My apologies for reviving an old thread but I just came across it.
I suspect my web page article on the CM15A was where the suggestion of using the RM1SG came from.
The reason for the significant increase in range is probably due to two factors. The RM1SG has 103dB receive sensitivity (about the same as the HiMARK RX3310A) and is very wideband compared to the HiMARK which is a superheterodyne receiver (and even wider than most other superregenerative receivers) so it
hears poorly tuned X-10 transmitters that are not transmitting right on 310MHz better than the HiMARK. I haven't measured the bandwidth but it was able to receive from a Chamberlain 312MHz garage door indicator. (The supplier gave me the bandwidth but I cannot recall it offhand.)
I also tried to persuade the supplier to route pin 3 of the op-amp to one of the two RX pins so it could be used for an RSSI indicator but had no luck - I suspect the gear is manufactured by a different company than the one selling it.
The well documented mod is one more way to get acceptable RF range with a CM15A. Of course, it voids the warranty but it looks like X-10 may not be honoring warranties, anyway. A contact in China is trying to find the manufacturer.
It performs even better with an eggbeater antenna BUT it seemed to be overwhelmed when a preamp was added. I did not do extensive testing and the preamp was adjacent to the receiver which may have been the cause. Or, it may just be that the wider bandwidth was being buried in noise due to the reduced selectivity. If I get some time, I'll try to test it anew with the preamp located remotely. I should have prototype
RR5x5TM PCBs in about two weeks and I'll breadboard one with an RM1SG.