data. Go a bit further and have them send the data to an aggregator and a simple web site; still a lot of time and I need to disconnect devices to make them plyable. This is why this experiment really is intriguing - but my Electronics Engineering colleges all see it as a toy that won’t measure up to real professional power tracing. I never saw Sense as that - so I’m fine with it being a “toy” from that angle.

So far, sense to me has been quite expensive. A $2000+ electricians bill (new sub-panel, re-running wires etc), sense itself and hours spent in frustration trying to get some data out of sense. Well, I got SOME and that’s pretty interesting already, but not the ones I really wanted. Our AC went bust last year and our new one is supposedly very energy efficient compared to our old unit - but we don’t see that on the electric bill after the unit was replaced. So Sense seemed the right tool (toy) for the job - if it could split out the main (big) stuff it would really help to see what caused the power-use and if it really was the AC.

This is where I have to remind me about how this technology works, and how much guess work there is - or rather, it needs a rather large customer base to have enough data to make reasonable decisions. And to be honest, I don’t think that will happen - the price point is too high (not just the price of sense). And because of that I probably have to get used to not seeing the AC or kitchen appliances non which are “standard” - rare models or at least uncommon use models will not show up in Sense’s DB and I understand that. I see it for what it is - and I take the data I can get and enjoy it. I do wish there was more, and if I don’t see some of the real large power consumers soon, I’ll have to do it the old proven way to get some answers.

If someone bought sense thinking it was complete and proven technology, yeah that’s a mistake. It’s about as beta as it comes, and until the customer base increases dramatically the DB and knowledge of devices will be limited. And if $300 as a price point seems expensive, then this is NOT the device for you - there are more costs than just that. From proper wifi, networking, electrical setup etc. - you can end up spending quite a lot before Sense will work.

1 Like

Much of this thread was before my time at Sense, so let me take some time to digest the whole thing and I’ll definitely get back to you! I just don’t want you to think I’m ignoring your comment.

I’m the original poster of this thread. I didn’t feel like digging into the panel again and removing the sense so I just left it. The times it worked, it was interesting. Not a particularly helpful gadget but at least interesting.
I got a wifi extender, programmed it and put it near the sense. Now, our sense is online MORE than it is offline. It’s still offline a lot though. It’s super frustrating when I open the app and want instant info to see how much usage turning an appliance of or on causes. Usually, the app just tries to connect and tries and tries. Sometimes it will connect immediately but usually it connects minutes later or even won’t connect at all.

Another thing that really drives me nuts is that there’s no QUICK OR EASY way to see past usage spikes by minute & hour during each day. I have an arcade (with 45 games on freeplay) that I allow my guests to use 24/7 on their own. I like to know exactly when & how long the guests use the arcade… because it’s fairly expensive to run.

So, I go to TRENDS then USAGE THIS WEEK and then USAGE BY DAY. You can click a certain day but the info here is nearly useless! There’s a small green bar graph that goes from 12AM to 12 noon and 12AM. There’s no other info! It doesn’t show hours of the day (although there are 24 individual green bars. Maybe they are the hour dividers?) or the amount of kWh used during each hour. So, you are shown a different height green bar for each hour but the height of the bars do not correspond with kWh so you really have no idea exactly when, how long an appliance is used or how much kWh were used.

HOWEVER, you can go to “NOW” (current use) and scroll backwards (literally by the minute which takes forever!) to see when spikes occurred and for exactly how long. So, If I want to see when and for how long the arcade was used last night, THE ONLY WAY is to scroll backwards (from the current time of day) under “NOW” for a painfully long time until I see the spike on the graph. One finger swipe is equivalent to roughly 4 minutes!!! If it’s 11 am and and I want to see what happened at 11PM last night, I have to swipe my phone screen backwards through 12 hours using 4 minute intervals! That’s 15 swipes per hour. 15 swipes x 12 hours = 180 swipes! That’s ridiculous.

Am I doing this wrong or is there a better way to see when and for how long a spike happens? Yes, I know I could move the sense to the panel IN the arcade but then I lose all info before the garage panel (panels in our home).