Sorely Disappointed in UPB

[chambro]

Hello All-

This is my first post on this board, and unfortunately I'm not sharing good news. After months of researching, I decided to go with UPB for the new home we're constructing next door to our current house. I received my first order of about a dozen switches (mostly SAI US2-40's) installed 3 of them in the new house and fired up UPStart to test it all out. The first screen that popped up told me I had severe noise levels (>28) that will likely prevent the system from working. The noise was so bad, I couldn't even create an initial network, let alone see any of the switches. The house is in its last stages of construction and is unoccupied, so although there aren't many appliances running, there were several carpenters there and I figured maybe some of their tools could be causing the problem.

I came back after everyone left at about 6pm, shut down all the flourescent and low voltage lights, and tried it again. Althought the noise dropped from 28 to 22 with the "noisy lights" shut down, it is still severe and unusable. I decided to see if I could isolate the circuits generating the noise, so I took my laptop (running on batteries, no AC power supply) down to the circuit breakers in the basement. It's a big house (about 8000 sf), so there is a main panel, 2 subs, and 6 "subs of subs". There is also a generator with an automted transfer swich. I plugged the PIM into the outlet coming off of one of the subpanels, opened the UPStart Noise meter, and started flipping breakers off. The initial reading in the basement was 18, a little lower than the main floor. For the most part, each shut breaker had no effect on the noise level, but to my great suprise, 3 cirucuits actually caused an increase of about 2 counts each in noise as they were shut off. By the time I was done, I had every single breaker in the entire house off execpt for the outlet the PIM was plugged into (it is the only item on that circuit), and my noise level was now 22 instead of the initial 18. I also completey shut down the generator all of its circuits, with no impact either. I'm assuming I have a major external source of noise.

I came back the next afternoon (the trades were all there with their saws running) and now, for no apparent reason, the noise level was down around 6. I was actually able to create a network, configure the switches, and create some links (required a lot of retries). The link was working about 4 times out of 5 immediately after I set it up. I came back around 5pm (after everyone left) and the links did not work at all; fired UPStart up again and noise is back to about 20.

I tried this out in my current house next door and got similar results; severve noise most of the time with windows of barely usable quality. Obvioulsy, this is not even close to workable.

I've heared about these prototype and/or home built "whole house filters" that may help with external noise sources. Does anyone have any knowledge or experience with these? I'm a little reluctant to possibly throw good money after bad if this is a low probability fix. Any other suggestions on dealing with external noise (except poking around inside all my neighbors houses!) would be greatly appreciated. I'd really like to make the UPB work, but based on the results so far I am not optimistic and am seriously considering buying a few Z-wave switches to try out.

Thanks everyone.

 

[nov0798]

Is the house currently being run off of a generator, or regular power?

Any possibilty the unit your using to measure the noise is in fact bad?

 

[123]

UPB is just the messenger; no need to shoot it. If I was in your shoes, I'd be disappointed with your electrical supply and not with UPB. UPB is clearly indicating that your AC supply is too noisy for proper operation. Unless that situation is mitigated, all PLC-based lighting will be disappointing. RF or hard-wired technologies may be the viable solution.

 

[fcwilt]

I've got about 50 some UPB devices in my 9000 sf home, 3 main panels, 2 sub panels, generator.

I have no problem with noise. I have a few UPB units on my dock as well which is about 250 feet from the house. The signal level there is a bit on the low side but still reliable.

I do have a number of filters on UPS's, computer-like devices, etc.

Best of luck.

 

[Steve]

I know exactly how you feel and what you are going through. It's no fun and very frustrating. Unfortunately externally generated noise is the toughest one because you have no control over it. But to try to figure it out, I would look at a few things - is there other construction going on around there where it could possibly be some temporary construction equipment, perhaps lights, generators or some other semi permanently wired construction items? What about the noise - can you find a pattern to it, is it cyclical or time based? Like does it get strong for 30 minutes once or twice a day then go away or get low or is it there pretty much all day long to various degrees? You can setup your laptop to log it for as long as you can. What kind of utilities do you have in the neighborhood - underground or overhead, do you know where the transformer that feeds your house is? Do you know exactly which houses that transformer feeds? If you can, take a PIM and laptop to some neighbors. Compare your readings in a house on same transformer to a house on a different transformer. If the noise is only present in the houses on the same transformer you've at least narrowed it down to a few houses, or the transformer itself. Sounds like a new neighborhood, but an old transformer that is breaking down can itself generate the noise.

Your system will always be most reliable if you can find and eliminate the noise. Short of that you unfortunately have Gen I hardware. I don't know why SAI is sticking to their guns on it, but Gen II is proven to be more tolerant of noise and I have proven this myself. SAI/Gen I stuff will start having alot of trouble past around 5 'Lesters' (term coined from the UPB inventor - Marshall Lester) of noise. Gen II stuff has been shown to work decently up to around 25 Lesters of noise. PCS, HAI and Web Mountain (SAI hardware with Gen II firmware) are all Gen II. In a house your size its a bit more difficult because you may need phase couplers and/or repeaters to help things along. It's also a little too big and complex to do breaker/phase alignment. But all of those things are almost moot if you have 28 Lesters oif noise almost constantly throughout the day.

There is a pool company in Cali that has some sort of attenuator that does cut down noise dramatically and it has no amperage rating so it may (I have no idea really) be able to go in the main or subpanels. It was really designed for 220V variable speed pool pumps like the Pentair Intelliflow that generate noise on the line. Brandon at Automated Outlet has info on it or I can track down the phone number somewhere. Whole house suppressor with emi and transient filters don't help the noise.

So, if you wanted to dig a little more and try to make UPB work, the first thing I would do is forget about the SAI Gen I hardware and go Gen II. I personally prefer the PCS stuff. Yes, it is a little more expensive, but it is high quality Gen II hardware. PCS will also give you a money back guarantee if it won't work. You can try wherever you purchased your SAI stuff, they should take it in as a trade or again Brandon or Martin at Automated Outlet can hook you up with PCS.

If you decide to give up on UPB and I wouldn't blame you, I'm part way there myself, although with PCS equipment and some phase aligning my system is not too bad right now, you have either wireless or hardwired to try. I imagine its too late for hardwire especially in your size house at this point. So then you have ZWave, Zigbee and some proprietary wireless. ViziaRF, RadioRa and Ra2 coming and Homeworks seem to be the highest quality. I believe the Vizia stuff is ZWave compatible but RadioRa and Homeworks are proprietary. I would also not rule out Zigbee. While there are not HA Profile 'standard/interoperable' devices quite yet, there are some nice other systems that can be used such as Centralite Jetstream. They still have an issue with the feel of their paddles, but that should be worked out in a month or so and should be no big deal.

Do you have a central controller, like an Elk M1 or HAI OmniPro? If so, that may dictate a bit what system you go with. Also keep in mind that with a controller in the mix you can have a hybrid system of several of the technologies and they will interoperate via the controller. Depending on how much you and the family put form over function will also help choose a tech. As an example, with ViziaRF (and maybe some others) the switch paddle only clicks in one spot, the bottom, so it is not a typical rocker like the UPB devices. Jetstream is also a toggle but it is more of a flat plate than a half rocker like Vizia. Then there is the LED - green or blue, the feel of the switch, etc. All those may be factors in your decision.

Good luck and keep us posted with your results. And, Welcome to CocoonTech!

Edit: As usual, it took a while to type this. When I started there were no replies, so some of my thoughts may now be duplicated with prior posts (have not read those yet).

 

[drvnbysound]

Is there anything to test for this noise, before purchasing the products, other than an oscilloscope?

 

[Digger]

If you buy a PIM and use the free Upstart software that you can download you should be able to check your lines for noise. If you also buy one plug in module you should be able to check various circuits in your home for signal strength.

 

[chambro]

UPB is just the messenger; no need to shoot it. If I was in your shoes, I'd be disappointed with your electrical supply and not with UPB. UPB is clearly indicating that your AC supply is too noisy for proper operation. Unless that situation is mitigated, all PLC-based lighting will be disappointing. RF or hard-wired technologies may be the viable solution.

To me, your perspective is a little strange. UPB is not a "messenger", it's a technology product I paid good money for based on the performance claims the manufacturers made (namely it is very robust against noise). Your statement is like saying "The top on my new convertable leaks when it rains, but I'm NOT disappointed in the car, I'm disappointed in the weather." A reliable product needs to work well in the range of environments in which it will be used, not just the "perfect conditions".

I understand that many folks have no noise problems with their UPB setups, and I am very happy for them. But reliability is not about the folks who don't have problems, it's about the ratio of folks who do have problems to those that don't. Going through this board since I've had my issues shows me that there are a significant number of folks with UPB noise issues, with severity ranging from somewhat annoying to completely unworkable. Obviously, this annecdotal data is not good enough to calculate accurate reliabity statistics, but I believe it is reasonable to conclude that this technology is NOT yet even close to "reliable", especially if you use a traditional six sigma standard prevalent in the electronics industry.

So am I still disappointed in UPB? Yes. But I haven't given up on making it work. There seems to be a lot of us in the same boat, so hopefully there are some solutions out there.

 

[chambro]

Thanks Steve, your post is helpful.

I did not realize that SAI is still using Gen I hardware; I thought everyone switched over. I do have a couple of HAI switches (Gen II) that I can install and try, but as you stated, with the noise levels I'm seeing it may not matter.

I did purchase my hardware from Automated Outlet, so I'll give them a call next week about the whole house filter you suggested.

I'm using an Omnipro II controller for my communicating thermostats, so I planned on using it for everything. If I can't get the UPB working, what would be the next most compatable lighting control technology for Omnipro?

 

[Steve]

... it's a technology product I paid good money for based on the performance claims the manufacturers made (namely it is very robust against noise)....

... but I believe it is reasonable to conclude that this technology is NOT yet even close to "reliable", especially if you use a traditional six sigma standard prevalent in the electronics industry...

Well, I understand your frustration and I'm right there with you but I can't really agree with these comments.

First, there is no way SAI can claim their devices work in a noisy environment while they are still using Gen I, Gen I has no sensitivity settings or anything at all to deal with noise. In fact, their stance is that Gen I works great but what you need to do it find the source of the noise and eliminate it - not always an easy or even possible choice. At least Gen II does try to work around noise and PCS WILL back up their claims or offer your money back.

Your second statement kind of is misplaced in my opinion. While I have issues with it too, I am not saying UPB is not reliable. In fact I still think its the best PLC system out there. But sever noise will kill anything and right now there is nothing that deals with it any better than UPB. It's got the muscle to power through alot but no, its not perfect either. If you have an outside influence that it totally corrupting the signal, that does not make the technology unreliable.

Is satellite tv unreliable when there are atmospheric conditions that block the signal? Is fm radio all of the sudden unreliable if I erect a solid steel building in the path of the signal? Are there ways they may be able to make UPB more immune to noise? Maybe but nothing is perfect for all conditions. If you can't get UPB working reliably then it's time to move on to another option

 

[Digger]

SAI advertising is a little misleading. While any UPB device is more immune to noise then X-10 or Insteon and will work better in a noisy powerline it still can have problems. GEN II is even better than GEN I but even that may have problems at times.

For power line carrier technology UPB is the best. Zwave is RF so it is different. For some installs it is better and others not. Hardwired is probably the best of all but it is expensive and not always practical.

 

[signal15]

I had massive amounts of noise on my wiring in the house. I couldn't figure out what was causing it. About 2 months later, my green box (the pad) outside with the transformer started buzzing and popping. I would get power surges in the house that would burn out light bulbs, pop GFI's, and it fried my coffee pot. A couple of days later, the thing started bulging and smoking. The electrical guy that came out almost crapped himself, said it was about to crater, and had to shut the power off to the entire neighborhood while they replaced the pad.

It could be your transformer. And if your current house is seeing the noise problem also and you're off the same transformer, this could certainly be the issue. Unfortunately, if this is the problem, I think you're going to have a really hard time getting the electric company to replace it. You may have to lie a little bit and tell them you see 180V power spikes every couple of days, and that you'll hold them responsible for any fried computer equipment (they paid for my fried stuff). I couldn't even get them to come out until I told them mine started smoking.

Once they replaced the pad, my noise problem went away.

 

[123]

To me, your perspective is a little strange.

Not strange; just not your perspective.

.. is like saying "The top on my new convertable leaks when it rains, but I'm NOT disappointed in the car, I'm disappointed in the weather."

Can't blame the car if it leaks in a monsoon ... and that's the equivalent noise on your power lines. You need to fix that.

... needs to work well in the range of environments ... not just the "perfect conditions".

Agreed. Unfortunately, your electrical environment is 'out-of-range' and far, far from perfect. You need to fix that.

... I believe it is reasonable to conclude that this technology is NOT yet even close to "reliable" ..

All technologies have an operating range (temperature, pressure, voltage, current, etc) and, unfortunately for you, the background noise on your powerlines is outside the operating range of UPB. Which, again unfortunately for you, is one of the most noise-tolerant PLC-lighting technologies (especially Gen II).

...I haven't given up on making it work.

Excellent! Home Automation requires a lot of persistence. I use X10, the least noise-tolerant PLC technology, and it took boatloads of persistence to make it work (well, most of the time). PC power supplies, UPS devices, wall-warts, fluorescent and compact flourescent lighting, motors, LED nightlights, etc are the Usual Suspects when it comes to generating electrical noise. If you unplug everything and still measure a high-level of noise (and that appears to be the case) then the source is external and probably out of your control. The only option is to suppress the incoming garbage and I have no experience with whole-home noise-suppression devices ... a bulwark against the onslaught of electrical crud. Sounds like an expensive, difficult, and ongoing battle. My concern would be what happens if the suppression device fails? The noise comes through and renders your UPB investment unusable. Kind of like when my X10 repeater/coupler has a spasm and floods the lines with traffic ... nothing works until you power-cycle the repeater.

RF technologies, like zwave, might be your best option.

 

[AnthonyZ]

To me, your perspective is a little strange. UPB is not a "messenger", it's a technology product I paid good money for based on the performance claims the manufacturers made (namely it is very robust against noise). Your statement is like saying "The top on my new convertable leaks when it rains, but I'm NOT disappointed in the car, I'm disappointed in the weather." A reliable product needs to work well in the range of environments in which it will be used, not just the "perfect conditions".

I understand that many folks have no noise problems with their UPB setups, and I am very happy for them. But reliability is not about the folks who don't have problems, it's about the ratio of folks who do have problems to those that don't. Going through this board since I've had my issues shows me that there are a significant number of folks with UPB noise issues, with severity ranging from somewhat annoying to completely unworkable. Obviously, this annecdotal data is not good enough to calculate accurate reliabity statistics, but I believe it is reasonable to conclude that this technology is NOT yet even close to "reliable", especially if you use a traditional six sigma standard prevalent in the electronics industry.

So am I still disappointed in UPB? Yes. But I haven't given up on making it work. There seems to be a lot of us in the same boat, so hopefully there are some solutions out there.

"I have a nice new Ferrari that handles terribly on streets with potholes. I chose to live in Colorado and hence, suffer from a lot of potholes every spring. Ferrari should have foreseen that I would use the car in poor conditions and made the product ready for anything. I mean, what were they thinking..."

Obviously, this example is ridiculous but, you understand where I'm coming from. You have a problem with noise. Your UPB devices did not introduce the noise. The manufactures can't be held accountable for environmental conditions that exist in a small minority of installs.

Also, you did your research on the product, at least enough to be able to state, "...it's a technology product I paid good money for based on the performance claims the manufacturers made (namely it is very robust against noise". Despite that research, you still chose to go with the vendor who has undeniably done the least amount of work to ensure their product works in a wide range of installs. In my opinion, you're blaming the product for poor performance when at least some degree of the responsibility falls squarely on your shoulders.

I'm not trying to be a a$$ but, I think you've misplaced your disappointment...

 

[refinished]

FWIW - I have a rock-steady UPB network in my house - almost no noise on the meters. An extended run of the noise log shows low single digit noise blips - a couple a day - otherwise, zero. All comm strength in the teens or higher. I recently put in some new devices from brand X (Gen II) and had problems with them. Still no noise and high strength, but I had various issues that affected basic function. I replaced the devices with like devices, and even replaced them with like devices with newer firmware - same repeatable issues. I troubleshot with the retailer and manufacturer for quite a while. On a whim, I tossed in a switch from brand Y (very old Gen I) and the problems went away. Brand X and Y make great equipment and I have devices from both in the house now, but Brand X did not like the particular application environment at all. The techs said things like 'zero-crossing', and 'resonance', and other confusing words to explain possible causes. I know this is not a copy of what you have in front of you, I only offer it as data to show that you may be able to rework the approach to salvage the mission. I've been told that the rectification of the issue is most likely not a GenI/GenII thing, but more likely due to manufacturing and programming differences between the brands. Good luck!

Edit: Consider that if the cleanliness of the power is REALLY bad, you might have problems with anything electronic, including other HA technologies.