Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[JeffVolp]

A question has come up on how the XTBM displays high-level noise.  Because noise is sampled directly inside the X10 reception window, it cannot be sampled during reception of X10 “1” bits.  As a result, the noise readout is not updated when receiving a continuous high noise level that looks sufficiently like a X10 signal that the XTBM tries to decode it.  Since the XTBM is unlikely to decode a valid start pattern, the decode will quickly fail, indicating either “BSC” (bad start code) or ^NOISE^ (high noise).  In this situation, those warnings are the only indication that there is a serious noise problem that must be resolved.  Again, this issue only involves high-level noise that looks sufficiently like a X10 command for the XTBM to try to decode it.

As an alternative, I have a firmware mod that moves the noise sample point just after the X10 transmission window when receiving X10 “1” bits.  The downside is that the noise value can change when receiving a X10 command.  For example, the Cellet cellphone charger described in one of my troubleshooting tutorials produces more noise as the AC voltage rises.  So the noise displayed from that unit increases significantly during reception of a continuous series of X10 commands.  However, noise displayed from a continuous source, such as a “wireless” intercom, remains almost the same.

While this mod can result in some erroneous noise readouts, I think it will make the XTBM more useful for those of you battling a high noise level because the actual value of the noise level would be displayed at all times.

Your thoughts or suggestions on this issue are welcome.

Jeff

[Brandt]

I see BSC and ^NOISE^ all the time! but the Noise readout is never any higher than .01!!

[JeffVolp]

I see BSC and ^NOISE^ all the time! but the Noise readout is never any higher than .01!!

That is exactly what happens when continuous noise looks like a X10 signal, and the XTBM is trying to decode it.  There is a tradeoff between trying to focus only on noise in the X10 reception window (which is where it will effect X10 communication) and provide a continuous update.

Another approach is to display the potential X10 signal level as a noise level whenever one of those errors pop up, but that has its own set of warts.  I'm not sure what is the best way to deal with this, which is why I was looking for some feedback.

Jeff

[Brandt]

Maybe a mode switch (maybe P2 ON/OFF )to toggle between listening for noise, and listening for x10 signals

[Brian H]

The firmware mod may help those with high noise levels.
I think on a few occasions I have seen the high noise reading just sit there until the noise goes down.

[JeffVolp]

I can send out a few "beta" PICs with the alternate firmware if you want to try it.  Just contact me directly.

Jeff

[JeffVolp]

I can send out a few "beta" PICs with the alternate firmware if you want to try it.  Just contact me directly.

Nobody interested?

Jeff

[Brian H]

I will try it.
Email to follow.

[dbemowsk]

I have seen on occasion where a high noise level has shown ^NOISE^ on the display continuously until the noise has dropped.  That is how I found out that my well pump was causing a serious amount of noise.  I have since put a filter on the pump and have not had too many issues that have caused the ^NOISE^ display.  Though I have not had many issues recently where there has been a need to use my XTBM, but I know I will be doing some switch and module installs in the near future and am sure it will be needed.  I don't know that I would be the best candidate for testing the new firmware right now, but I am wondering what the new firmware chip would sell for in the future if I should want it?

[JeffVolp]

In testing on my own system there is no difference in the way the XTBM works.  However, if I inject 120KHz noise from my function generator onto the powerline, the new firmwave will track the amplitude as I vary the noise level.  Whether it makes any difference in your installation depends on whether you frequently receive the high noise or bad start code indications.

Since this really doesn't add any new capability to the unit, the cost of the upgrade will be pretty much the mailing cost if you send the chip back for reprogramming, and a bit more if you want a new chip.

Jeff

[Brian H]

Chip arrived and was installed safely.

Only thing I saw so far is I no longer get the 'I' indication on Insteon Signals. Just X10 bad code messages.

I did see the noise measurement vary as the noise levels changed.

[JeffVolp]

Thanks for the feedback.  I don't understand how changing the noise sample point while decoding a X10 message would have effected the Insteon indicator, but I'll look into it.

An Insteon signal should give a X10 bad start code.  The fact that the Insteon "noise" exists before the zero crossing, and goes away after the end of the X10 reception window, is how it identifies the Insteon signal as different from ordinary noise.

Unfortunately, I don't have any Insteon equipment here to test with, so I try to simulate the Insteon signal with my function generator.

This mod is really aimed at the few people who have very high noise levels in their installations because the actual noise readout should help them track down where the noise is coming from.

Jeff

[Brian H]

I did some added tests.
Put a TM751 in the AC Outlet on my XTB-IIR
Repeater Test where all OK. Good voltage and Frequency readings on the XTBM.
Went to a few outlets in the home and got good voltage readings and low 105KHz to 110KHz. frequency readings.

Put V1.03 back and got the samething.

I was hosed by a New Revision {DC:10E19} TM751. Most likely the new wimpy power supply and the strong X10 signal on the power line AC.
Moved it to an outlet not on the XTB-IIR and it works fine in the tests.

Installed a slightly older TM751 that is also a new rev {DC:9L53}all readings where OK in the XTB-IIR. Same for an old design [DC:00D17}.

V1.03 also doesn't give the 'I' all the time either.

Put V1.04 back with the old TM751 in the XTB-IIR and now all reading looked good and it also sometimes shows the 'I'. So don't over test for the 'I' thing.

I did with V1.04 sometimes get the Failed Selftest Message on power up.

[JeffVolp]

Thanks for the update.  Since the Insteon indicator is in a different piece of code, I was having trouble understanding the relationship.

Several people have reported the Fail Self Test in some receptacles.  That can happen if there is a nearby signal sucker or noise source.  I could loosen up the tolerance, but then it would be testing function only, and not accuracy.  If it just occurs in certain receptacles, take that indication to mean there is something nearby that could interfering with the XTBM's X10 transmission.  (The mod had nothing to do with any of the self test code.)

Jeff