Print

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

[toasterking]

I noticed a strange phenomenon recently.  Connecting an XPPF plug-in noise filter to the outlet on *some* appliance modules causes them to fail to turn off.  Well, they turn off, but immediately back on again.  The issue seems to be with the local control feature.  Even with no load connected to the XPPF, it still errantly activates local control and the appliance module switches on immediately after switching off.

Here are the appliance modules I experimented with.  ALL of these do seem to have local control, because physically disconnecting and reconnecting the XPPF in all cases causes the appliance module to transition from an OFF to ON state.

These modules appear to be incompatible with XPPF, i.e. do not remain in an OFF state when sent an OFF command:
AM466, date code 07I39
PAM01, date code 09I39
PAM02, date code 10I39

These modules do seem compatible with XPPF, i.e. do not exhibit the behavior noted above:
AM466, date code 04B06
AM466, date code 7K46
AM486, date code 9D17

Only the appliance modules with codes ending in I39 seem to have the incompatibility.  The XPPF I used for all tests has date code 09C10.  I suppose it is possible that there is another version of the XPPF that does not cause these problems at all.

UPDATE:  I now realize that it is not the "I39" part that is significant but the first numeral which indicates the year of manufacture.  The three modules that do not exhibit the behavior are the oldest ones.  More info on the date codes here:  http://forums.x10.com/index.php?topic=29215.msg163939#msg163939

[Brian H]

Attached is the schematic of the Leviton 6288 model of the X10Pro XPPF.
C1,C2 and C3 are across the power lines.

There is NO bleeder resistor across the capacitors. So unplugging an XPPF with no load on it when the AC was near peek. Can result in a shock if you touched the Input Pins. From the charged capacitors.

I did a test with four AM466 Three Pin Appliance Modules Date Code: 09K48 that are the revised CFL friendly models. I duplicated your findings. I originally thought it was the older model Local Control Sensing. I now am thinking. The charged capacitors are fooling the Output  On or Off sensor when turned Off. The charged capacitors make it look like it didn't go Off and it tried again. Turning it back On.
I did also try one old hardware version with local control sensing and it didn't trigger back On.

I don't have the Date Code information but X10WTI did come out with a revised Appliance Module. Sold as CFL friendly and Local Control Sensing having been removed. Long time since I tested any of the CFL friendly ones.

[Brian H]

I tried an older X10 Date Code: 01D16 AM14A Two Way Appliance Module. It didn't seem to want to trigger back On.
Same for two B&D Freewire FWAR Date Code: 200424-XT Appliance Modules and a FWLROD Date Code: 200402-XT Freewire Outdoor Appliance Module.

Now I do remember the later CFL Friendly models being reported to trigger back On with noise spikes from some loads going Off.

[dhouston]

Now I do remember the later CFL Friendly models being reported to trigger back On with noise spikes from some loads going Off.

Long before the flood of CFLs, I had a screw-in lamp module that would react to a 50+ year old tube fluorescent turning off. Changing to an LM14A fixed the problem.

[Brian H]

I did a few more tests with the four CFL Friendly Appliance Modules.

XPPF and a 330K bleeder resistor. All four still where going back On.
XPPF and a 33K bleeder resistor. All four where OK.

Smarthome Filterlinc and a 330K bleeder resistor. All four still went back On
Smarthome Filterlinc and a 33K bleeder resistor. All four where OK.

ACT AF120 and a 330K bleeder resistor. All four still went back On.
ACT AF120 and a 33K bleeder resistor. All four where OK.

[toasterking]

Brian,

How much power are your 33K bleeder resistors dissipating?  What power rating did you use and how warm do they get?  I didn't try any bleeder resistors, but I am sure that would have worked also.

It's interesting that the XPPF is triggering the output state sensor.  I don't recall the application, but I know that I have had trouble with that in the past.  If they only could have used a bistate relay with 2 sets of contacts, one for state feedback, they would have been so much better.

In case anyone is curious, this is how I happened upon this oddity.  I had a set of powered speakers switched by an appliance module:
   AM466 --> SMPS --> Powered speakers
While I was doing some cleanup, I identified the SMPS as a signal sucker with my XTBM Pro.  It had eluded me before because those speakers are not usually switched on.  So I added an XPPF:
   AM466 --> XPPF --> SMPS --> Powered speakers
That solved the signal sucking problem, but my other problems began.  Based on the info I posted, I changed to an older AM466 that didn't fire back on immediately:
   AM466, date code 7K46 (really old) --> XPPF --> SMPS --> Powered speakers
Sure enough, it didn't fire back on immediately, but about 6 seconds after switching off, the power LED on the speakers would pulse once weakly and then the AM466 would switch back on.  Definitely local control this time.  Some interaction between the SMPS and the capacitors in the XPPF, maybe?  Maybe the local control sensing current from the AM466 was involved too.  I don't know for sure, but it never did that before without the XPPF.  This is the combination that finally worked:
   AM466, date code 7K46, modified for no local control and no sensing current --> XPPF --> SMPS --> Powered speakers
Since modifying the AM466 for no sensing current, the power LED on the speakers no longer pulses after switching off either.

This is why this stuff is for hobbyists!  Most consumers won't put this much effort into making two products for the same system from the same company work together.  I didn't try appliance modules from another manufacturer, but they probably would have worked fine since other brands usually don't have the "unique features" of local control and single-pole bistate relays.  I am interested to see how some of these designs may change in the future now that Authinx has creative license.

[Brian H]

33K 1 watt is fine. As the power it dissipates is around .5 watts. Dave W got some 3 watt ones from an Ebay dealer.
Another member here. Used three 100K .5 watt resistors from Radio Shack, in parallel.

What you did on the 7K46 Appliance Module. Probably removed the sensing current as the LED on the power speakers no longer pulse.

For newer members not familiar with the Local Control Sensing removal. On older before CFL Friendly Appliance Modules:
The older appliance modules with Local Control Sensing. Can have the feature removed by cutting a diode or jumper.
Jumper removed the triggering on a change. The actual sensing current was still there and some CFLs and LED bulb still flashed.
The Diode killed most of the current just leaving the very small On Off sensor the electronics needed to function.
I had cut both the jumper and removed the diode in a few of mine.
Details can be found here. http://www.idobartana.com/hakb/index.htm

[toasterking]

Thanks, Brian, for the information on those resistors.

Indeed, the AM466 I modified no longer supplies sensing current to the load.  I cut the lead on the diode as mentioned in the link you provided.

Slightly off topic, but I also just posted modifications I made to disable the sensing current on a variation on PAM01 and PLM01 modules that I was not able to find documentation of elsewhere:


PAM01:
http://forums.x10.com/index.php?topic=29097

PLM01:
http://forums.x10.com/index.php?topic=29098