X10 Community Forum
🔌General Home Automation => Automating Your House => Troubleshooting Automation Problems => Topic started by: Lightmaker on March 16, 2012, 02:08:33 AM
-
Hello everyone.
I have a small but growing X10 setup with the following modules in use:
- X10 RR501 RF Transceiver (with its built-in appliance module)
- X10 LM465 Lamp module (new style with soft start)
- X10 Pro PLM21 2-Way Lamp module
- SmartHome 2002STW3 ApplianceLinc appliance module running some fluorescents (local control disabled)
- 5x HR12A PalmPad remotes
I recently replaced a TM751 RF Transceiver with the RR501, which works much more reliably with my two lamp modules at the far end of the house.
However, the RR501 has a strange quirk that each time a dim or brighten command is sent from a PalmPad remote, the command is sent TWICE over the power lines by the RR501. So each time I dim a light, I get to watch as it dims in two distinct steps, one after the other. It isn't a huge problem, but it is somewhat annoying visually, and the light control is more coarse with half of the dimmer settings being skipped. :-\
This isn't a lamp module issue, as it happens with both the LM465 and the PLM21. It doesn't look like a remote issue either, as all of my remotes have the same effect, no matter how quickly you press the buttons. With the TM751 transceiver in place, I can dim/brighten in single steps... but with the RR501, it repeats each command twice. :-\
Is this a known problem with the RR501? ???
-
All of the palm pads send their commands multiple times for signal integrity.
It depends on how the transceiver handles them.
I believe some versions of the RR501 may send it on the power lines twice.
I have an old CM15A and newer CM15A. One sends the commands multiple times the other only one time with a palm pad.
-
The problem is not in the RR501 and you'll see the same phenomenon with most X10 transceivers.
It's because nearly all of the handheld remotes send multiple copies of the RF code. Once the transceiver starts sending to the powerline it is deaf to further RF but, if the remote, is still sending RF when the transceiver finishes sending to the powerline, it will start anew.
This was the fundamental cause of a problem that used to exist in the TM751 where the programmer apparently took a shortcut that made it send PLC forever if it sensed any RF that would cause its superregenerative receiver to oscillate. Feedback from the 120kHz on the household wiring caused what I dubbed the "endless dim syndrome". If you did not see extra dims with your TM751, it may be that they've lengthened the time that it ignores RF.
The Security remote (SH624) and keyfobs (e.g. KR22A) send single RF codes if pressed/released quickly but most others send six or more RF copies which will usually cause more dimming/brightening than desired.
Each RF copy takes ~108mS while a powerline command takes about 1/2 second (varies with the command).
If you place the HR12A on a solid surface and then poke/release a key (like a bird pecking) as quickly as possible, you may be able to see this.
The real reason for sending multiple RF copies is the poor tuning/sensitivity of X10 transmitters/receivers. The extra copies help set the receiver AGC, improving the sensitivity after a copy or two.
There's a 'scope screenshot (non-X10 signal) showing a signal emerging out of the noise at the bottom of this web page...
- http://davehouston.org/rf-noise.htm
-
Thank Brian and Dave, I appreciate the informative replies. 8)
Brian-
Do you know how old your CM15As are, and which one of them responds favorably to the PalmPads? (Is it the newer one?)
My TM751 has a date code sticker on the back that reads "07F23", while my RR501's sticker says "08J43". So it looks like the RR501 is actually newer than the TM751... The RR501 is the smooth-front slightly rounded style, not the flat square style with the triangle-cut stripe of grooves on the left. (Does that description make sense?)
Do you know where I might find out how many revisions of the RR501 were made? I found another thread where a member lists the firmware version of their RR501 (date code: 02D16, firmware: P10283E), but they also claim it is on a sticker on the back. My unit doesn't have a firmware version sticker...
(The thread: http://forums.x10.com/index.php?topic=12826.0)
Dave-
Interesting technical analysis. I don't think I have seen that endless dimming problem with my TM751, although as I mentioned, I did seem to have general "not working" issues with the lamp modules at the far end of the electrical circuit. When it would quit working, the lights would always stick at whatever state they were in last - dimmed, full bright, off, etc. The RR501 seems to have completely solved this though, so I'm guessing it has a stronger output signal?
For the PalmPads, do you know if there is any way to change the number of repeats in the RF transmission? Since my modules all dim or brighten twice, I know empirically that it doesn't need to be sending that many repeats for the modules to receive and interpret them, even at the far end of the electrical run. ;)
Thanks again for the help. 8)
-
HR12A one fast tap of button. B1 On;B1 Off; B Bright and B Dim.
Monitored by a Smarthome 1132CU and Smarthome Essentials Software.
XTB-IIR repeater/coupler Off.
CM15A Date Code 05C10 Firmware P10792M:
R: B1 - 6:36:45 PM 3/16/2012
R: BOn - 6:36:46 PM 3/16/2012
R: B1 - 6:36:46 PM 3/16/2012
R: BOn - 6:36:47 PM 3/16/2012
R: B1 - 6:36:48 PM 3/16/2012
R: BOff - 6:36:48 PM 3/16/2012
R: B1 - 6:36:48 PM 3/16/2012
R: BOff - 6:36:49 PM 3/16/2012
R: BBright - 6:36:52 PM 3/16/2012
R: BBright - 6:36:53 PM 3/16/2012
R: BDim - 6:36:54 PM 3/16/2012
R: BDim - 6:36:54 PM 3/16/2012
CM15A Date Code 04J44 Firmware P10792E
R: B1 - 6:40:09 PM 3/16/2012
R: BOn - 6:40:10 PM 3/16/2012
R: B1 - 6:40:11 PM 3/16/2012
R: BOff - 6:40:11 PM 3/16/2012
R: BBright - 6:40:13 PM 3/16/2012
R: BBright - 6:40:13 PM 3/16/2012
R: BDim - 6:40:14 PM 3/16/2012
R: BDim - 6:40:15 PM 3/16/2012
RR501 Date Code 02F23:
R: B1 - 6:53:41 PM 3/16/2012
R: BOn - 6:53:41 PM 3/16/2012
R: B1 - 6:53:42 PM 3/16/2012
R: BOff - 6:53:43 PM 3/16/2012
R: BBright - 6:53:44 PM 3/16/2012
R: BBright - 6:53:45 PM 3/16/2012
R: BDim - 6:53:46 PM 3/16/2012
R: BDim - 6:53:46 PM 3/16/2012
RR501 Date Code 10G28:
R: B1 - 6:55:04 PM 3/16/2012
R: BOn - 6:55:04 PM 3/16/2012
R: B1 - 6:55:06 PM 3/16/2012
R: BOff - 6:55:06 PM 3/16/2012
R: BBright - 6:55:07 PM 3/16/2012
R: BBright - 6:55:08 PM 3/16/2012
R: BDim - 6:55:09 PM 3/16/2012
R: BDim - 6:55:09 PM 3/16/2012
TM751 Date Code 00D17
I have no clue as to why M got into the Dim and Bright.
I tired two of them with the same M transmissions.
R: B1 - 6:59:41 PM 3/16/2012
R: BOn - 6:59:42 PM 3/16/2012
R: B1 - 6:59:44 PM 3/16/2012
R: BOff - 6:59:44 PM 3/16/2012
R: BBright - 6:59:48 PM 3/16/2012
R: MDim - 6:59:48 PM 3/16/2012
R: BDim - 6:59:52 PM 3/16/2012
R: MDim - 6:59:52 PM 3/16/2012
TM751 Date Code 10E19:
R: B1 - 7:01:03 PM 3/16/2012
R: BOn - 7:01:03 PM 3/16/2012
R: B1 - 7:01:06 PM 3/16/2012
R: BOff - 7:01:07 PM 3/16/2012
R: BBright - 7:01:09 PM 3/16/2012
R: BBright - 7:01:10 PM 3/16/2012
R: BDim - 7:01:12 PM 3/16/2012
R: BDim - 7:01:13 PM 3/16/2012
-
@Lightmaker - I haven't seen any complaints about endless dimming for several years so X10 must have fixed their firmware. You would have to replace the µController in the HR12A with one you programmed yourself to change it's response to a button press.
@Brian H - you either need an itchier trigger-finger or a different interface doing the reporting. It was too far back for me to recall details but I tested using one DSO 'scope channel on a 310MHz RF receiver and the other on the powerline through Act's high-pass filter so I could see the number of RF copies and the number of Dims on the powerline. And, I could only get 5 copies about 20% of the time - it takes a very fast press/release. I documented it thoroughly at the time but that and a lot more were lost when a network HDD failed a few years ago. Again, I do not recall the details but believe the 1132 reports dim/bright differently than most X10 interfaces (i.e. I don't think it adhere's to the patterns I detailed at http://davehouston.org/micro-dim.htm). Also, you might try it using a keychain remote if you have one.
-
Dave-
I did see your post in the other thread:
...a single IC marked "HiMARK, RX3310A-LF...
That explains why you never saw the endless dim problem. The Hi-Mark RX3310A is a superheterodyne RF receiver that replaced the X10 designed superregenerative RF receiver that was at the heart of the problem.
So that is good to know. 8)
Brian-
I hadn't even thought about monitoring the power line to see what it is doing.
I do have a Smarthome 2414U ("SmartLabs PowerLinc V2 USB"), but I have never had the software for it. In the past, any time I have needed computer control I have just used a CM17A FirecCacker with Heyu under Linux. Of course that is not bidirectional though, so it wouldn't fit this need.
Do you have any suggestions for software for the 2414U? The offerings from Insteon are $137 (mControl on PC) and $190 (Indigo on Mac), which is way outside my budget just to troubleshoot a remote control issue. :o The Smarthome Essentials from your 1132CU doesn't appear to be compatible with the 2414U. Searching on the Linux side, I keep finding dead ends (such as the "not supported and probably never will be" from MisterHouse)... I tried ThinkingHome on the Mac, and while it detected the 2414U, it would crash as soon as it actually tried to use it. B:(
Based on what I've found in my searches, it looks like people generally end up purchase alternate hardware devices instead of finding software to make their 2414Us work. :-\
-
The 2814U Timer Essentials will work with the 2414U.
Not sure if it would do what you want. 2814U {Icon version of the 2414U} is discontinued but the software links on the page still work.
http://www.smarthome.com/2814UB/Smarthome-Manager-Essential-Timer-USB-by-Smarthome/p.aspx
Ken Millers InHomeFre also works with the 2414U, 2414S and 2814U. Ken stopped development on it after some issues with Smarthlabs but you still maybe able to find it. I used InHomeFre for my main basic control for awhile.
Both are for Windows. I have not seen much under Linux.
I also have a TM751 that starts Dim firestorms as it warms up. ???
-
Dave, I hit the buttons as briefly as I could. I will see if I can dig out a keyfob or try my UR19A and see if things look differently.
Also as you pointed out. Could be the way the 1132CU processes thing. I know the LED on the HR12A flashed more than two times.
-
... a CM17A FirecCacker with Heyu under Linux. Of course that is not bidirectional though, so it wouldn't fit this need.
If your software allows you to control the number of RF copies generated by the CM17A, it should demonstrate the difference with six RF copies and anything less than six RF copies.
What happens to the lamp level if you use it to send a Dim command? It's been far too many years for me to remember whether it automatically sends a certain number of copies or whether that is controlled via the supervising software.
@Brian H - a technique that is also helpful sometimes is to use a true RMS meter (or Kill-A-Watt) connected to the outlet of a lamp module. You can easily measure the steps in voltage supplied to the lamp. And, with a bit more precision than an eyeball allows.
-
No True RMS meter but I can use my Kill-A-Watt
-
You need a short extension to plug the Kill A Watt into the module outlet. Then plug the lamp into the Kill A Watt and measure voltage to the lamp.
I was surprised to find that my $25 no-name Right Tech 53408 meter made in Korea is true RMS.
-
So I looked all over for InHomeFre, without success. :-\ There are a LOT of sites and forum posts where people recommend it, but they all link back to a few of Ken's websites, which no longer exist. Do you know of any places where this software is still available for download?
I did however have some success on another front. A number of sites that talked about InHomeFre noted that the "SmartHome Device Manager" software was required for InHomeFre to work. Searching for SDM, I found this website:
http://fredricksensoftware.us/Insteon/
which has five different versions of it for download. I installed their recommended version, 235, and it detected my 2414U right away.
Luckily for me, the SDM software logs the activity on the power line! :)%
So here is a log, where I used the HR12A PalmPad remote and the RR501 transceiver to interact with a LM465 lamp module:
(Startup)
downloadTimer started
timerActivate enabled
SDM:DM=active,1.01.88.235,3/17/2012 1:14:50 PM,responded,11.F1.9F,2.13
("ON" for unit A2 pressed)
PLC:eventraw=08
PLC:receiveX10raw=00 6E
PLC:receiveX10=A2
PLC:eventraw=08
PLC:receiveX10raw=01 62
PLC:receiveX10=A On
("Dim" pressed)
PLC:eventraw=08
PLC:receiveX10raw=01 64
PLC:receiveX10=A Dim
PLC:eventraw=08
PLC:receiveX10raw=01 64
PLC:receiveX10=A Dim
PLC:eventraw=08
PLC:receiveX10raw=01 64
PLC:receiveX10=A Dim
PLC:eventraw=08
PLC:receiveX10raw=01 64
PLC:receiveX10=A Dim
("Brighten" pressed)
PLC:eventraw=08
PLC:receiveX10raw=01 65
PLC:receiveX10=A Bright
PLC:eventraw=08
PLC:receiveX10raw=01 65
PLC:receiveX10=A Bright
PLC:eventraw=08
PLC:receiveX10raw=01 65
PLC:receiveX10=A Bright
PLC:eventraw=08
PLC:receiveX10raw=01 65
PLC:receiveX10=A Bright
("Off" for unit A2 pressed)
PLC:eventraw=08
PLC:receiveX10raw=00 6E
PLC:receiveX10=A2
PLC:eventraw=08
PLC:receiveX10raw=01 63
PLC:receiveX10=A Off
So it seems as though the RR501 is forwarding the RF commands out onto the power line FOUR times, not the twice that I had originally thought. ???
So the HR12A sends SIX RF commands (as per Dave), which the RR501 hears and forwards to the power line as FOUR PL commands, which the lamp modules interpret as TWO commands? ???
Do "standard" powerline modules send and trap repeats of the signals for better reliability, like the RF controls do? (In other words, if I had one of the plug-in controllers with the buttons on it, and pressed "dim", would it only send it out on the power line once? Or would it send it multiple times quickly, just like the RF controllers do?)
-
I used SDM3 308 or 278. 308 was needed for the original HouseLinc and 278 was what the Developers Group reccommended.
I have InHomFre and have tried to send it to folks without much success. I also have Ken's SetSHPort that allows you to easily switch between the 2414U and 2414S.
I started my Insteon Developers Membership when the 2414U/S was the only controller made.
All X10 powerline commands are sent two times for better integrity.
Dave Houston has many good informational pages on his site.
http://www.davehouston.net/
Uncle Phil from ACT Inc also has lots of data on their site.
http://www.act-remote.com/PCC/uncle.htm
JV Engineering also is a trove of knowledge.
http://jvde.us/x10_troubleshooting.htm
-
So the HR12A sends SIX RF commands (as per Dave), which the RR501 hears and forwards to the power line as FOUR PL commands, which the lamp modules interpret as TWO commands?
There are two many cooks spoiling the broth. It's nearly impossible to pin down the source(s) of the unwanted dims or even get an accurate count w/o using an oscilloscope to simultaneously watch the output of an RF receiver and the PLC commands on the powerline.
I did most of this exploration 10-12 years ago and, as noted earlier, have lost my notes documenting details so we are stuck with what I can dredge up from 70 year old grey cells.
I always suspect how Smarthome interfaces report dim/bright (and how their switches/modules respond) as they do not follow the conventions I documented here...
- http://davehouston.org/micro-dim.htm
One quick and dirty test is to see how many quick press/release HR12A/RR501 cycles are required to go from full dim to full bright and vice versa. That will give you a fairly good idea of the number of PLC dims being sent each time. The test I suggested with a true RMS meter or Kill A Watt measuring the output of a lamp module is the best test as eyeballs have trouble at the extremes near full dim and full bright.
I doubt the RR501 is seeing more than 6-7 RF copies with a quick press/release and that should result in two standard steps (a triplet on the powerline, a PLC doublet is one step). The interface and/or software may be inventing the others by misinterpreting what's on the powerline. IIRC, they do not understand that the first dim/bright in any contiguous group is little more than a preamble and has only a 0.6% effect on the level of the lamp, in effect, interpreting a single dim/bright as two.
-
A meter would definitely be needed if the LM465 was Soft Start. They don't dim to 0%. They stop with about 14.2 VAC {non True RMS Meter} across a 40 watt bulb dimmed all the way down.
-
I will have to find my Kill-A-Watt and do some better testing... but neither my LM465 nor my PLM21 will dim to "zero", I can say that for sure. They hit a minimum threshold, and further dim commands are ignored.
By eye, both the LM465 and PLM21 appear to have 15 levels (counting the two ends, full bright and full dim). This is taking the double-step into account - there are really only 7 levels available to me. :-\
Curiously though, I did some more testing with SDM3 and the 2414U, and found that if I hit the buttons on the HR12A *REALLY* fast, I could get SDM3 to report either THREE or sometimes just TWO repeats of the dim/brighten command! I could not reliably repeat this though, so I wonder if it is a timing/sync artifact, or as Dave mentions, a difference or irregularity in how the commands are interpreted by the Smarthome box (the 2414U). My other thought was that it could even be a case of poor switch debouncing in the HR12A itself, and not necessarily (or entirely) a firmware issue?
Dave or Brian, do either of you happen to have an older RT504 ("Sixteen Plus") remote, or the equivalent Leviton 6313 ("Hand-Held Wireless Controller")?
It is times like these that I wish I owned an oscilloscope. :-\
-
My testing was done back in the days before the soft-start LM465 was available. I believe I did note the minimum level (~5%) of the LM14A which was the earliest soft-start on the page I referenced earlier.
15 levels are about what you can see with the naked eye. And 7 levels are typical with the HR12A. There were 17 (metered) levels with the soft-start LM14A.
I believe I probably sent the RF and/or PLC commands with software of my own and a small RF transmitter module and PL513 to eliminate the remotes when measuring lamp module output levels. You can send single PLC dims with the commandline and a CM11A but that's tedious.
Charles Sullivan and I compared notes and were pretty much in agreement. I probably would never have realized my cheap meter was true RMS had he not been using a more costly Fluke while I was using the Kill A Watt. I tried my meter and was pleasantly surprised that it agreed step for step with the Kill A Watt.
I have a dimmable CFL on an LM14A. I can get discrete levels with a KR22A and SH624 but always get multiples with an HR12A. It's in the room with my flatscreen TV and I use it more as a nightlight than anything else. I had to experiment with lamp modules and remotes to get soft-start that didn't start much higher and/or adjust much more coarsely than I wanted.
No, I don't think I ever had the RT504. I had another 16 address remote but can't recall who made it and cannot locate it so it either got culled or misplaced.
I have kept a very old W98SE PC going because the motherboard had both PCI & the ISA slots which my Protek 220 oscilloscope card needs. It will not work with newer flatscreens (I even tried a couple of graphics cards that claimed to work with W98SE) so I've also had to keep a CRT monitor, too. I'm planning to finally replace it in a couple of weeks with a reasonably priced USB 'scope that has nearly the same specs (and better low-level programing) so I can finally recycle the old monsters and reclaim some space on my workbench.
- http://www.pdamusician.com/dpscope/features.html
IMPORTANT: You need isolation when 'scoping X10 modules or the powerline to avoid getting a hot 'scope chassis.
-
Thanks Dave. I will certainly keep your safety warning in mind! :o
What did you think of the Protek 220? Aside from being trapped in the old computer, did it do everything you needed? You will have to keep me posted on how the newer USB one works out. Are you going to get the DPScope, or the DPScope SE?
Also, what have you used (if anything?) to monitor the RF transmissions? I have both a USB RF transceiver (CM19A) and the serial RF receiver (MR26A), but I have never used either one of them... I'm not sure if they are useful for this sort of low-level troubleshooting, or if they would simply mimic the results of the power-line transceivers?
-
I was happy with the Protek - it was more than adequate for almost anything that would arise in Home Automation. It did lack reliable triggering.
I'm just going for the DPScope.
I use RF receiver modules that let me see the RF signals directly w/o interpretation by X10 who filter necessary details. There are several shown here...
- http://davehouston.org/modules.htm
although some links are out of date. I also like the RM1SG (search eBay) as i's fairly wideband. All are easily tuned to 310MHz.