Can't launch vehicles lift more than one satellite into orbit at once? Being too small for Directv's satellites is a show stopper, and I hadn't realized that was the case with the Falcon 9. Being too big, on the other hand, seems like less of a problem, assuming the remaining capacity of the Falcon Heavy can be filled - which maybe I shouldn't assume.
According to Wikipedia, the Falcon Heavy (they renamed it from the Falcon 9 Heavy) is capable of lifting 21,000 kg to geostationary transfer orbit, at a cost per launch of $77 - $135 million, versus $56 million for Falcon 9 with a lift capacity of just under 5000 kg. If Directv was taking up a third of the payload of the Falcon Heavy, that would be as little as $25 - $45 million. Any idea how that compares to what they're paying now?
The fact SpaceX is booked up for a few years doesn't seem to be too much of a problem, considering that there haven't been any announcements for satellites beyond D15. Given that D14 was announced in June 2010 and will be launched almost four years later, if they announced D16 today they would be looking at a launch in 2016 or more likely 2017. According to the launch manifest on SpaceX's site, there is one Falcon Heavy launch currently scheduled for 2016, and one for 2017.
They may not choose to use SpaceX, especially if the next satellite is announced in the near future, but they should at least have the choice available to them given how long it takes from announcement to launch.
I am not 100% sure but believe that Arianespace has done that with the Ariane5. A payload would consist of 2 or even 3 satellites at one time
That has happened many times and not just with the Ariane. There was a LEO launch last month of a Minotaur-1 carrying 29 (small) satellites.
There are several issues related to multiple-vehicle deliveries to GTO. First, all of the multiple payloads have to be delivered to the launch site in fairly close time-proximity to one another; no one wants to have a multi-hundred million dollar payload just sitting around in a clean room (not to mention that most launch locations don't have an excess of clean room space for storage anyway).
Second, all three payloads have to be fueled and processed with operating fluids, batteries, etc. prior to integration with the launch vehicle. This takes time and is generally considered fairly hazardous if the satellite has hypergolic fluids onboard for attitude control (most do). So that has to be done in a hazmat facility; most launch operations don't have multiple hypergolic loading sites and crews to process multiple payloads in parallel. Doing them serially takes longer and puts more strain on ground crews to sustain that level of intensity in the hazmat ops. Then the fueled and ready spacecraft have to be physically and electrically mated to their payload adapter, and then the entire cluster of payloads encapsulated for launch inside the fairing. Look at the Arianespace site for a cutaway view of how Ariane 5 does it; it's not terribly simple. I don't think SpaceX has even shown concept art or advertised their intentions of developing this capability for F9H at all. Their target market for this vehicle is PROBABLY single large LEO payloads for the USAF and NSA, and large-payload GTO missions, using the excess delta-v for reusability operations if those prove to be viable and economically advantageous.
And third, cost. The prices quoted by SpaceX are taken with a large grain of salt by people who know the GTO launch business. They're akin to the barebones prices you see advertised on newspaper sales flyers; that price typically wouldn't even include a rocket payload adapter, or pay for the very expensive ground processing and integration ops. Now even so, there is no doubt SpaceX promises lower prices than competitors for the payloads it can fly. The real question will be if they can maintain regularly-scheduled missions, one after another, for several years on end, and still keep their prices as low as promised. IF they do that, the entire launch industry will be turned upside down. One can hope.
Anyway, back to the main topic - multiple payload operations for GTO missions has been a staple for Ariansepace but the operational and timing issues have proven to be big enough to lead to the development of Ariane 6, a smaller booster for single GTO missions, which is still a ways away. So really, for now anyway, SpaceX is off the radar for typical CONUS Directv payloads. F9H may be marketed for that purpose in the future, but that vehicle hasn't even flown yet. Time will tell.