Print

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

[JeffVolp]

Some of you know we have an Ocelot for our home automation controller.  That with the XTB-523 working as its powerline interface had provided virtual 100% reliability since we moved into this house.  But over the last several months reliability had deteriorated.  While still above 99%, occasionally the hall light controlled by an extended code was not at the proper brightness for that time of day.  Then over the last week it seemed like ghosts invaded the house, with lights switching on and off randomly.

Since years ago the power module had failed, I swapped in a new module.  But then the Ocelot would no longer communicate with the PC.  Hmmm...

I swapped in the backup I picked up on eBay.  It would not communicate with the PC either.  So thinking I had somehow damaged the serial port, I tried a different port.  Still no go...  Then I noticed the case on the Ocelot seemed warmer than usual.

I pulled the cover, and the electrolytic capacitor was hot.  Monitoring the rectified voltage with a scope showed the capacitor was not filtering the rectified voltage at all.  Checking the original Ocelot showed it had a similar problem with a dead electrolytic.  I replaced both electrolytics, and both Ocelots are again working as they should.

So, if you have an Ocelot that begins to misbehave, check the electrolytic in the upper corner.  If warm at all, it should be replaced.

Jeff

[dave w]

Everything gets old and starts to deteriorate. even us.....sigh.

[Brian H]

Thank you for the information.

Electrolytic Capacitors seem to frequently cause issues. Changing capacitors in many of my failing devices has fixed most of them.

[JeffVolp]

Everything gets old and starts to deteriorate. even us.....sigh.

Yeah, I'm getting old too, but I'm planning to do the rim-to-rim again in a few weeks.  I did it two years ago camping one night at the bottom, but this time I'm planning to do it in one day.

Jeff

[dave w]

but I'm planning to do the rim-to-rim again in a few weeks.  I did it two years ago camping one night at the bottom, but this time I'm planning to do it in one day.

You must be buying that "Texas Super Food" stuff advertised on the radio. <wink> My idea of strenuous exercise is the treadmill set to 3.8 mph. 
Be safe.

[brobin]

Had a similar problem with my Stargate last week. It couldn't control the alarm panel or the shade controller which are both connected to serial ports.  Common to those two ports is a single MAXIM 232 chip which JDS Technologies had the foresight install in sockets.  Popped in a new one and back in business.  I've had a number of devices where caps in the power supply failed.  The old ReplayTV's had that problem as well as the Insteon Hub and the Insteon 2413 PLM.  Seems like if they just spent a penny or two more and used caps with a higher WVDC they'd a avoid a lot problems. 

[dave w]

Seems like if they just spent a penny or two more and used caps with a higher WVDC they'd a avoid a lot problems. 

They did avoid problems for themselves. They bought caps that would outlast (barely) the one year warranty. 

[Brian H]

With the 2413 PLM and 2443 Hardware V2 Access Points.

They tried a few value and VDC rated capacitors. C7 and C13 in the unregulated 12VDC. Where originally 10uF/35V.
As of the V2.4 they are now 100uF/50V and a different serial board with a better ESD rated serial chip and some protective networks on the Serial Port pins to the outside world.

They had a $29.99 sale on the 2413S/2413U PLM last week and from all the folks receiving them. The serial version is now V2.5. The USB version may also be V2.5 now. My 2413S from the sale should be here next Tuesday. Needles to say it will be opened and observed when it gets here.

The 2450 I/OLinc also as had a few capacitor updates. Original C7 was 10uF/25V very close to the supply output voltage.
V1.8 was updated to a 10uF/50V. V2.3 that latest I have. C7 is now a 230uF/50V

The 2413 PLM and I/OLinc have a small switching power supply and I don't think all the band aid fixes may have used capacitors rated for a switching power supply.

[JeffVolp]

Seems like if they just spent a penny or two more and used caps with a higher WVDC they'd a avoid a lot problems. 

Actually, temperature is much more important than operating voltage as long as the capacitor is operating within its voltage spec.  A couple of technical papers I read in the past indicated life expectancy doubled for every 10C decrease in operating temperature below its rating.  Unfortunately, electrolytic capacitors are usually located near the heat generating components in the power supply.  And AC ripple also increases its core temperature.

In the Ocelot the electrolytic is adjacent to the regulator heat sink.  It is rated for 85C, well above what it should see during normal operation.  But using a 105C capacitor instead would increase its life expectancy by about a factor of four.  I installed 105C capacitors as replacements.

Jeff

[joesteed]

I use an Ocelot with an XTB-IIR interface, controlled by HALhms software on my computer. Received x-10 from the powerline started being reported erratically, often showing only a received house code A regardless of the command. Recently stopped reporting incoming x-10 altogether. Is this an example of the kind of behavior that could be related to the capacitor? If so, how complicated is the fix and can you recommend a specific capacitor to use? I mostly avoid replacing soldered components these days, but am not eager to buy a new ocelot if I can avoid it. Thanks!