Edit: My apologies to the OP. It was not my intent to knock your post off the rails. Nonetheless, that appears to be the effect that I've had. If there are further powerline related troubleshooting questions, I will attempt to reply. If there are further Insteon vs UPB related questions, I would suggest a new thread.
Digger, JDale, and Lou,
I'm sorry, but my post was not intended to ignite the old Insteon/X10/UPB controversy. I was not trying to single out any protocol. Instead, I was trying to focus on the limitations of the Powerline as a communication medium. To that end:
1) All of these technologies must obey certain laws of physics line losses and noise are a fact of life.
2) By the numbers (the transmission line impedances I presented), all of these technologies "work" in a home environment.
3) The technology will break down when faced with multiple "problem items" on a circuit. This is what can make troubleshooting so difficult. You are not looking for a single problem, but rather a combination of problems that combine to take down the communication.
The signal absorption numbers that I presented above were for the Insteon communication Frequency of 130 KHz. They assumed one responder at the end of a long line. In reality, there would likely be multiple units spaced over the length of the circuit. Each would repeat the message, thereby restoring signal level.
I may also have left the impression that I don't care for Insteon - that is not at all the case. I have been with Insteon since the beginning and use it alongside my X10 system in a 4500 Sq foot home. Like Digger, I would say I have above the normal number of noise generators/signal absorbers (6 PC's, 2 home theater setups, 4 additional TV's, the list goes on). Nonetheless, I currently have no filters installed. I also have only 1 Accesspoint installed - it's simply used as a RF receiver.
The point that I was trying to make is that, if you do have problems with an Insteon system, it is more difficult to troubleshoot. I will agree with JDale that, in a properly functioning system, Accesspoints add additional redundancy through multiple signal paths. My point is that, in a properly operating system, AP's should not be required. The message hopping nature of Insteon should be sufficient to overcome normal line loss problems in your home. If it can't, you have an abnormal problem that should be "fixed". Too often I see people trying to fix the problem by "hopping around it" with an accesspoint. In my mind this doesn't "fix" anything. The problem still exists, and will likely come back at a later point in time to take a big bite out of ...
The output voltage level of the Insteon products (3.2 Vp-p in 5 ohms) has also been questioned, and compared to the 40 V-p level of UPB outputs. This really is not a fair comparison because of the difference in technologies.
1) UPB devices place a single 40 V capacitive spike on the powerline every half cycle. Two bits of information are conveyed depending on the time placement of the spike. Since data is only transferred every half cycle, the capacitive circuit have ample time to re-charge in preparation for the next data bit. Since two bits are communicated every half cycle, UPB supports a data rate of 480 bits/second.
2) Insteon devices put a 130Khz carrier on the powerline @3.2 Vp-p. Insteon communications occur at each zero crossing (every half cycle) and comprise 24 bits of information. Raw data rate (including header information) is therefor 2880 bits/second.
The point here is that these are completely different communication approaches. UPB couples 40V spikes onto the line. While this is a high instantaneous power, it isn't sustained (average power is very low). Insteon sustains 1W of power over a period of 1.8 ms - this requires significant "capacity" from the power supply. Note also that the output is rated at a 5 ohm load - this load is generally accepted as the "low" level impedance of a home wiring system. Typical levels are on the order of 10 ohms while high limits are on the order of 30 ohms. While I don't have details on the output stage configuration for Insteon units, it would be reasonable to assume that they would be typically higher than the quoted 3.2 Vp-p (up to a max output stage voltage limit).
Could SH increase the output level? Absolutely. From a very high level, options might include:
1) Decreasing the data rate - lower the sustained output power level (complete re-write of the Insteon protocol - incompatible with current devices).
2) Increasing device power supply capacity
A) increase the form factor - this would present problems for J-box installed devices (I do like their current size).
B) move to "high capacity" storage devices - increase cost.
Quite honestly, I don't like any of the trades above and I'm totally unconvinced that they are necessary.
As a data point - I have one of Jeff Volps X10 plug in boosters (XTBR). This devices boosts my X10 to roughly 32 Vp-p in my home and is quite simply a godsend to mixed Insteon/X10 systems. In order to provide that output level, Jeff went to a transformer based power supply design (it's rather large and heavy).
The following comparison is unfair because Jeff was designing a single point plug in repeater. His design goals were completely different from a higher output Insteon device. Nonetheless, his design is the only data point I have for a high power output communication interface:
Insteon "bursts" last 1823 us while X10 bursts last 1023 us. Simple physics mandate that power output = power input * efficiency. Jeff's device would need to grow a bit to provide headroom for a transmission that is 1.8 times as long.
I am by no means a power supply designer, but I do not see how you can increase the output level significantly without affecting Items 1 or 2 above.
Digger, JDale, and Lou,
I'm sorry, but my post was not intended to ignite the old Insteon/X10/UPB controversy. I was not trying to single out any protocol. Instead, I was trying to focus on the limitations of the Powerline as a communication medium. To that end:
1) All of these technologies must obey certain laws of physics line losses and noise are a fact of life.
2) By the numbers (the transmission line impedances I presented), all of these technologies "work" in a home environment.
3) The technology will break down when faced with multiple "problem items" on a circuit. This is what can make troubleshooting so difficult. You are not looking for a single problem, but rather a combination of problems that combine to take down the communication.
The signal absorption numbers that I presented above were for the Insteon communication Frequency of 130 KHz. They assumed one responder at the end of a long line. In reality, there would likely be multiple units spaced over the length of the circuit. Each would repeat the message, thereby restoring signal level.
I may also have left the impression that I don't care for Insteon - that is not at all the case. I have been with Insteon since the beginning and use it alongside my X10 system in a 4500 Sq foot home. Like Digger, I would say I have above the normal number of noise generators/signal absorbers (6 PC's, 2 home theater setups, 4 additional TV's, the list goes on). Nonetheless, I currently have no filters installed. I also have only 1 Accesspoint installed - it's simply used as a RF receiver.
The point that I was trying to make is that, if you do have problems with an Insteon system, it is more difficult to troubleshoot. I will agree with JDale that, in a properly functioning system, Accesspoints add additional redundancy through multiple signal paths. My point is that, in a properly operating system, AP's should not be required. The message hopping nature of Insteon should be sufficient to overcome normal line loss problems in your home. If it can't, you have an abnormal problem that should be "fixed". Too often I see people trying to fix the problem by "hopping around it" with an accesspoint. In my mind this doesn't "fix" anything. The problem still exists, and will likely come back at a later point in time to take a big bite out of ...
The output voltage level of the Insteon products (3.2 Vp-p in 5 ohms) has also been questioned, and compared to the 40 V-p level of UPB outputs. This really is not a fair comparison because of the difference in technologies.
1) UPB devices place a single 40 V capacitive spike on the powerline every half cycle. Two bits of information are conveyed depending on the time placement of the spike. Since data is only transferred every half cycle, the capacitive circuit have ample time to re-charge in preparation for the next data bit. Since two bits are communicated every half cycle, UPB supports a data rate of 480 bits/second.
2) Insteon devices put a 130Khz carrier on the powerline @3.2 Vp-p. Insteon communications occur at each zero crossing (every half cycle) and comprise 24 bits of information. Raw data rate (including header information) is therefor 2880 bits/second.
The point here is that these are completely different communication approaches. UPB couples 40V spikes onto the line. While this is a high instantaneous power, it isn't sustained (average power is very low). Insteon sustains 1W of power over a period of 1.8 ms - this requires significant "capacity" from the power supply. Note also that the output is rated at a 5 ohm load - this load is generally accepted as the "low" level impedance of a home wiring system. Typical levels are on the order of 10 ohms while high limits are on the order of 30 ohms. While I don't have details on the output stage configuration for Insteon units, it would be reasonable to assume that they would be typically higher than the quoted 3.2 Vp-p (up to a max output stage voltage limit).
Could SH increase the output level? Absolutely. From a very high level, options might include:
1) Decreasing the data rate - lower the sustained output power level (complete re-write of the Insteon protocol - incompatible with current devices).
2) Increasing device power supply capacity
A) increase the form factor - this would present problems for J-box installed devices (I do like their current size).
B) move to "high capacity" storage devices - increase cost.
Quite honestly, I don't like any of the trades above and I'm totally unconvinced that they are necessary.
As a data point - I have one of Jeff Volps X10 plug in boosters (XTBR). This devices boosts my X10 to roughly 32 Vp-p in my home and is quite simply a godsend to mixed Insteon/X10 systems. In order to provide that output level, Jeff went to a transformer based power supply design (it's rather large and heavy).
The following comparison is unfair because Jeff was designing a single point plug in repeater. His design goals were completely different from a higher output Insteon device. Nonetheless, his design is the only data point I have for a high power output communication interface:
Insteon "bursts" last 1823 us while X10 bursts last 1023 us. Simple physics mandate that power output = power input * efficiency. Jeff's device would need to grow a bit to provide headroom for a transmission that is 1.8 times as long.
I am by no means a power supply designer, but I do not see how you can increase the output level significantly without affecting Items 1 or 2 above.
