Print

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

[andyd]

I have several SS13A Stick-A-Switch units.  Four are currently in use and I have several unused spares.   I have been unimpressed by the reliability of these units getting commands to my CM15a so I decided to measure the output frequency of several of them.  The nominal X10 frequency is 310.0 MHz.  I found all the measured SS13A were higher than this - 312.56. 312.58, 312.90, 312.88, 312.83 for the ones I have in front of me as I type.  It's hot here in Arizona and I wondered if they were all off frequency because they were warm.  I put one in the freezer and it increased in frequency when very cold so that didn't explain the frequency error.

I decided to try to get at least a few of my SS13A on the right frequency.  I started with a unit that was almost useless at 312.35 and tweaked it to 309.68.  It now works well and has a useful range.  I tweaked a second unit from 312.56 to 309.67 with similar results.

Opening the SS13A requires gently prying open the 6 latches that hold the back in place.  This can be done with a small flat blade instrument screwdriver.   The visible back side of the circuit board has a small hole that allows access to capacitor VC1.  Mark the current position before touching it.  The capacitor change required to shift the frequency is tiny.  It may take several attempts to get an acceptable result but it seems to hold once set.

Given that all the measured units started at 312.xx I have to wonder if the test equipment used to calibrate them in the factory was in error.

Note that VC1 appears to be quite fragile.  One unit did not work after attempting to adjust the frequency.  I found a fractured part protruding from the side of the capacitor.  I'll need to find a replacement capacitor to get this one working again.

Andy

[Brian H]

Thank you for the information.
Most  of the X10 transmitters are not that close to frequency. Except the later modules with a 310MHz Saw Resonator for frequency determination.

The power line transmitters are also not too close to the 120KHz frequency on some modules I have.

The schematic of a SS13A is in the FCC Database along with internal photos and a parts list.
The schematic in the database does not sound like yours. It is possible it is a later or earlier hardware revision SS13A.
https://apps.fcc.gov/oetcf/eas/reports/GenericSearch.cfm
Grantee:B4S
Product Code:SS13A
Pick details tab.

[dhouston]

You may also find this helpful.
https://www.laser.com/dhouston/X10_feng_shui.html

[andyd]

"The schematic of a SS13A is in the FCC Database along with internal photos and a parts list.
The schematic in the database does not sound like yours."

Yes, I had looked at all the FCC data and found that it does not apply to my SS13A units.   The circuit is similar but not the same.  The FCC circuit is dated 1999 and all my units were purchased much later than that.

If I need better RF performance I'll add a resonant dipole antenna to the CM15A but it seems to be ok with the re-aligned SS13A.

I measured a KR19A at 310.2 and two HR12A at 309.9 and 310.5, so the SS13A stand out as being badly off frequency.

The reason I got into this is a bit convoluted - I'm an amateur radio operator so lots of RF in the house.  I found that operating on the 6 meter band caused the clock of my CM15A to gain time which messed up several time sensitive functions.  I opened the CM15A and put a clamp on ferrite on the mains power wires.  That seems to have fixed the clock problem but it made the CM15A less responsive to my SS13A switches.  I assume that some of the SS13A signal had been getting to the CM15A by power line coupling and I killed that with the ferrite which forced the CM15A to reply completely on the antenna coupled signal.  All my SS13A are wall mounted right next to the hardwired light switches so they (unintentionally) couple RF into the house wiring.

Andy

[Brian H]

Some of us have added four decoupling capacitors on the main board power supply and IC's of the CM15A to improve performance.
It is the only design where I have seen no normally used decoupling capacitors. I know the old X10WTI was make it as cheap as possible but decoupling capacitors is a bit overboard.

https://www.laser.com/dhouston/CM15A.pdf

[andyd]

Thanks for the tip on the decoupling caps   I'll get that done ASAP.  This morning my clock was on time but my macros did not run.  That was after transmitting on 6 meters yesterday.  The ferrite seems to have helped but is not a complete fix.

The last time I saw the macro failure I did a full reload of the CM15a.  This time I connected the computer, opened AHP, checked the macros ran ok, closed AHP, and disconnected the computer.  The macros now run ok.   

Andy

[Brian H]

Forgot to mention. The CM15A is also in the FCC Database. Including photos and schematic.
It is for the earlier ones with a separate RF Receiver and separate RF Transmitter Daughter board.
The latest one has surface mounted IC's and the RF features are also on the same main board.

[andyd]

Thanks.  The circuit diagram is a lot clearer in the FCC original. That modified one was hurting my head!.  I have 3 CM15a all of the type shown in the FCC docs.  One is back in service with ferrite and decoupling cap mods.  We'll see how it holds up to my 6 meter transmissions.

Back to the original topic - I tweaked some more SS13a.  I found it was easier to close the case if the board was held in place with masking tape.  Without that it can pop out of position as the rear plate latches are being teased closed.

I also noticed that the transmitted frequency is higher when dim/bright is activated than when 1 on/off is activated.  I attempted to bracket 310.0 MHz.

Andy

[Brian H]

I am glad it help.
I am not too surprised the frequency shifted with different commands. A string of Dims would keep it On longer and I don't think the circuits are that stable.

Some of us also did an antenna mod on the CM15A.

[andyd]

Image of SS13A open for frequency adjustment.  Note 6 tabs on case back and location of VC1 adjustment. Masking tape holds board in place when back replaced.  It is not original fit.

Andy

[andyd]

It turns out that the frequency of the SS13a is not so much dependent on which button is being pressed  It is much more dependent on the hand/finger position when the button is pressed.

The oscillator inductor is a circuit board track that runs close to the switch buttons.  Hand capacitance shifts the oscillator frequency (more hand, more capacitance, lower frequency). 

The shift due to hand capacitance is much smaller than the frequency error that prompted me to adjust my switches.

Andy






Andy

[Brian H]

Thank you for the added data.
I am sure it may assist others.