apostolakisl Posted April 8, 2017 Posted April 8, 2017 Any reason to suspect that the radio would stop working after doing the fix? At least one of the two I repaired does not seem to have a working radio. The other I haven't tested yet. Of course I can't say if the radio ever worked on either of them since I have lots of dual band devices through the house which would put com onto the airwaves and vice-versa independent of a working radio on the plm.
Brian H Posted April 8, 2017 Posted April 8, 2017 (edited) The fix is in the power supply and not in the RF section. Though it maybe possible something in the RF area of the PCB got physically damaged during the power supply fix. I busted a PCB run in one of my repairs. In the power supply area and had to use a piece of wire wrap wire to bridge my problem. I do know that my Hardware 2.1 Firmware 9E PLM has RF issues with the old 2440 RemoteLinc's. It doesn't seem to process the RF communications between them properly. They have to use a different Dual Band Module to be process the RF and then use the power line back to the PLM. So it is also possible your problem one never worked 100%. Edited April 8, 2017 by Brian H
Brian H Posted April 16, 2017 Posted April 16, 2017 If anyone on this list has one of the post-January 2017 units, please crack that puppy open and let us know what's different. As a minimum, it would be nice to know what values they used for the capacitors we've all had to replace (C3, C7, C8, C11 and C13). I just received a Hardware: 2.3, Date Code: 4216, Firmware: 9E, PLM for a new project. Will try and give it a look in the near future.
Brian H Posted April 21, 2017 Posted April 21, 2017 (edited) 2413S PLMHardware 2.3, Date Code:4216, Firmware 9EC3 6.8uF/250V Can't see brand.C11 100uF/25V FujuconC8 10uF/16V FujiconC7 and C13. Now 100uF/50V Fujicon capacitor. { My Hardware V2.1 has 100uF/35V }Mounted parallel to the PCB with leads bent 90 degrees. Fastened with component glue so they don't move with normal everyday use.Physically too large to fit in PCB Pads location vertically.I had seen vague reports of the serial chip on the daughter board having issues.Not sure if the brand was changed or normal production parts availability change.It is now a TI brand MAX232EI. { My hardware 2.1 has a ST232 }. The TI MAX232EI meets the IEC61000-4-2 ESD Protection Standards on the RS232 I/O Pins to the outside world. Edited April 21, 2017 by Brian H
billchurch Posted May 29, 2017 Posted May 29, 2017 About to replace mine on a 2413U. I previously had a 2413S which I think I already chucked in the bin... Wishing I had saved it now. The 2413U is a v1.B, purchased in 2013, so almost 4 years... Not too bad. Will report back shortly. I also ordered 10 of everything... Probably should have ordered 30 of the EKZM500ETD101MHB5D, but oh well... I'm thinking I can kit these up as well and just do US mail ($0.42) for anyone else who might need them to save on the shipping.
Brian H Posted May 30, 2017 Posted May 30, 2017 (edited) If you have any of the hardware 2.0 and above 2443 Access Points. They are built with the same 2413 main board. They also also fail shortly after the 2 year old mark. Both of mine started sending garbage messages and the LED on the side was real dim. I had to repair both of mine with the same power supply repairs. My findings on what was changed in the hardware 2.3 is in the post 329 of this thread. Edited May 30, 2017 by Brian H
billchurch Posted June 1, 2017 Posted June 1, 2017 She's back up and running! Another success... The 2413U v1.B for the record. Album of some repair photos: http://imgur.com/a/8vb9Z
billchurch Posted June 2, 2017 Posted June 2, 2017 I made a blog post about it with some pictures, not by any means a guide but might help some out? https://1projectaweek.com/blog/2017/6/2/repairing-a-failed-insteon-powerlinc-power-line-modem-plm-2413u
fasttimes Posted June 2, 2017 Posted June 2, 2017 My PLM still works, but it is unreliable. Have people been able to repair their working but unreliable PLM by replacing components?
Breezyken Posted June 2, 2017 Posted June 2, 2017 Yes, as the capacitors degrade the DC voltages decrease and the device become unreliable. You can open it up and check the voltages to confirm the caps are failing. I posted earlier a schematic in this forum of the power supply with the voltages listed. You may want to just go ahead and replace those caps.
fasttimes Posted June 2, 2017 Posted June 2, 2017 (edited) Are there any how-to guides buried in here? I have a meter (and basic knowledge on using it) and can handle a soldering iron. Ah, found this http://forum.universal-devices.com/topic/13866-repair-of-2413s-plm-when-the-power-supply-fails/page-15?do=findComment&comment=203182 But I just realized I have a 2413U. Do they use the same caps? And this! Just published today in fact! https://1projectaweek.com/blog/2017/6/2/repairing-a-failed-insteon-powerlinc-power-line-modem-plm-2413u Edited June 2, 2017 by fasttimes
Breezyken Posted June 2, 2017 Posted June 2, 2017 You found the info necessary to do the repair. The latest list of part numbers in post #333 should be good and you may be able to buy them from him and avoid the shipping that costs as much as the parts. There are different versions of 2413, but they are similar except for the very early version. The only difference between the 2413U and 2413S is the plug-in daughter board that contains the USB or Serial interface. Let us know how you make out.
Brian H Posted June 2, 2017 Posted June 2, 2017 (edited) Same base 2413 PLM main board in the 2413U,2413S and hardware V2 2443 Access Points. Same capacitors are in all three. Later hardware had C7 and C13 changed from the 10uF/35V to 100uF/35 volt. The latest V2.3 has 100uF/50V but due to size. Laying horizontally against the the board with a small dab of component glue to keep them from moving. I would go with 100uF 35 Volt or 50 Volt even if my originals are 10uF/35 volt. Only exception would be the original 2413 hardware V1.0. They only had one capacitor on the board. They added a coil and capacitor found in later revisions. As rework hanging freely in the air with fly wire jumpers to the main board. I also had to do my pair of 2443 Access Points. Flaky operation and LED real dim Edited June 2, 2017 by Brian H
billchurch Posted June 5, 2017 Posted June 5, 2017 You found the info necessary to do the repair. The latest list of part numbers in post #333 should be good and you may be able to buy them from him and avoid the shipping that costs as much as the parts. There are different versions of 2413, but they are similar except for the very early version. The only difference between the 2413U and 2413S is the plug-in daughter board that contains the USB or Serial interface. Let us know how you make out. If he doesn't have any I have some as well, just PM me our info and I can send them by US Mail...
m6bigdog Posted July 12, 2017 Posted July 12, 2017 FWIW, the Insteon Model 2443 Dual-Band Access Point has the same power supply board as the PLM and fail in the same way. I just repaired my PLM 2413S V1.B, date code:1325 (while waiting for a new one to arrive) by replacing C7 only with a 10uf 50v cap (105degC) that I had in by parts bins. I did a lot of Electronics Engineering, to include Switching Power Supply design, during my 40 year electronics career, hence I have tons of spare parts.) While I was testing/checking the system , I saw one of the 6ea. Access Points in my system had a dead LED. So, I pulled it apart and sure enough it has the same power supply and C7 was bad so I replaced it and the Access Point was back online. C7 is the first cap in the Pi filter, after D6, the half-wave rectifier and has the most ripple current; hence it will get the hottest from ESR. This power supply a DC to DC converter with a 200khz step-down transformer drive circuit.
Brian H Posted July 13, 2017 Posted July 13, 2017 (edited) Thank you for reinforcing the data posted, in this very long thread. I too have had 2443 Hardware V2 Access Points {V1 hardware are the older design with power transformer supply} have the same bad caps and very low unregulated 12 volt supplies. Usually around 18-20 VDC. Mine where <7VDC and the logic supply IC was out of regulation. Fixed a few myself. Even in my newest {V2.3} PLM they are still using the same questionable brand caps. C7 and C13 are still 100uF but upped the working voltage to 50VDC. Older hardware 2413 PLM and hardware V2 Access Points have the 10uF 35VDC C7 and C13. They are mounted horizontally on the PCB and held with a glob of component glue. Physically to large to mount normally vertically to the PCB. My V2.3 also has a new RS232 interface chip on the Daughter Board. Has a IEC61000-4-2 ESD Protection rating. Edited July 13, 2017 by Brian H
heyfrank Posted September 7, 2017 Posted September 7, 2017 Looks like my PLM just fried. Going on my 4th or 5th one, will insteon ever fix this issue?
Brian H Posted September 7, 2017 Posted September 7, 2017 In the 2413S Review Section. They claim that a problem component was found and corrected in V2.3. I have not had mine in use to see if the two year failures will be happening. I know the RS232 chip on the serial daughter board was updated and the caps where a higher voltage rated.
Teken Posted September 7, 2017 Posted September 7, 2017 Looks like my PLM just fried. Going on my 4th or 5th one, will insteon ever fix this issue? Considering the hardware revision has moved from 2.0 to 2.3 ~ someone is trying. But, its clear to me the lack of any serious engineering standards are being applied. This comes across to me either the powers that be are incompetent or out and out just don't care. They see the 2413S PLM as a regular revenue stream. I am at a complete loss how any company would allow one of the primary components in any serious home automation system continue to have a black eye toward the Insteon brand. I have to assume the new guy who bought the company is working on this or is completely clueless to this problem.
Brian H Posted September 7, 2017 Posted September 7, 2017 I am now also seeing HUBs going South. I believe at two years and a few months. Since it has the same small switching power supply chip and same brand capacitors as a 2413S. I was not surprised. What gets me is. If some penny bean counter has allowed a slightly more costly quality capacitor. Their two year warranty cost savings. Would easily covered the small BOM price increase.
apostolakisl Posted September 7, 2017 Posted September 7, 2017 I am now also seeing HUBs going South. I believe at two years and a few months. Since it has the same small switching power supply chip and same brand capacitors as a 2413S. I was not surprised. What gets me is. If some penny bean counter has allowed a slightly more costly quality capacitor. Their two year warranty cost savings. Would easily covered the small BOM price increase. Have the hubs really been out two plus years already? Man time flies.
Teken Posted September 7, 2017 Posted September 7, 2017 I am now also seeing HUBs going South. I believe at two years and a few months. Since it has the same small switching power supply chip and same brand capacitors as a 2413S. I was not surprised. What gets me is. If some penny bean counter has allowed a slightly more costly quality capacitor. Their two year warranty cost savings. Would easily covered the small BOM price increase. Yes, I saw mention of a few on the older SH forums as well. I was a little taken aback considering it was a brand new product and *Thought* someone would have sat down and said: Lets make this a real solid and kick aszz value controller Sadly, like many other products from Smartlabs it seems once its been released into the wild there is very little in ways of updates. This single life release ideology really makes me shake my head as to what drives this company. We already know Joe Dadda and team were completely hopeless in almost every way in terms of being serious in the HA field. Only a fool would have sat back and let others (Z-Wave / ZigBee) take market share every single year!! As of this writing I don't see any positive movement from the new guy who bought this company besides go toward the cheap end. From firing previous Devs to outsourcing all sales and technical assistance to the freaking Indians. HOW CAN YOU MAKE THINGS WORSE: Have people who can't speak or understand English . . .
The_Penguin Posted September 16, 2017 Posted September 16, 2017 I am now also seeing HUBs going South. I believe at two years and a few months. I just lost an outdoor on/off module. The old non dual-band 2456SE. Dead as a doornail. It's been plugged in outside since Jan 2014. I'll post a new thread about the repair, when complete. Looks like 3 bad caps. 6.8uf @ 250v 10uf @ 50v and 100uf @ 25v. I've replaced the 10 and the 100, don't have a 6.8 in stock so it'll have to wait until my next Digikey order. Had to saw it open with a Dremel cutoff wheel, not pretty. Don't think there's a more civilized way to open these.
fisix Posted September 26, 2017 Posted September 26, 2017 Hmm. You know, if all we need is 5V... those usb charger warts that Apple ships with their iPhones and iPads are very nice, and very small, and they seem to last forever (as long as you don't get one of the fake aftermarket knockoffs). I haven't looked inside to trace where the communication signals get patched back into the power lines, but would it be possible to just remove the power supply circutry that's in there and replace it with one of those warts? I think they're about the same size as the transformer in there.... any other voltages generated by the current PS? Anyone know what the current draw is on the 5V rail? Just a thought. The engineering Apple (and others) have done on those small warts is superb, and it'd be great to leverage it for something as important as the PLM. I've had to repair old Gb switches (that I didn't want to replace) with crappy PS units, and I've had good luck just finding a contemporary PS that generates the right voltages, can supply more than the rated current, and fits within the case. Newer isn't always better, but from the right manufacturer with the right philosophy, it almost always is. For these PLMs, I'd contemplate something that didn't fit within the original cover and just 3d printing a tumor-like mesh cover to keep fingers from shorting things unintentionally. Unfortunately, such fix wouldn't generally work with other devices that have to fit within a typical electrical box...
stusviews Posted September 26, 2017 Posted September 26, 2017 Hmm. You know, if all we need is 5V... those usb charger warts that Apple ships with their iPhones and iPads are very nice, and very small, and they seem to last forever (as long as you don't get one of the fake aftermarket knockoffs). I haven't looked inside to trace where the communication signals get patched back into the power lines, but would it be possible to just remove the power supply circutry that's in there and replace it with one of those warts? I think they're about the same size as the transformer in there.... any other voltages generated by the current PS? Anyone know what the current draw is on the 5V rail? Just a thought. The engineering Apple (and others) have done on those small warts is superb, and it'd be great to leverage it for something as important as the PLM. I've had to repair old Gb switches (that I didn't want to replace) with crappy PS units, and I've had good luck just finding a contemporary PS that generates the right voltages, can supply more than the rated current, and fits within the case. Newer isn't always better, but from the right manufacturer with the right philosophy, it almost always is. For these PLMs, I'd contemplate something that didn't fit within the original cover and just 3d printing a tumor-like mesh cover to keep fingers from shorting things unintentionally. Unfortunately, such fix wouldn't generally work with other devices that have to fit within a typical electrical box... What you're talking about is the power supply for the ISY which can be anything from 5VDC to 35VDC. I'm currently using a 12VDC supply. But this topic is about the PLM power supply which is not external to the PLM. Rather, it's internal parts.
Recommended Posts