ChrisWalker
Active Member
I had originally deleted my post here because it was a little harsh on the UPB testing we did, but due to popular demand (hee hee, see below) I'll share at least a bit of it so you can get a good feeling for our experience with UPB these last two days.
First of all, our company (ControlThink) is a software and services company. We do two things: create super-reliable software, and provide services where we test things in depth to make sure products work well generally and with software.
To this end, a company named Simply Automated (SA) provided us UPB switches (which we paid for) to test in one of our dealer's $3 million show homes. We installed 44 master switches and a dozen or so slaves. And we also installed a few Z-Wave devices (thermostat control, most notably) to show off the seamless bridging of protocols.
SA seems like a great company. They seem to really care about quality. They've been very responsive to the issues we're sharing with them, and they seem very interested in fixing the ones they can fix (i.e. the ones not related to the protocol particularly).
Here are a few of the high points and low points of our experience over three days:
High Points:
1. BEAUTIFUL SWITCHES: When the SA remote (slave) switches catch up to their 2nd generation (like the master ones), it's going to be a very pretty system. The blue glow is awesome on the (fairly expensive) masters.
2. SEEMS TO WORK WELL: Once [painfully] set up, the network seems to work well. The real-time status updates seem like they're going to have troubles when integrated with a GUI in certain circumstances, but when operated without an automation system the home is very nice.
Low Points:
1. HIGH FAILURE RATE: 10% of switches either had bad UPB hardware (couldn't communicate) or bad triacs (couldn't control their load). One of those burned out when a short developed down the line, and one burned out randomly after about an hour. None of them were wired hot. Additionally, 15% or more of the rockers were bad, although I understand that issue was possibly an isolated batch and has been fixed on all new shipping rockers.
2. UNACCEPTABLE LINE NOISE: The audio/video system, which cost hundreds of thousands of dollars, now crackles and sounds like the UPB pulses are going to damage it. The demo home will now be receiving a power conditioning upgrade on those two dedicated circuits at a cost of tens of thousands of dollars.
3. SWITCHES NOT COMMUNICATING WHEN ON: For about 5%-10% of the switches, they communicate fine when they're off, but do not respond when they are on. Additionally, the attached lights flicker like a scene from a horror movie when they are trying to communicate while on. It's trippy.
4. INTERESTING DEFAULT CONFIGURATION: The default configuration on the switches, when coupled with a single rocker, was to basically turn off the whole home whenever the switches were turned off. This can be fixed via the UPStart tool, but that was a painful 6-8 hour process (just to get the light switches working like regular light switches) during which the software crashed more than a half dozen times, sometimes losing our work.
5. BRUTE FORCE REQUIRED FOR RELIABILITY: We set up a 3-way virtual setup for the master bedroom, since there were no traveler wires. We had to ratchet the retries up to the max (4) to get the switches to communicate reliably. This seems to possibly be in part because UPB lacks collision detection, but I'm not sure.
6. SETUP VERY SLOW: We're used to setting up 60 Z-Wave switches (i.e. after they're in the wall) in 15-30 minutes. It took us 6-8 hours. Figure out an average hourly rate for an installer on that, and add in the $1,000 extra paid to the electricians to deal with all the problems, and this install cost about $1,500 to $2,000 more than a similar Z-Wave installation would have cost. Which means the installation may end up costing as much as the switches themselves.
We found out a lot more during the process. And Simply Automated seems dedicated to fixing almost every one of these issues. They can't deal with some of the core collision/reliability issues, since those are UPB--but hopefully they make some recommendations there. We will probably be communicating with PCS as well.
As a testing firm, I think we can safely say that we still see huge advantages in the Z-Wave world in terms of ease of installation and use, reliability, and affordability. But, short-term at least, and in particular controlled situations, there are certainly times when UPB makes sense in a mixture of technologies.
Chris
First of all, our company (ControlThink) is a software and services company. We do two things: create super-reliable software, and provide services where we test things in depth to make sure products work well generally and with software.
To this end, a company named Simply Automated (SA) provided us UPB switches (which we paid for) to test in one of our dealer's $3 million show homes. We installed 44 master switches and a dozen or so slaves. And we also installed a few Z-Wave devices (thermostat control, most notably) to show off the seamless bridging of protocols.
SA seems like a great company. They seem to really care about quality. They've been very responsive to the issues we're sharing with them, and they seem very interested in fixing the ones they can fix (i.e. the ones not related to the protocol particularly).
Here are a few of the high points and low points of our experience over three days:
High Points:
1. BEAUTIFUL SWITCHES: When the SA remote (slave) switches catch up to their 2nd generation (like the master ones), it's going to be a very pretty system. The blue glow is awesome on the (fairly expensive) masters.
2. SEEMS TO WORK WELL: Once [painfully] set up, the network seems to work well. The real-time status updates seem like they're going to have troubles when integrated with a GUI in certain circumstances, but when operated without an automation system the home is very nice.
Low Points:
1. HIGH FAILURE RATE: 10% of switches either had bad UPB hardware (couldn't communicate) or bad triacs (couldn't control their load). One of those burned out when a short developed down the line, and one burned out randomly after about an hour. None of them were wired hot. Additionally, 15% or more of the rockers were bad, although I understand that issue was possibly an isolated batch and has been fixed on all new shipping rockers.
2. UNACCEPTABLE LINE NOISE: The audio/video system, which cost hundreds of thousands of dollars, now crackles and sounds like the UPB pulses are going to damage it. The demo home will now be receiving a power conditioning upgrade on those two dedicated circuits at a cost of tens of thousands of dollars.
3. SWITCHES NOT COMMUNICATING WHEN ON: For about 5%-10% of the switches, they communicate fine when they're off, but do not respond when they are on. Additionally, the attached lights flicker like a scene from a horror movie when they are trying to communicate while on. It's trippy.
4. INTERESTING DEFAULT CONFIGURATION: The default configuration on the switches, when coupled with a single rocker, was to basically turn off the whole home whenever the switches were turned off. This can be fixed via the UPStart tool, but that was a painful 6-8 hour process (just to get the light switches working like regular light switches) during which the software crashed more than a half dozen times, sometimes losing our work.
5. BRUTE FORCE REQUIRED FOR RELIABILITY: We set up a 3-way virtual setup for the master bedroom, since there were no traveler wires. We had to ratchet the retries up to the max (4) to get the switches to communicate reliably. This seems to possibly be in part because UPB lacks collision detection, but I'm not sure.
6. SETUP VERY SLOW: We're used to setting up 60 Z-Wave switches (i.e. after they're in the wall) in 15-30 minutes. It took us 6-8 hours. Figure out an average hourly rate for an installer on that, and add in the $1,000 extra paid to the electricians to deal with all the problems, and this install cost about $1,500 to $2,000 more than a similar Z-Wave installation would have cost. Which means the installation may end up costing as much as the switches themselves.
We found out a lot more during the process. And Simply Automated seems dedicated to fixing almost every one of these issues. They can't deal with some of the core collision/reliability issues, since those are UPB--but hopefully they make some recommendations there. We will probably be communicating with PCS as well.
As a testing firm, I think we can safely say that we still see huge advantages in the Z-Wave world in terms of ease of installation and use, reliability, and affordability. But, short-term at least, and in particular controlled situations, there are certainly times when UPB makes sense in a mixture of technologies.
Chris