The following solution applies to all modules that do not drive a load (like a lightbulb or a motor). It is relatively expensive and cumbersome, but is very easy and straightforward. The transformer solution can be applied to:
- computer interfaces: CP290, CM11A
- Powerflash module
- Universal module. Here, a dry contact is made and the load driver is external.
- Chime module
- Controllers: mini controller, maxi controller, Telephone transponder
Use a 220V-110V step down transformer and add a capacitor of around 0.1uF/600V across input output so the 120 kHz can be transfered. Capacitor should be polyester, MKT or similar and 400V unit would work.
In 220V
--------o-------,|||
| )|||
0.1uF
| )|||
600V
--- )|||
--- )|||
| )|||
| )|||
Out 110V --------o-------)||| Transformer
)|||
)|||
)|||
)|||
)|||
Common 0V ----------------'|||
Making a replacement CM11A cable
The following is a description of the CM11A pinouts, for a RS232 version.
- The notation for RJ11 (phone style plug) is as follows: Viewing the plug from it's contact side, the wire order is from left to right:
Yellow-Green-Red-Black
-
The notation for the RS232 pin is a standard D9 female plug, usually marked on the plug itself.
RS232 pin RJ11(phone style) pin
2 Yellow
3 Red
5 Black
9 Green
source: Dave Huston, post in Newsgroup.
There are four types of lockups. Two can be easily induced,
another is fairly easy to induce, and one is fairly infrequent but is the hardest to
cure.
1. This one can be induced by using another transmitter to send an old style extended code without any following extended data. This will lock up all CM11As 100% of the time. It can be cleared by using the other transmitter to put any command on the power line.
2. This one occurs when an older CM11A (older than about 24 months - with firmware 7 as returned by the Status request detailed in the CM11A protocol documentation) and a two-way module such as the LM14A are powered up simultaneously (e.g. after a power outage). This also will lock up the older CM11As 100% of the time. It can also be cleared by using another transmitter to put any command on the power line.
3. A third type occurs when the RS-232 cable is disconnected from the PC but not from the CM11A. This frequently, but not always, causes lockups and other weird behaviors. The lockups can usually be cleared in the same manner as above. This is probably the most frequent cause but it was the last to be identified and documented.
4. A fourth type is fairly infrequent and seems to be related to a dangling RS-232 cable and/or brownouts. It cannot be cleared in any manner other than removing the CM11A from the power line, removing the batteries and waiting a lengthy period. In most case 45 minutes is adequate but, in some cases, it takes longer.
All are totally independent of the software used. None have ever been acknowledged by X-10 but we've come to expect silence from X-10.
Credit: Jerry Hancock, Personal Correspondence
If the symptom is that the CP290 is able to load programs and it looks like it is is working but nothing is coming out on the line - try this fix. C1 (.222uf) was blown. This is the capacitor near the transformer that puts the signal on the line. Replaced with two high voltage .1uf caps available anywhere.
Improved: Increasing the clock accuracy of the CP20 by adding a timer chip
source: Steven Bloom , comp.home.automation
Description:
This modification is inexpensive (about $8.58), easy to implement (about 15 min, which is
quicker than the the following mod), easy to calibrate (none required), and improved
(garanteed to +/- 1 minute per YEAR @ 0Cto 40C).
The mod is based on a neat chip from Dallas Semiconductor called DS32KHZ which is a
temperature compensated crystal oscillator (TCXO).
Procedure:
- cut the trace to pin #39 of the 80C48.
- "deadbug" glue the DS32KHZ chip onto the top side of the circuit board (there is a nice blank spot right next to the 80C48
- Use the space toward the center of the board since the case sloops to the outside edge of the board.
- wire the following pins (I just soldered right to the top of
the ic's pins):
DS32KHZ 80C48
------- -----
#5 (vbatt) #40
#12 (out) #39
#4 (gnd) #20
#13 (vcc) #20 Yes, vcc to gnd
Increasing the clock accuracy of the CP290 by changing the crystal (never tried)
The CP-290 is clocked by a two-transistor crystal oscillator running at 32.768 kHz. The power consumption of this oscillator is very low, which allows the CP-290 to run on battery backup for long periods. This oscillator is used to run the clock whether or not it is connected to AC power, i.e., the power line frequency is not used.
To open the CP-290, unplug it and remove the battery. Next, remove the four screws holding the case together, one in the battery compartment and three under the rubber feet in the remaining corners. Remove the battery compartment and the screw near the power transformer that holds the main circuit board in place. When removing the main circuit board, watch out for the ribbon cable. It's pretty stiff, and the wires might break if you flex it too much.
To improve the accuracy of the clock, I removed the original 100 ppm watch crystal and replaced it with a 20 ppm part (Digi-Key sells several). Next,I removed 33 pF capacitor C17 and replaced it with a 15 pF fixed capacitor in parallel with a 5-15 pF variable capacitor. I used NPO ceramic capacitors for maximum temperature stability.
The oscillator circuit is located along the right side of the circuit board, about halfway between the ribbon cable and the battery compartment. The crystal is a small cylinder about 1 cm long and a few mm in diameter, and is covered by a blob of silicone to keep it in place. After installing the new parts, I removed the solder flux with alcohol so the stray capacitance of the flux wouldn't affect operation.
To trim the oscillator, I used a frequency counter with 0.1 ppm accuracy. The clock signal can be probed at the end of R35 nearest to the battery compartment, or at wire jumper L21 near tuned transformer TC3, or on pin 39 of the 80C48 microcontroller IC1.
People who don't have access to a good frequency counter/standard will have to trim the oscillator the old fashioned way: by trial and error. If this is the case, you might want to program the CP-290 to turn off a non- existent module once an hour. The LED on the front panel will blink each time the CP-290 transmits, which will allow you to determine whether the clock is drifting.
When adjusting the frequency, keep the CP-290 at the same temperature at which it will normally be operated, and allow the circuitry to warm up for at least 10 minutes first. It doesn't matter whether you run it on battery power or AC because the oscillator's supply voltage is regulated. I recommend using battery power for safety.
You may have to use different capacitor values depending on how far and in what direction the frequency is off. Lower values of C17 make the oscillator run faster, and higher values make it run slower. If you make C17 too small, the oscillator won't run at all. You can tell whether the oscillator is running because the front panel LED will blink every few seconds when the CP-290 is on battery power. If it doesn't blink, the oscillator isn't running. Also, if you try to change the frequency too much, the oscillator will become unstable and drift with temperature. This happened to me with the original crystal, which is why I replaced it with a better one.
Increasing
Range and Reliability
Credits: )) Sonic ((
http://siber-sonic.com/X10/X10world.html
PLC Frequency Adjustment
- Unplug CP290 and remove the 3 under-foot screws and the one in the battery compartment holding the two body halves together.
- Position the unit in a fashion which allows easy adjustment of the two transformers which will be adjusted.
- Connect frequency counter to powerline signal sensor (not CP290). Power up counter and signal sensor, allow counter to stabilize.
- Short transistor TR5 base to emitter to allow the 120kHz oscillator to free-run.
- Connect CP290 to A.C. line.
- Adjust transformer TC2 for 120kHz.
- Unplug the CP290, powerline signal sensor, and disconnect the frequency counter. Leave the free-run short circuit in place.
PLC Output Amplitude Adjustment
- Move the CP290 to the location where it will be used. Bring along oscilloscope and powerline signal sensor; connect these to a separate circuit, or at least an electrically distant outlet on the same circuit.
- Set up oscilloscope and powerline signal sensor to monitor the A.C. line.
- Connect CP290 directly to the A.C. line.
- Adjust transformer TC1 for maximum 120kHz signal amplitude. This is likely to be a broad, “low-Q” peak.
- Unplug/disconnect all. Remove the B-E short on TR5.
- Connect CP290 to computer and A.C. line. Run self-test. Reload programming. Test for normal operation.
- Unplug and reassemble CP290.