BraveSirRobbin
Moderator
OK, are those motion sensors set up to trigger the "next" unit code after motion is detected?
Hardware Feature comparison?
- Thread starter WayneW
- Start date
Hi kwilcoxkwilcox said:
I have had problems with W800 too (W800 with HCA), the number of right answers is very low !!
On the other hand, HouseLink is discontinued, so I have to buy another interface, that´s why I asked about CM11A and PowerLinc Controller (or the Controller USB that has memory). I can learn from your experince that PowerLinc Controller USB is better than CM11A, thanks !!!
kwilcox
Active Member
Ok, the test results are in. I apologize in advance for the somewhat OT nature of this post.
I ended up using the W800 test program that's on the wglassociates web site. While hyperterm works, it displays junk due to the bytestream nature of the W800 interface protocol. The test program decodes the W800 data stream into meaningful X10 commands.
Anyway, when two hawkeyes trigger simultaneously, neither are received. The house/unit code they are assigned makes no difference. This is not an HCA problem; rather it affects all of us who use the W800 regardless of the controller. As was postulated earlier, the two signals apparantly collide in much the same way as their powerline counterparts would.
In my test scenario, I tried two hackeyes set to D10 and D11 then repeated the test with the devices set to D10 and B11. Results were identical. Interestingly, this problem only occurred when the devices fired at precisely the same time. It may be that HCA doesn't recognize W800 reports when they occur very close together; I will continue to test this scenario, perhaps this is why posters with different systems are reporting that they aren't experiencing this issue.
I ended up using the W800 test program that's on the wglassociates web site. While hyperterm works, it displays junk due to the bytestream nature of the W800 interface protocol. The test program decodes the W800 data stream into meaningful X10 commands.
Anyway, when two hawkeyes trigger simultaneously, neither are received. The house/unit code they are assigned makes no difference. This is not an HCA problem; rather it affects all of us who use the W800 regardless of the controller. As was postulated earlier, the two signals apparantly collide in much the same way as their powerline counterparts would.
In my test scenario, I tried two hackeyes set to D10 and D11 then repeated the test with the devices set to D10 and B11. Results were identical. Interestingly, this problem only occurred when the devices fired at precisely the same time. It may be that HCA doesn't recognize W800 reports when they occur very close together; I will continue to test this scenario, perhaps this is why posters with different systems are reporting that they aren't experiencing this issue.
In my house, I am pretty sure it's almost impossible for my motion sensors to trigger at the same time since they are so widespread, and there are only 2 people, unless the mice start dancing at the same time one of us is moving around the house, but I still haven't experienced these missed commands (eventho it makes perfectly sense). Thanks for testing this!
I was having the problem with a motion sensor and a DS10A. The Ds10A controls my closet light, and the motion sensor the bedroom lights. The closet door was right next to the motion sensor. Wh I opened the closet door the sucess rate was very low. I just repositioned the motion sensor and the problem went away.
T
T
rocco
Active Member
Don't apologize. I, for one, appreciate your taking the time to test and report back.kwilcox said:
An X10 RF transmission only takes 108 milliseconds. So if your motion sensors are triggered more than a tenth of a second apart, your OK.kwilcox said:
I find that my dancing mice are too small for my motion sensorselectron said:
Maybe its just me, but I've had pretty dismal experiences with these devices (hawkeye). Thanks for the idea of cutting the light sensor though. That might be enough to save my sanity with them.
What I've been seeing is that they seem to work for a while then slowly go insane. They start sending messages every few seconds. It makes a real mess of the RF when several of them are trying to transmit an 'on' signal every 3 to 5 seconds or so.
I've never really got to the bottom of it. I've usually pulled the batteries out and been >< close to ripping them off the walls and attacking them with a hammer in the garage.
So far, 4 of 6 have done this. Replacing batteries and reprogramming doesn't seem to have an any effect. Maybe I got some from a bad batch? (Or maybe just a bad box of batteries?)
I will try cutting the light sensor and another fresh set of batteries.. Maybe that's the ticket?
(I had been using a WGL v572a, but I also have a W800RF32 to try as well)
What I've been seeing is that they seem to work for a while then slowly go insane. They start sending messages every few seconds. It makes a real mess of the RF when several of them are trying to transmit an 'on' signal every 3 to 5 seconds or so.
I've never really got to the bottom of it. I've usually pulled the batteries out and been >< close to ripping them off the walls and attacking them with a hammer in the garage.
So far, 4 of 6 have done this. Replacing batteries and reprogramming doesn't seem to have an any effect. Maybe I got some from a bad batch? (Or maybe just a bad box of batteries?)
I will try cutting the light sensor and another fresh set of batteries.. Maybe that's the ticket?
(I had been using a WGL v572a, but I also have a W800RF32 to try as well)
And now for something on topic. I have X10 (smarthome products mostly), UPB and Insteon in my home.
My personal likes/dislikes:
X10:
+ Ubiquitous
+ Every device imaginable exists
- LOUSY reliability (even with repeaters etc)
+ If you can tear your house apart, a Lightolier compose firewall might be the ticket. (But I would not be suprised in the least if it turns out that it will out Insteon 131KHz carrier signals as well....)
- System degrades as 2-way devices are added. (Each transmitter sucks more signal away)
- Carrier signal seems to be readily absorbed by many devices in the house these days.
- The high efficiency ballasts for my aquarium tank lights cycle at 60,000Hz. They seem to totally destroy X10 signals nearby (X10 running at 120KHz).
UPB:
+ Exceptional reliability. Absolutely supurb! I'm even using them on the wrong side of a GFCI on the wrong phase, outdoors on a 50-ft extension cord in a waterproof box with 100% reliability. I have something like 10 appliance modules, most controlling fish tank lights and water management. I actually trust it.
+ Carrier signal seems to be quite immune to signal absorbtion so far.
- Carrier signal didn't jump to other phase with any real reliability, as the promo material suggested it should. A cheap passive signal coupler solved that trivially though.
+ Excellent signal propagation. We get "excellent" signal from my computer controller to our neighbor's house. A priceless April fools day prank resulted this year.
+ Computer programming interface is nice. I wrote my own software with no hassle.
- Light switches suck! The body is larger than even the old smarthome devices and won't fit in this house's J boxes. I have to replace the J-boxes to make them fit. The bodies are too tall.
- Light switches double suck! The rockers try and detect single vs double tap. They do this by waiting for something like 1/4 to 1/3 of a second. This is really annoying when both cases are programmed to do the same thing (turn load on instantly). This adds significant user-visible delay to button presses. I dearly wish there was a way to disable the double-tap detection delay. But there isn't. Aargh!
- Expensive! (Well, compared to X10 and smarthome stuff. Not so expensive compared to Z-wave it seems)
- Very limited device availability. I really desperately want a 4 or 8 way relay device. eg: like wgldesigns' relay8 or the smarthome iolinc. I also really desperately want a UPB version of the Rain8. I could also use some in-wall 8-button controllers with lights (like smarthome's keypadlinc stuff).
Insteon:
+ Smarthome seem committed to produce a wide variety of devices.
- Hardly any are available yet. Even less than UPB.
+ switchlinc v2's have excellent WAF. Especially for our instant-on kitchen lights. She dislikes the lag from the SA UPB switches.
+ In theory, system reliability should improve as devices are added due to the mesh system.
- In practice, putting lone insteon devices on the far end of circuits with 8-10 X10 transmitters and no other devices to help them hear remote signals has been less than spectacular here. X10 transmitters and other X10 signal absorbers seem to like the taste of insteon carrier too and seem to absorb it as well. I don't have enough experience with this to be sure though.
+ In theory, adding more signalink RF repeaters should be an acceptable temporary workaround for signal void problems.
+ SALad and the programming environment for the controllers is very interesting. Now, if they allow SALad on more devices, not just the controllers... well, that would be really something!
- Relatively new and little track record in the real world.
- I don't trust smarthome's QA. I deeply distrust the smarthome designed devices after finding protocol bugs still in their late 2004 products. eg: cause three devices to transmit simultaniously and they lock up the X10 network forever, untill you power cycle them. (This is easy to provoke. Suppose you have switchlink-relay's at C1, C2 and C3. Transmit a grouped command like "C1 C2 C3 C-status-request" and BOOM! your X10 network turns to sh*t with 120KHz noise as all three devices try and yell over each other.) I deeply distrust how they'll implement and debug insteon.
- Insteon devices are not going to add reliability to an unreliable X10 system because they don't repeat the X10 signals. Backwards compatability is useless in an unreliable X10 setup.
- Insteon devices will degrade a working X10 system with few transmitters. All insteon devices are X10 transmitters and will suck away X10 signal, without repeating it.
- I dont see how Smarthome can deliver on their promise for a smooth upgrade of an X10 system without needing to replace devices. I personally fear that once you've got more than a handfull of devices, it'll be the point of no return and you'll be committed to replacing everything you can.
My bottom line is that for lighting control, I'm leaning towards replacing all the X10 switchlinc devices with insteon switchlincs. I can't live with the UPB double-tap detection delay. I could replace all the J-boxes, but the double-tap delay is the killer for us. PCS don't seem to care about problems like this. But for any equipment control, where reliability is critical (eg: floods due to missed 'off' events), UPB is my preferred system.
Hard wired signal systems (RS422/485 and friends) would be better, but it just isn't an option in this house.
My personal likes/dislikes:
X10:
+ Ubiquitous
+ Every device imaginable exists
- LOUSY reliability (even with repeaters etc)
+ If you can tear your house apart, a Lightolier compose firewall might be the ticket. (But I would not be suprised in the least if it turns out that it will out Insteon 131KHz carrier signals as well....)
- System degrades as 2-way devices are added. (Each transmitter sucks more signal away)
- Carrier signal seems to be readily absorbed by many devices in the house these days.
- The high efficiency ballasts for my aquarium tank lights cycle at 60,000Hz. They seem to totally destroy X10 signals nearby (X10 running at 120KHz).
UPB:
+ Exceptional reliability. Absolutely supurb! I'm even using them on the wrong side of a GFCI on the wrong phase, outdoors on a 50-ft extension cord in a waterproof box with 100% reliability. I have something like 10 appliance modules, most controlling fish tank lights and water management. I actually trust it.
+ Carrier signal seems to be quite immune to signal absorbtion so far.
- Carrier signal didn't jump to other phase with any real reliability, as the promo material suggested it should. A cheap passive signal coupler solved that trivially though.
+ Excellent signal propagation. We get "excellent" signal from my computer controller to our neighbor's house. A priceless April fools day prank resulted this year.
+ Computer programming interface is nice. I wrote my own software with no hassle.
- Light switches suck! The body is larger than even the old smarthome devices and won't fit in this house's J boxes. I have to replace the J-boxes to make them fit. The bodies are too tall.
- Light switches double suck! The rockers try and detect single vs double tap. They do this by waiting for something like 1/4 to 1/3 of a second. This is really annoying when both cases are programmed to do the same thing (turn load on instantly). This adds significant user-visible delay to button presses. I dearly wish there was a way to disable the double-tap detection delay. But there isn't. Aargh!
- Expensive! (Well, compared to X10 and smarthome stuff. Not so expensive compared to Z-wave it seems)
- Very limited device availability. I really desperately want a 4 or 8 way relay device. eg: like wgldesigns' relay8 or the smarthome iolinc. I also really desperately want a UPB version of the Rain8. I could also use some in-wall 8-button controllers with lights (like smarthome's keypadlinc stuff).
Insteon:
+ Smarthome seem committed to produce a wide variety of devices.
- Hardly any are available yet. Even less than UPB.
+ switchlinc v2's have excellent WAF. Especially for our instant-on kitchen lights. She dislikes the lag from the SA UPB switches.
+ In theory, system reliability should improve as devices are added due to the mesh system.
- In practice, putting lone insteon devices on the far end of circuits with 8-10 X10 transmitters and no other devices to help them hear remote signals has been less than spectacular here. X10 transmitters and other X10 signal absorbers seem to like the taste of insteon carrier too and seem to absorb it as well. I don't have enough experience with this to be sure though.
+ In theory, adding more signalink RF repeaters should be an acceptable temporary workaround for signal void problems.
+ SALad and the programming environment for the controllers is very interesting. Now, if they allow SALad on more devices, not just the controllers... well, that would be really something!
- Relatively new and little track record in the real world.
- I don't trust smarthome's QA. I deeply distrust the smarthome designed devices after finding protocol bugs still in their late 2004 products. eg: cause three devices to transmit simultaniously and they lock up the X10 network forever, untill you power cycle them. (This is easy to provoke. Suppose you have switchlink-relay's at C1, C2 and C3. Transmit a grouped command like "C1 C2 C3 C-status-request" and BOOM! your X10 network turns to sh*t with 120KHz noise as all three devices try and yell over each other.) I deeply distrust how they'll implement and debug insteon.
- Insteon devices are not going to add reliability to an unreliable X10 system because they don't repeat the X10 signals. Backwards compatability is useless in an unreliable X10 setup.
- Insteon devices will degrade a working X10 system with few transmitters. All insteon devices are X10 transmitters and will suck away X10 signal, without repeating it.
- I dont see how Smarthome can deliver on their promise for a smooth upgrade of an X10 system without needing to replace devices. I personally fear that once you've got more than a handfull of devices, it'll be the point of no return and you'll be committed to replacing everything you can.
My bottom line is that for lighting control, I'm leaning towards replacing all the X10 switchlinc devices with insteon switchlincs. I can't live with the UPB double-tap detection delay. I could replace all the J-boxes, but the double-tap delay is the killer for us. PCS don't seem to care about problems like this. But for any equipment control, where reliability is critical (eg: floods due to missed 'off' events), UPB is my preferred system.
Hard wired signal systems (RS422/485 and friends) would be better, but it just isn't an option in this house.
I admit the UPB switches are tall and are a tight fit into a box (so far, I haven't had to replace any boxes) - but I haven't noticed the delay issue you speak of with the tap vs double tap.
I program my switches so single up tap ramps to 85% (.8 seconds). Double tap is full bright (ramp .8 seconds). Bottom tap ramps off (.8 seconds), and bottom double tap goes to nightlight levels (5-20% depending on light, ramps over .8 seconds).
Since all of my actions ramp - the delay is unnoticable. My wife likes the ramp effect, and it's easier on the bulbs. I've programmed the switches as described above so I don't need to train our babysitters/in-laws... it works as they expect (they don't usually know about double taps).
I have never had a single comment (including wife) that these switches have a noticable delay. I'm betting the ramp is why...
I program my switches so single up tap ramps to 85% (.8 seconds). Double tap is full bright (ramp .8 seconds). Bottom tap ramps off (.8 seconds), and bottom double tap goes to nightlight levels (5-20% depending on light, ramps over .8 seconds).
Since all of my actions ramp - the delay is unnoticable. My wife likes the ramp effect, and it's easier on the bulbs. I've programmed the switches as described above so I don't need to train our babysitters/in-laws... it works as they expect (they don't usually know about double taps).
I have never had a single comment (including wife) that these switches have a noticable delay. I'm betting the ramp is why...
BraveSirRobbin
Moderator
This statement caught my eye in Peter's post (excellent post BTW):
I thought Insteon's claim to fame was that they were immune to these types of (same X-10) problems!!
kwilcox
Active Member
@PeterW:
Here's some good news on the UPB iolinc equivalent. It's the SAI UMI-32. Here's a link:
Simply Automated UMI-32
It's got 3 inputs and two outputs. The inputs can be a variety of contact closure/voltage sense. Retail is around the same as the iolnic. Automatedoutlet stocks them too. Search for umi32.
I too haven't noticed the double tap issue but I also haven't programmed double-tap sequences on my switches. I'll have to play a little with that to see what you mean.
I've also not noticed any size issues with my SAI switches. The one's I ordered were about the same length as a switchlinc 2380 and not quite as deep. I'm using US1130W's, the single rocker only version. I'd imagine the 4 contact version would be a bit bigger since it's got 4 times the capability of a switchlinc.
Here's some good news on the UPB iolinc equivalent. It's the SAI UMI-32. Here's a link:
Simply Automated UMI-32
It's got 3 inputs and two outputs. The inputs can be a variety of contact closure/voltage sense. Retail is around the same as the iolnic. Automatedoutlet stocks them too. Search for umi32.
I too haven't noticed the double tap issue but I also haven't programmed double-tap sequences on my switches. I'll have to play a little with that to see what you mean.
I've also not noticed any size issues with my SAI switches. The one's I ordered were about the same length as a switchlinc 2380 and not quite as deep. I'm using US1130W's, the single rocker only version. I'd imagine the 4 contact version would be a bit bigger since it's got 4 times the capability of a switchlinc.
kwilcox
Active Member
Let me just add that another reason I'm probably not noticing tap/double tap issues is because nobody at our house actually touchs any switch to turn on lights. I've got hawkeyes in every room that activate the lights when they sense motion if its dark outside. I'll have to play with the office lights to see what you mean. I can't remember noticing anything unusual in the office before I installed the sensor there, but it may have just been because I was too giddy about the reliability of the switches to notice it.
kwilcox
Active Member
@PeterW
Well, I played with my office UPB switches over lunch and to tell the truth, I'm mystified. For a reference point, my US1130W's have a rocker whose tactile response feels and sounds identical to that of a switchlinc. My testing was against the switch's directly connected load which in my case was the office overhead lights; a set of 4 cans with dimmable CF bulbs.
That being said, I tap the rocker and then the light comes on. It occurs right after the rocker returns to the neutral postion. It's kind of like you hear the sound "click-a" (which are the two sounds the rocker makes when you tap it so imagine this for your own timing references) then the light lights. So the overall effect is sort of like "click-a-bling!" where "bling" is the lamp lighting. Is this the delay you are talking about? I must've turned my lights off and on a dozen times but I couldn't figure out anything else. Sorry. Maybe the US1130's don't suffer from the delay issue?
Edit: Oops, I was going to respond to your hawkeye woes too!
I'm thinking your problem is that the WGL V572A actually ends up placing all the hawkeye traffic onto the powerline because it interfaces to a TW523 or simular powerline interface. In this scenario, you're absolutely correct: it's a recipe for disaster! A large hawkeye setup will flood the powerline and cripple your installation when connected in this manner. In fact, I originally gave up on Hawkeyes for that same reason. However, the W800RF is an entirely different story since it sends Hawkeye traffic directly to your controller via its serial interface which then acts upon it by starting programs, triggering lights etc. When that device's support was added to HCA I wasted no time purchasing a W800 and now have a very large Hawkeye installation up and running with no major issues. This because their traffic now occurs at wireless speeds and is sent directly to HCA. I'd try the W800RF interface out if I were you. It makes a night and day difference.
Well, I played with my office UPB switches over lunch and to tell the truth, I'm mystified. For a reference point, my US1130W's have a rocker whose tactile response feels and sounds identical to that of a switchlinc. My testing was against the switch's directly connected load which in my case was the office overhead lights; a set of 4 cans with dimmable CF bulbs.
That being said, I tap the rocker and then the light comes on. It occurs right after the rocker returns to the neutral postion. It's kind of like you hear the sound "click-a" (which are the two sounds the rocker makes when you tap it so imagine this for your own timing references) then the light lights. So the overall effect is sort of like "click-a-bling!" where "bling" is the lamp lighting. Is this the delay you are talking about? I must've turned my lights off and on a dozen times but I couldn't figure out anything else. Sorry. Maybe the US1130's don't suffer from the delay issue?
Edit: Oops, I was going to respond to your hawkeye woes too!
I'm thinking your problem is that the WGL V572A actually ends up placing all the hawkeye traffic onto the powerline because it interfaces to a TW523 or simular powerline interface. In this scenario, you're absolutely correct: it's a recipe for disaster! A large hawkeye setup will flood the powerline and cripple your installation when connected in this manner. In fact, I originally gave up on Hawkeyes for that same reason. However, the W800RF is an entirely different story since it sends Hawkeye traffic directly to your controller via its serial interface which then acts upon it by starting programs, triggering lights etc. When that device's support was added to HCA I wasted no time purchasing a W800 and now have a very large Hawkeye installation up and running with no major issues. This because their traffic now occurs at wireless speeds and is sent directly to HCA. I'd try the W800RF interface out if I were you. It makes a night and day difference.
Similar threads
