Althougth I think I understand the concept of governor mode--where rpm is kept constant independent of load (up to the battery limit), I sure don't understand how the ESC zero's in on the exact rpm you want.
Here's an example, you take off at some fixed power (throttle) setting of the ESC. As the plane accelerates, the rpm would normally rise--but if you are in governor mode it would seem to me that you would be stuck at the rpm the motor was turning when you released the plane. I don't think this is true, but then I don't know.
I have read the Castle-Creation's literature, but it too doesn't really explain. The only CC ESC I have is a small 10A (Phoenix) one on a Sig E-Force (RC), so I have never tried it. I guess I could try it and see how it works.
I am sure there are those here who understand what actually happens.

Hi Alan,
   I'm not sure how I missed this thread? I am sure Dean could explain this more in depth.

   The governor controls a predetermined throttle % not a RPM. The ESC uses a pulse width signal to control this throttle setting. A pulse width is a pulse width weather the motor is loaded or unloaded. The RPM is a result of this pulse width.

   I believe if you set the governor to say 90% throttle on the ground and the motor spins 10,000k @ 36amps when unloaded in the air the only change would be the current draw (say 30amps), not RPM. If you point the nose up, increasing the load the rpm should stay ~10,000, but he amp draw should go up. If you point the nose down the load decreases and the RPM should stay at ~10,000, but the amp draw will go down.

   I may be wrong, and this may be a poor explanation, but from my experience (limited) this is how I understand it to work. Hope this helps.

Alan,
   I understand what you are asking now  . I guess I didn't explain the pulse width changes to keep the RPM constant. Otherwise the RPM would change under load. 

   Unfortunately, I don't know how the ESC knows to change the pulse width. 

The actual "throttle" or voltage drive duty cycle (% of full) is varied by a control loop that compares the actual RPM to the RPM you asked for with the receiver pulse.

That is what Alan was looking for.
 
I am more familiar with an external governor that uses a hall effect sensor (or other type of external sensor) to receive feedback. Very simple (at least to me) compared to a CC ESC with internal circuitry that does the work.

Thanks for the info.   

As I mentioned on another thread, I just replaced the stock Brodak speed control/timer with a JMP-2 and Castle Creations Phoenix 35 ESC. I know understand why the JMP-2 is named so--you have to jump through a lot of hoops to program it.
Anyway my main idea is to run a higher pitch prop than I am running now. Right now a Graupner 9-4 works pretty good with the stock Brodak ESC/timer. My hope is a slower turning higher pitch prop will pull the plane at the same airspeed I was running on the Graupner, but with higher efficiency. With the Graupner, I was burning ~2500mAhrs per flight. That was a 6 minute time on the Brodak timer. 5 minutes would probably work too, if I would shorten the lines a little to speed up the pattern. The only difficulty here was that I am not sure if it is "legal" for the assistant to actually "start" the motor in a real PAMPA competition. The Brodak timer doesn't allow a gap of more than a few seconds after the button is pushed before the motor starts at full (and expensive high loaded static power).
I first tried to replace the Brodak timer with the JMP-2, but it didn't work. My guess is that the JMP-2 wasn't dancing the right steps to arm the Brodak ESC. So I went and bought the Phoenix 35--Tower Hobbies had them for $69 which looked like a good deal to me. I also bought the USB programming interface--really a necessity after my experience with the JMP-2. I put it in High Governor mode, because that's what everone else is doing with their AXI's
So I set up the JMP-2 to give me ~12krpm. I think that is a nominal rpm for my APC 9-4.5 Thin Electric prop to give me just less than 5.0s laps. We will see. I also have some APC 9-6 and 8-6 Thin Electric props along with 9-6 and 8.5-6 Master Airscrew E prop. We will see! To get the 12krpm, I set the JMP-2 to give ~1/4 throttle. This corresponds to ~16 ticks of the full scale 64 divisions which you can program the JMP2.
Also setup for a 5m30s full time flight. This is above the 30 second wait after the button is pushed, so the judges will see a total time just about 6 seconds. I did hear the 2 blips --of course I was holding the plane, so I think I will have plenty of warning.
So after all this is done, maybe I will have an idea what battery to buy next. The best (and safest) bet seems to be a 7.8oz 7/8" thick 3200maHr battery. One constraint is that the battery needs to slide in the wing, and a lot of batteries (like a nice 1 inch thick 20C 4 oz 2500 maHr) are too thick. The stock Brodak 16C 11oz 4000maHr battery is less than 7/8". However it would be good to lose a few oz. Currently I am flying at about 33oz total with the stock setup.
I am thinking about the next electric project. Once I understand a few more things, I will see where I stand.

Phil,
You are right. It is actually funny to watch the Phoenix ESC try to start the motor from zero. It doesn't know which way to go for very low amps. It is even more amusing to note that if you reach over and spin the outrunner can in the right direction, you can even get the thing running a little sooner than the automated sequence does. Almost (but safer) than flipping the prop!