apostolakisl Posted January 13, 2018 Posted January 13, 2018 I am in the process of remove/replacing my older switches with new ones. After removal I bench tested the pulled devices. I just pulled 4 V4.5 switches and 3 of them do this odd behavior. First off, this is a factory reset switch. The deal is, when the switch is de-powered for any length of time (allowing all the caps to discharge), upon re-applying power, it seems to go through this "powering up" phase that lasts several minutes. The led indicators keep "charging up" as seen in the video until finally the switch works properly. First just the bottom led, then 2, then 3, etc. I am guessing bads caps? Any ideas? And, yes, the load wire actually does what the leds indicate (if you had a light attached it would flicker). I made a youtube video and will link it shortly. Feel free to jump forward in the video, no need to watch all 6 minutes continuously.
Teken Posted January 13, 2018 Posted January 13, 2018 Yes bad caps and sometimes a combination of zener diode and the timing crystal. On the dimmer there are a couple of SMD resistors that go bad. Sent from my iPhone using Tapatalk
apostolakisl Posted January 13, 2018 Author Posted January 13, 2018 I suspect it is a cap. I wouldn't expect a resistor to "warm up" and then work normally.
Teken Posted January 13, 2018 Posted January 13, 2018 I suspect it is a cap. I wouldn't expect a resistor to "warm up" and then work normally. SMD components will react differently than analog devices when energized. Similar to Zener diodes when the break over voltage exceeds the spec odd things happen. For a SMD resistor if the component ever over heated for a brief period it's tolerances are compromised and thus will operate over / under its defined state. Which in a complex circuit provides / does not provide the reference operating voltage other systems are looking for. Most engineers plan and design their systems to operate in a wide range to offer a larger buffer. Before the multi voltage (100 ~ 277 VAC) Insteon PSU. Prior models had poor voltage and current handling capability never mind a reasonable amount of buffer tolerances. Cheap and poorly engineered products was the goal back in 2007 ~ 2014 under Joe Dadda & CO. Sent from my iPhone using Tapatalk
apostolakisl Posted January 13, 2018 Author Posted January 13, 2018 whatever it is, it is obviously the same thing on all 3 of them. If it is just a simple cap or diode replacement, I could fix them in 5 minutes. But I would need to know what part.
Teken Posted January 13, 2018 Posted January 13, 2018 whatever it is, it is obviously the same thing on all 3 of them. If it is just a simple cap or diode replacement, I could fix them in 5 minutes. But I would need to know what part. If I can locate the pictures from my blog I'll upload the components that failed here. Sent from my iPhone using Tapatalk
apostolakisl Posted January 13, 2018 Author Posted January 13, 2018 Just took it apart. It is a little harder than I anticipated because they riveted the voltage regulator to the casing. Not sure how to best handle that. Might be more trouble than its worth.
Teken Posted January 13, 2018 Posted January 13, 2018 Just took it apart. It is a little harder than I anticipated because they riveted the voltage regulator to the casing. Not sure how to best handle that. Might be more trouble than its worth. I just drilled it out took about ten seconds with a good metal bit via dremel. You'll need a steady hand or a vice because you don't want the bit to walk and mar the case. Don't ask how I know . . .
apostolakisl Posted January 13, 2018 Author Posted January 13, 2018 .120 in drill bit is perfect. Doesn't touch the case at all, just cleans out the rivet.
Brian H Posted January 13, 2018 Posted January 13, 2018 Some of the dark areas maybe PC runs on the back of the board or between the front and back sides. If there is a run in between the back and front. I have seen some Insteon models in the areas where power handling components like zeners get dark from the constant heating. Some of the early designs. Didn't have enough heat sinking copper runs and they got hot. Had a few 30 volt zeners. In 2456S3 and 2456D3 modules burn up and fall off of the PCB.
apostolakisl Posted January 15, 2018 Author Posted January 15, 2018 Is this the end of the line on this thread?
Brian H Posted January 16, 2018 Posted January 16, 2018 Just looked at your video. That looks like a power supply problem. Capacitors are always my first choice of a failure. When strange multiple symptoms are found. I had one fail in a 2456S3 ApplianceLinc. Pulsed On and off at about a one second cycle. I believe the 2476D switch has two boards in it and you may find some electrolytic capacitors on one of them.
apostolakisl Posted January 16, 2018 Author Posted January 16, 2018 Just looked at your video. That looks like a power supply problem. Capacitors are always my first choice of a failure. When strange multiple symptoms are found. I had one fail in a 2456S3 ApplianceLinc. Pulsed On and off at about a one second cycle. I believe the 2476D switch has two boards in it and you may find some electrolytic capacitors on one of them. Just one board. The picture is the backside, hidden behind the paddle. The other side is visible through the case, but here is a pic of it anyway. There are two big obvious capacitors. There are I believe 16 total as the highest "c" number is c16. But those little dinky ones probably don't have anything to do with the power supply.
Brian H Posted January 16, 2018 Posted January 16, 2018 I would look at the large black cased electrolytic marked RK on the right edge of the photo. The large Mylar cap closer to the left edge. Is used to limit the current into the power supply. Though they can fail. The large electrolytic would be my first thing to look at. I had a few {different brand} fail in 2457S3 ApplianceLincs. They got strange and one started cycling On and Off at a one second cycle.
apostolakisl Posted January 16, 2018 Author Posted January 16, 2018 I would look at the large black cased electrolytic marked RK on the right edge of the photo. The large Mylar cap closer to the left edge. Is used to limit the current into the power supply. Though they can fail. The large electrolytic would be my first thing to look at. I had a few {different brand} fail in 2457S3 ApplianceLincs. They got strange and one started cycling On and Off at a one second cycle. I spent $1.28 having 10 of those caps sent from China (tell me there isn't some crazy subsidizing going on here). Grant you, these are probably not great caps, but I assume that at least when new they will work properly. If they actually fix the switch, I'll go ahead and get some nice quality ones. I'll need to figure out how to re-rivet the heat sink on the voltage regulator back to the case to re-assemble. It doesn't look like a standard rivet will have clearance.
mwester Posted January 16, 2018 Posted January 16, 2018 I spent $1.28 having 10 of those caps sent from China (tell me there isn't some crazy subsidizing going on here). Grant you, these are probably not great caps, but I assume that at least when new they will work properly. If they actually fix the switch, I'll go ahead and get some nice quality ones. I'll need to figure out how to re-rivet the heat sink on the voltage regulator back to the case to re-assemble. It doesn't look like a standard rivet will have clearance. I did the same with the capacitors -- I rationalized that I would replace the cheap one with the proper low-ESR cap if the repair worked. That was six months ago... I didn't re-attach the SCR/Triac back to the heat sink, since I expected that I would just be opening it up again to replace the capacitor. Instead I tagged the red load wire with a label to remind me not to use a load on this switchlinc, and put it in service in one of my outbuildings as one of the switches without connected loads in a 4-way light switch arrangement. All is well... and I suspect that at this point I'll just leave it for the next year or two until the replacement capacitor fails.
apostolakisl Posted February 2, 2018 Author Posted February 2, 2018 Got the caps. Replaced the big one. It worked.
Tim McDermott Posted February 4, 2018 Posted February 4, 2018 Just to be clear, we are talking about replacing the "black cased electrolytic marked RK," and not the mylar, correct?
apostolakisl Posted February 4, 2018 Author Posted February 4, 2018 Just to be clear, we are talking about replacing the "black cased electrolytic marked RK," and not the mylar, correct? yes. I also had another switch of that same circuit board which was flat out dead and replacing the cap fixed it too. A 5th one that had a different board but the same cap was also dead and stayed dead after replacement.
Teken Posted February 5, 2018 Posted February 5, 2018 From personal experience all of the capacitors need to be replaced for various reason(s). That can include problems relating to the switch buzzing, LED's ramping up / down, not responding to the AC, to losing its programming.
Brian H Posted February 5, 2018 Posted February 5, 2018 I always suspect electrolytic capacitors. My test 1132CU stopped sending X10 signals. Swollen, bulging capacitor. My LED monitor. Getting touchy. 8 capacitors in it. Six swollen, bulging and one blew electrolyte on to the PCB. I have also have had 30 volt surface mounted Zener diode issues. In the older hardware designed Insteon modules.
apostolakisl Posted February 6, 2018 Author Posted February 6, 2018 None of the caps I replaced had any physical signs of failure. Still need to figure out how to rivet the voltage regulator heat sink back to the body of the switch.
apostolakisl Posted June 24, 2018 Author Posted June 24, 2018 Just did the same repair to a KPL. It had failed as if it were getting no power. I opened it up to see what caps it might have, and it has the same cap. I replaced that one capacitor and the KPL is back up and running. Again, I wish there were away to rivet the heat sink back to the body. Any ideas?
Recommended Posts
Archived
This topic is now archived and is closed to further replies.