Jump to content

Repair of 2413S PLM When the Power Supply Fails


danu1964

Recommended Posts

Happy to report that my PLM V1.0 is back up and working.  Brian, your suggestion to tie the (-)'s of C7 & C13 worked.  Although when I powered it up, I got nothing.  No LED lights, nothing.  Pulled the unit apart again and then proceeded to replace C11 and C3, primarily because they were the next two easiest to replace.  When I plugged it in this time, Got a short red flash from the LED and then in a few seconds a continuous green.  Plugged in the ISY and opened the Admin Consol.  Low and behold, everything was registering and communicating.  We'll wait and see how permanent the repair is.  If I get another few years, I'll be thrilled.  Thanks to all on this board and especially Brian and Shawn (plm.repair@gmail.com).

  • Like 1
Link to comment
Share on other sites

Smartlabs has changed out several capacitors to help improve the two year failure rate. They have also removed the ALL ON / ALL OFF commands to help resolve that issue. All of the latest production hardware have also removed the same command sets.

 

 

=========================

 

The highest calling in life is to serve ones country faithfully - Teach others what can be. Do what is right and not what is popular.

Link to comment
Share on other sites

  • 1 month later...

Version 1.B here, I have C7 and C13 as the older 10uF 35V. Should I switch them to 100uf / 35V?

If you make sure the 100uF 35V replacements are a Low ESR type . They should work.

 

My new one from Smarthome seems to have the same questionable brand in it. Just changed from 10uF to 100uF.

 

I did not see any other component changes in later one i have with the 100uF 35V ones.

Using the original 10uF 35V Low ESR replacements would also be fine.

Link to comment
Share on other sites

If you make sure the 100uF 35V replacements are a Low ESR type . They should work.

 

My new one from Smarthome seems to have the same questionable brand in it. Just changed from 10uF to 100uF.

 

I did not see any other component changes in later one i have with the 100uF 35V ones.

Using the original 10uF 35V Low ESR replacements would also be fine.

 

Digikey label a lot of them as "General Purpose", found one with impedance of 250 mOhm. 

 

Sweet, I love the internet. All is well now, back up and running. I also cleaned a lot of flux on the bottom side of board, pretty sloppy work Insteon !

 

Digikey

C11 = 493-3276-ND  (100uF @ 25V)

C7,C13 = 1189-1860-ND (100uF @ 50V)

C3 = 493-12079-1-ND (10uF @ 400V)

C8 = 493-10329-1-ND (10uF @ 16V)

Edited by MolsonB
Link to comment
Share on other sites

  • 2 weeks later...

So nobody has been able to determine if this issue has been fixed yet with newer units?

They are just getting close to the age where older ones started to go bad.

So there is not too much data yet.

I do occasionally take the top off of mine and measure the unregulated 12 volt supply and look for swollen or leaking caps.

So far nothing found. About time to look again.

Firmware 9E {maybe higher now} was supposed to help with the All On events some where seeing.

Link to comment
Share on other sites

I just wanted to drop a note here... I replaced the caps on my old, broken 2413S (v1.5 with a build of 1130).

 

I followed this whole thread and went to Mouser to pick the parts.  Here's what I ended up getting which varied only slightly from the very first post:

1 x ESX106M400AH4AA 10uF 400V
2 x EEU-EB1H100S 10uF 50V
1 x USV1C100MFD 10uF 16V
1 x UHE1E101MED 100uF 25V
 
The only one I had to find an alternate for was the 10uF 16V since the one from the original post is no longer available.  The other ones were all in stock at Mouser when I ordered, and I got enough to replace 4 sets if I had to.  Makes sense... the cost per cap isn't high and you're paying for shipping anyway... might as well spend a couple extra bucks for spares.
 
The desoldering was the hardest part for me.  I already had a solder sucker and desoldering braid, but my solder kung fu was weak and it just took longer than I thought it would.  I did finally get all 5 done with some patience.
 
I did one cap at a time and I started with C7 & C13 since it sounded like those are the most likely to be the bad ones.  I was tempted to call it quits then and just see if it worked, but I figured I might as well persevere and replace the rest while I had everything setup.
 
In the end, I have a working 2413S PLM which is great.  It's been sitting in my parts bin for nearly 2 years (my new one is v2.0 dated 1444, so almost 2 years exactly).  I tested the old one with Houselinc software just to make sure it was communicating okay.  Paired it with a Lamplinc I had lying around and it seems fine.  I now have a good spare I can use if/when my new v2.0 dies.
 
There was one odd thing... when the repaired unit is sending signals, I can actually hear it.  I had it plugged into an extension cord right next to my keyboard while I was testing, and it kind of "buzzes" in time with any data it's modulating onto the powerline.  It's not super loud and if it wasn't right next to me I wouldn't have heard it.  I don't know if it ever did that before, or if anyone else has ever noticed.  I've heard electronics "buzz" in my experience and it could indeed be related to the caps.  It's not going to hurt anything and it still worked, and like I said, it's actually pretty quiet unless your ear is right there, but it was strange.
 
Anyway... chalk up another successful re-cap to this thread, and thanks to the original poster!
  • Like 1
Link to comment
Share on other sites

I find the best way to un-solder is to use my air compressor.  Of course you first get the solder liquefied, then "poof".  I use blue tape to mask off stuff I don't want solder to spray on.  But even if you don't the solder spray doesn't bond, you can just wipe it off with your finger.  It works very well and gets the holes through the board essentially 100% clear of solder in a split second.

 

Update on my repair.  Still working, more than 6 months online now.

Link to comment
Share on other sites

  • 2 weeks later...

First, let me say how much I love this thread and OP (danu1964) and Brian H for all their hard work.

Second let me say how disappointed I was to see everyone say they could be added to the list of people with broken and fixed devices only to find out there was no actual list...so here's your list...

 

danu1964 (OP) 3 failures at least 1 fixed
Panda88 1 failed 1 fixed
LeeFleishman 1 failed 1 fixed
LeeFleishman 2 failed 2 fixed 1 Preventive fix
Steven 1 failed 1 partial repair and fixed but didn't last
LeeFleishman 0 failed 0 fixed 1 Preventive fixed
apostolakisl 2 failed 2 fixed
ergodic 1 failed 1 fixed
fmark 1 failed 1 fixed
GeorgeRufle 1 failed 0 fixed
uy6n 1 failed 0 fixed
ertyu 1 failed
arw01 1 failed 1 fixed
lithiumus 1 failed
bfranske 1 failed 1 fixed
dmwolff 1 failed 1 fixed
taterb 1 failed 0 fixed
mcheavens 2 failed 2 fixed
whistlepigger 1 failed 1 fixed
ixlr8 1 failed 1 fixed
deaconwc 1 failed 0 fixed
wojo 1 failed 1 fixed
Kentinada 1 failed (I assume)
jgorm 1 failed
wrj0 1 failed 1 fixed
BCreekDave 1 failed 1 fixed
Jay 1 failed
avwuff 1 failed 1 fixed
tome 1 failed 1 fixed
chelmite 1 failed
jakekooser at least 2 failed (could have been a 3rd failed?)
jumon 1 failed 1 fixed
jackman 1 failed 1 fixed
2112 1 failed
benderdave 2 failed 1 fixed
dlorenzo 2 failed 2 fixed
i814u2 1 failed 1 fixed
dexterdom 1 failed 1 fixed
MikeD 1 failed 1 fixed
dacross 1 failed
Harold 1 failed
diggler 1 failed 1 fixed
jackery 1 failed
orrd 1 failed
thehose 1 failed
drinkwater 2 failed 1 fixed
plm.repair 1 failed 1 fixed
smbenson 1 failed
gravymaker 1 failed 1 fixed
franklyn7895 1 failed 1 fixed (You are way closer to the 66th than the 6th)
rjbur 1 failed
robr 1 failed 1 fixed
Grizzy 1 failed
eduardo_garcia 1 faied 1 fixed
zurakowg 3 failed 3 fixed
randyf 1 failed
BMonfils 1 failed 1 fixed
jondaddio 1 failed 1 fixed
BMFurrButt 2 failed 1 fixed (Counting your brother)
The_Penguin 1 failed 1 fixed
dimensionzero 1 failed 1 fixed
ptcolombo 1 failed 1 fixed
scuba357 1 failed
MolsonB 1 failed 1 fixed
MPB 1 failed 1 fixed
gruefood 1 failed
Brian H 3 failed 3 fixed (including 2 2443 Hardware 2.5 Access Points)
waffles 3 failed 2 fixed
jch 3 failed (2 2413S and 1 2412S if I read that right (Updated by jch))
Breezyken 1 failed and 1 fixed.
Morris Hansen 1 failed and 1 fixed IOlinc
tomd 1 failed and 1 fixed EZIO40 4
Pat525 1 failed and 1 fixed
jtara92101 1 failed 1 fixed
JackL 1 failed (I don't think you actually fixed it but only used a spare)
classicmac 1 failed 1 fixed
MarkB 1 failed 1 fixed
fmk2 1 failed 1 fixed 
billchurch 1 failed 1 fixed
m6bigdog 1 failed and fixed PLM and 1 failed and fixed AP

 

98 total confirmed failed and 63 total confirmed fixed.  If it wasn't completely obvious to me that you fixed yours I may have left yours off.  Just send me a note if you want to be added!
Edited by gruefood
Link to comment
Share on other sites

Seems R5 might have exceeded normal operating temperatures...I've ordered the caps for a fix but I'll probably have to replace R5 and maybe C6 as well.  I also ordered a new PLM that should be here tomorrow, (I love Amazon's under promise over deliver mentality, it arrived today), but would be nice to fix this one as a back up.

 

http://i.imgur.com/yZ1YEpn.jpg

 

Of course, there may be way more wrong that just what you can see...

Edited by gruefood
Link to comment
Share on other sites

Gee that doesn't look too good.

Kind of burned stuff partially clouding the run layout in the photo.

If C6 is in series with R5 and they are parallel to D6. It maybe a noise reducer across D6.

I would definitely look at C6 is failing.

Link to comment
Share on other sites

  • 2 weeks later...

Add me to the list of folks with a failed 2413S.  I ordered this one just about exactly three years ago.  Before that I went through two 2412S units in five years.

 

Caps are on order.

Link to comment
Share on other sites

Was looking in my cabinet of HA parts and found my last 2413S, it's a V1.0.  When it died I opened it up looking at why it was damaged.  Maybe I should try replacing the caps in that one too.

Link to comment
Share on other sites

I actually emailed INSTEON tech support on this issue,2 weeks later I received a reply stating they have  replaced the capacitors in question with higher grade capacitors  and this should  no longer be an issue.  I believe they are aware of this issue.

  • Like 1
Link to comment
Share on other sites

Based on comments elsewhere in this thread, the newer Insteon PLMs appear to have higher value capacitors, but alas, there's no indication that they are any higher quality (same manufacturer, same type, same physical size, just higher capacitance).

 

I, for one, am highly skeptical of their claim that the problem is fixed.

  • Like 1
Link to comment
Share on other sites

Unless this upgraded capacitor brand is very recent.

My V2.1 has the same questionable brand capacitors as my old V1.5.

Just went from 10uF to 100uf for C7 and C13.

 

Recently we have also seen a few reports of new 2413S PLMs failing in a few months.

Edited by Brian H
Link to comment
Share on other sites

Has anybody had s self repaired PLM fail yet? The repairs have been out there for two years now? I would think some would have failed by now.

 

 

Sent from my iPhone using Tapatalk

I also have not seen any reports yet.

 

If good quality name brand, low ESR capacitors where used.

I would expect them to run much longer. Over the questionable brand Smartlabs used in the PLM. Where they failed. Making the power supply die.

 

If the repair was not done well. With poor soldering, overheating PCB runs, solder shorts during capacitor replacement. Then I could see some early  failures or the repair failed completely.

 

If other parts where stressed due to the failing power supply. I could see some short time failures with the replacement capacitors.

Edited by Brian H
Link to comment
Share on other sites

I also have not seen any reports yet.

 

If good quality name brand, low ESR capacitors where used.

I would expect them to run much longer. Over the questionable brand Smartlabs used in the PLM. Where they failed. Making the power supply die.

 

If the repair was not done well. With poor soldering, overheating PCB runs, solder shorts during capacitor replacement. Then I could see some early failures or the repair failed completely.

 

If other parts where stressed due to the failing power supply. I could see some short time failures with the replacement capacitors.

Not sure if this applies but I had done only a partial repair of two caps. It lasted about a month before the others went. I have been running on the repaired plm since then. Added two vent holes top and bottom as well.
Link to comment
Share on other sites

 

First, let me say how much I love this thread and OP (danu1964) and Brian H for all their hard work.

Second let me say how disappointed I was to see everyone say they could be added to the list of people with broken and fixed devices only to find out there was no actual list...so here's your list...

 

danu1964 (OP) 3 failures at least 1 fixed
Panda88 1 failed 1 fixed
LeeFleishman 1 failed 1 fixed
LeeFleishman 2 failed 2 fixed 1 Preventive fix
Steven 1 failed 1 partial repair and fixed but didn't last
LeeFleishman 0 failed 0 fixed 1 Preventive fixed
apostolakisl 1 failed 1 fixed
ergodic 1 failed 1 fixed
fmark 1 failed 1 fixed
GeorgeRufle 1 failed 0 fixed
uy6n 1 failed 0 fixed
ertyu 1 failed
arw01 1 failed 1 fixed
lithiumus 1 failed
bfranske 1 failed 1 fixed
dmwolff 1 failed 1 fixed
taterb 1 failed 0 fixed
mcheavens 2 failed 2 fixed
whistlepigger 1 failed 1 fixed
ixlr8 1 failed 1 fixed
deaconwc 1 failed 0 fixed
wojo 1 failed 1 fixed
Kentinada 1 failed (I assume)
jgorm 1 failed
wrj0 1 failed 1 fixed
BCreekDave 1 failed 1 fixed
Jay 1 failed
avwuff 1 failed 1 fixed
tome 1 failed 1 fixed
chelmite 1 failed
jakekooser at least 2 failed (could have been a 3rd failed?)
jumon 1 failed 1 fixed
jackman 1 failed 1 fixed
2112 1 failed
benderdave 2 failed 1 fixed
dlorenzo 2 failed 2 fixed
i814u2 1 failed 1 fixed
dexterdom 1 failed 1 fixed
MikeD 1 failed 1 fixed
dacross 1 failed
Harold 1 failed
diggler 1 failed 1 fixed
jackery 1 failed
orrd 1 failed
thehose 1 failed
drinkwater 2 failed 1 fixed
plm.repair 1 failed 1 fixed
smbenson 1 failed
apostolakisl 1 failed 1 fixed
gravymaker 1 failed 1 fixed
franklyn7895 1 failed 1 fixed (You are way closer to the 66th than the 6th)
rjbur 1 failed
robr 1 failed 1 fixed
Grizzy 1 failed
eduardo_garcia 1 faied 1 fixed
zurakowg 3 failed 3 fixed
randyf 1 failed
BMonfils 1 failed 1 fixed
jondaddio 1 failed 1 fixed
BMFurrButt 2 failed 1 fixed (Counting your brother)
The_Penguin 1 failed 1 fixed
dimensionzero 1 failed 1 fixed
ptcolombo 1 failed 1 fixed
scuba357 1 failed
MolsonB 1 failed 1 fixed
MPB 1 failed 1 fixed
gruefood 1 failed
Brian H 3 failed 3 fixed (including 2 2443 Hardware 2.5 Access Points)
waffles 3 failed 2 fixed
jch 4 failed (2 2413S and 2 2412S if I read that right)
 
87 total confirmed failed and 52 total confirmed fixed.  If it wasn't completely obvious to me that you fixed yours I may have left yours off.  Just send me a note if you want to be added!

 

You can update me to 2 failed and 2 fixed.  I am running off of one of my fixed units for a while now, maybe 6 months.  I repaired it in 4/2014, tested it for a month, and then put it away as a backup.  The new one I purchased to replace it died after 2 years and 2 months, in June of this year.  So I swapped it out with my repaired backup unit and have been going fine since.  I then fixed the newer one and tested it for about a month before putting it in storage as a backup.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.

×
×
  • Create New...