dougclind wrote:1. You weren't able to show really close channels on the JR. You only cycled the JR 3 or 4 times. How close have you seen the channels get in your testing? I have to think that there's SOMETHING in the firmware which keeps the separation to some kind of minimum.
The example shown
in the article was typical of the "overlapping channels" scenario. because this only happens very occasionally I didn't spend too much time trying to observe any closer spacing than this (for fear of wearing out the transmitter's power switch
With channel spacing as close as shown in that article, a video-sender can almost completely envelop the two channels and thus cause a loss of the link. I tested with a 600mW video sender but another engineer in Australia was able to kill the link with just a 50mW transmitter in the same circumstances.
2. The DSM2 signal is a spread spectrum signal which has a certain immunity to interference just because it is SS. It kind of looks like the bandwidth of the JR signal is as much as 50 MHz. Have you tried clobbering both channels to see at what point the rx loses the signal?
DSSS Spread-spectrum signals do have a high immunity to noise generated by other spread-spectrum sources but video-senders are not SS sources. They don't transmit in short bursts with gaps, as other SS equipment does (thus allowing some degree of interleaving), they have a 100% duty-cycle which means their effect on a SS signal is more pronounced.
You'll also notice that although the DSM2 signal does spread quite a bit near the base of the spike, the bulk of the power is spread over a range of only about 1-1.5MHz which is not very much when trying to cope with a non-SS source of interference.
3. You used a 600 mW video transmitter which looked like it put out a 100MHz wide signal. An NTSC TV channel only has a bandwidth of 6 MHz. So was that any kind of realistic test? Are there things in the 2.4 band that generate that kind of broadband signal?
The video sender uses about 15MHz of the band, the entire scan of the analyzer is only 100MHz. The DSM2 channels were spaced at just under 10MHz apart.
4. Do you have any idea of what kind of power our rc transmitters put out? Maybe we need kilowatt amplifiers. (just kidding)
Most 2.4GHz RC transmitters have an effective radiated power (ERP) of about 100mW. This is usually obtained by using a 60mW transmitter and an antenna with a gain of almost 2. The DSM2 transmitters seem to have an ERP of little more than 100mW (about 125mW by my measurements).
The reality is that, under good conditions, the DSM2 system is as good as any other on the market -- it's just that when you use DSM2 you've got less headroom (insurance against unfavorable conditions).
Couple the low headroom from an RF perspective with the low headroom from an operating-voltage perspective and it's easy to see why DSM2 equipment is over-represented in the "unexpected loss of control" statistics.
When DSM2 users have a perfect power-setup (good batteries/regulators) and fly in relatively benign RF environments then they will have no problems at all. This explains why the vast majority of DSM2 users have flown for years without a single lockout or glitch.
However, the reality is that not everyone's batteries are 100% and not all regulators can deliver the required currents all the time. What's more, as we see a massive increase in the use of the 2.4GHz band for RC and other applications, the RF environment we find ourselves flying in is sometimes far from benign -- downright hostile in fact!
Users of other RC systems such as FASST and even many of the cheap Chinese options have fewer problems because they have more headroom. Those systems are better able to handle the effects of less than perfect batteries and a hostile RF environment.
When you buy DSM2 you buy a system that will work well under ideal conditions but you get far less insurance against the unexpected than is the case with almost all other brands.
DSM2 was developed in a different era -- a time when the 2.4GHz band was much quieter than it is now. To be totally honest, it's really past its "best-by" date. I had hoped that they'd use the launch of the DX8 to either switch to DSMJ or at least something like DSM3 with an extra RF channel or two that would help mitigate a noisy RF environment. Both of those changes could have been made 100% backwards compatible.
Why they didn't take the opportunity to enhance their firmware and provide customers with much greater headroom is a question only they can answer -- and they refuse to even acknowledge my emails on the subject