More UPB signal / noise / repeater questions

[danjb]

As I mentioned in another post, I am trying a new UPB installation in my house. I have a PCS split-phase repeater installed, and am trying an installation of about 15 devices before I put in a bunch more.

I ran into some issues that looked noise related (mentioned in this post) that I suspected might be related to the repeater. I could never get totally reliable communications, but things seemed to be better when the repeater was disabled.

I recently tried installing a number of SA UPB filters on know problems areas in my house for PLC communications (surge protectors, UPS's, outlets with a preponderance of electronics on them). This improved things quite a bit, but I was still having issues with some devices occasionally not being reachable by UPStart.

I went through the circuit breakers in the panel, turning off circuits one by one to try to isolate potential noise sources that I may not have filtered, with no different results, unfortunately. I then went to UPStart and decided to turn off the "Automatically make use of it" option for the phase repeater, and change it to "Never make use of it" instead. Before I had put filters in, this improved things a bit, but never got everything working consistently.

Lo and behold, now all the devices seem to be reachable by UPStart! However, communication tests show that some devices have a low signal strength. Repeated tests show single digit (as low as 4 or 5 in one case) signal strengths at some devices, while other devices show average signal strengths over 80. The split-phase repeater shows a signal strength of over 130!

So, my questions are:

1. Why would telling UPStart to use the PCS SPR cause it to be unable to reach sime devices, but when it doesn't use the repeater it can reach those devices? Is the repeater defective? Do I need to do some sort of special configuration of it? Changing its receive sensitivity from HI to LOW does not improve things.
2. I assume the low signal strength of 4 or 5 for some devices is a problem. However, none of the UPStart communications tests shows any noise registering (at least right now). Does it seem reasonable to run with some devices having this low of a signal strength?

I'm getting the impression this active phase repeater is really more trouble than it's worth for this installation. Perhaps I would be better off with a passive coupler like a Simply Automated SA-ZPCI-W to just improve the signal strength across the phases a bit.

Any thoughts?

[nightwalker]

I wish I could give you more to go on as you seem to be taking the logical steps but as far as the repeater goes. I've installed UPB in somewhere around 300 houses or so, only one needed a repeater. That one was a house that was very spread out, and 8000sf or so, it also had 400A service using 2 200A panels in different locations. The repeater helped tremendously on that house but it's the only one I've ever needed it on. I do always use a phase coupler, in the same panel that the CIM is in.

I've run into more problems from loose connections at the panel and end up re-tightening the lines before i even start, just as a matter of course. I also put the CIM on it's own circuit as opposed to just plugging it into any old outlet. This does two things for me, gets my CIM off a circuit that may have noise on it, and allows me to kill every other circuit to make sure that if I am getting noise I know it's not coming from the outside. There is little I can do about noise coming in but if it's in the house you can filter. I'll be honest, either I've been really lucky or just leading an about to crash charmed life but i use very few filters. Mostly I use HAI devices but have installed other brands when needed, In my house i have a mixture of all of them.

[Pete]

Here my SA passive couple lowers my signals and the PCS PPC1 improves my signals.

I did do a test yesterday enabling and disabling the HAI SPR at the breaker and using upstart. I saw better signal strength with it on than off from my mix of devices.

I use a UPB USB floater around the house with my laptop to look for noisy circuits.

Here too the HAI SPR, UPB passive coupler and PIM(s) are on their own breakers.

The UPB PIM is literally connected to the fuse panel a few inches away in an automous gang box with only the PIM plugged into the double outlet.

The second UPB PIM is also connected to its own circuit next to the fuse panel.

I also just installed a 4th UPB PIM on the second floor via a Quatech box to just look at signals from one room in the house away from the electrical panel.

As the passive couplers, SPR and PIMs are on different circuit breakers I will post signal results using the two passive couplers on or off, SPR on or off for the HAI, PCS and SA switches to get you an idea of the signals that I get here.

I ended up using many split tandem breakers for these additional devices. One of the mistakes I have made though was putting the SPR on the same phase because of the additional split breakers are a bit overwhelming.

[Uncle Michael]

300 houses!!!

Dan,
This problem is definitely perplexing. If you are game, I think a phase coupler is worth a try. (Maybe you should try to persuade Marty to stop by for an expert consultation. )

[Pete]

Yes; you should hire Marty (our resident UPB guru) to help you with your endeavor.

[nightwalker]

They weren't all this year, I've been doing this for a long time.

BTW it's a long way to Texas from here, of course it is warmer there and the older i get the better warmer looks to me.

[danjb]

I wish I could get Marty down here! Unfortunately there doesn't seem to be a consistent explanation for the behavior. I tried a passive phase coupler in place of the SPR today, and it did produce some minor improvements in signal strength as reported by UPStart. For example, one lamp module reported an average signal strength at the PIM of 5.0 and an average signal strength at the device of 7.7. With the phase coupler in place, these numbers became 9.0 and 9.9 respectively.

However, that's still pretty low strength, especially compared to strengths in the 50's/60's or even 80's for some devices on the same phase as the PIM. I had imagined the phase coupler would have produced a bigger difference. I don't want to get hung up on numbers in UPStart if everything is working, but I did have one of the weak signal devices miss a comm test in a 10 repeat test, so there must be actual issues.

I'm beginning to wonder if it's a problem specific to the circuit the current interface is on. I'm going to experiment with moving it around. Any suggestions on other things to try?

[tenholde]

there is also a difference in phase couplers. Some invert the signal passed over to the other phase, and some do not.

Depending upon your house, one may work better than the other.

tenholde

[danjb]

It is indeed an inverting phase coupler (Simply Automated ZPCI). Are those worse somehow than other couplers? Reading its own ad copy makes it sound a lot better ( Surprise!).

[nightwalker]

Have you tried a little bit of what Pete does and put upstart on a laptop and take various readings around the house to see if some circuits have noise while others don't. It could very well be either something plugged into those circuits or a simple loose connection or even a breaker that is not doing as well as it should and causing some noise that won't effect electricity but may effect data getting through.

I had one house that was giving me fits, I knew the noise was coming from outside but could never get the power company to do anything. Finally after the Sandy storm the transformer got flooded and they were forced to replace it. All of the sudden noise levels dropped to nothing and signal strength went way up.

Maybe next time I have that kind of issue I'll just turn the hose on the transformer till it really fails.

[Uncle Michael]

It is indeed an inverting phase coupler (Simply Automated ZPCI). Are those worse somehow than other couplers? Reading its own ad copy makes it sound a lot better ( Surprise!).

I also tried the ZPCI coupler, and found it to be ineffective. The PCS PPC coupler is a non-inverting coupler, and it works much better for me. I also suspect that an inverting coupler may conflict with a split phase repeater, inverting the signal in a way that causes it to cancel itself. (That's just a layman's guess, not based on any real knowledge, so take it for what it's worth.)

[Pete]

I pulled the cover off of the fuse panel to validate connectivity and did a signal test using the HAI SPR, SA Inverting Phase coupler and the PCS PPC non-Inverting Phase coupler. The switch tested is a UPB lighting module. It is a Simply Automated "Gen 1" UML-20 lighting module. It sits on the opposite side of the fuse panel two stories up with bedroom over the garage.

I have 200 Amp service with many tandem breakers numbering 55 total breakers / circuits in a bit under 4000 SF two story colonial style home.

I have three UPB PIMs (including the floater right now) plugged into outlets adjacent to the fuse panel and on their own circuits. I have the SPR, PCS and SA Phase coupler all on separate circuits inside/adjacent to the fuse panel.

The testing PIM is an SA USB UPB PIM model number: USW-USB-W

After this test I give the combo SPR and PPC Phase coupler a slight edge over the SPR by itself or the SA by itself or the SPR and SA on together.

BTW and unrelated. I had a granular look at the neighbors electric. He has had the problem that breakers keep tripping; Christmas lights when the TV goes on in the basement, Christmas lights going out when the treadmill was used, on and on. It appeared to me from looking at the fuse panel and the electric that all of the newer electric was jumpered off the old electric (wherever the boxes were) and no new circuits were added to new additions to the house. I suggested he hire a master electrician to pull new wire through the old conduit to new circuit breakers in the panel. This is very time consuming depending on the number of bends in the conduit and the distance from the boxes to the fuse panel. That and it may include new conduit if the old pipes are already full. IE: I ran a new circuit for two outdoor outside outlets. I fished the wire inside then box to box (5 boxes) until it went to the fuse panel where I connected it to a new breaker. This took most of one day and there was only two bends in the conduit from start to finish.

Tested - Pictures are in same order as numbered tests below.

1 - 061-NOSPR-NOPCSPhaseCoupler
2 - 061-NOSPR-PCSPhaseCouplerON
3 - 061-NOSPR-SAPhaseCouplerON
4 - 061-ONSPR-NOPCSPhaseCoupler
5 - 061-SPRON-PCSPhaseCouplerON
6 - 061-SPRON-SAPhaseCouplerON

[Uncle Michael]

Very interesting experiment, Pete!

As l interpret it, #2 and 5 have the best signal strength results.
Does #4 have a total loss of signal everywhere, or just where you were testing?

[Pete]

Number 4 must have been the location of the PIM maybe? Or because it is a Gen 1 switch that is not supposed to work with an SPR coupler?

The SPR was on for that particular test.

I have not see that same particular switch fault as it connected to one set of Christmas lights that are working every night as scheduled from another UPB PIM.

While I had the fuse panel cover off; revalidated the labeling on the circuit breakers to the PIMs and Phase Couplers and SPR.

I did notice that when the PPC is connected correctly (one wire to each phase) the LED goes on. If you connect the wires to the phase the LED goes out. With the SA phase coupler 3 LEDs; there is one LED per line and one labeled surge. If all three are on you have connected everything OK. If you connect the two line wires to the same phase the line LEDs go on but the surge LED does not.

[Pete]

I was going to try the same test above with the HAI UPB serial PIM and maybe try a PCS serial PIM on the same circuit as previously tested.

I did read somewhere that the PCS serial PIM works well.

I do have to remove the cover off of the fuse panel to test as I have disconnected the SA phase coupler and connected the PCS phase coupler.