Hi Keith

In the case of CC ESC:s it easiest to buy the Castle Link cable and software and hook the ESC up to a computer. With the help of the link software you can set the ESC first in Control Line mode and then in Set-RPM mode. Once that is done you can write in the desired RPM(s). It is very easy to do and I highly recommend the CC Phoenix ESCs.

Andreas

Hi Andreas,
Thanks for the info on the CC ESC's. It's all very well setting rpm with a PC, but that means that you need to have a laptop/notebook etc. at the field if you want to make changes. This is a bit of a hassle, so do they supply any small programming devices as well?

Keith R

Keith,
CC will be coming out with a programming box that can be used at the field.  I wouldn't hold my breath as to when.  CC said it is in the "works".
As far as a laptop at the field, yes it is a bit of a hassle, but when you get your set up "dialed in" I almost never use it.  There are other ways to accomodate different rpm settings of the CC esc without using a laptop.

Besides, it seems that you're having a bit of a hassle now getting the governor working on your set up.   

Hi all!

Yes, CC will most probably release a programming box, they says its under development right now. Mr. Ford at CC wanted one himself and that is a good sign.  Too bad he doesnt already have one thou...

As Alan (the E-CL oracle  ) said, use all three RPMs and switch between them with the timer.

/Andreas

Hi Keith
I have 2 Old Time Stunt machines set up as you describe, with simple ESC's (no governor) using Will Hubin's timers with throttle setting using a pot (like a needle valve) and with throttle-up capability.  He sells two versions of that timer: one that adjusts RPM by some fixed schedule, and another that lets you determine how much gain you get over the course of the flight.  The  fixed mode timer holds lap speed to within a about .1 to .15 seconds over course of the OTS flight (shorter than F2B) so that is pretty good.

Electric motors all govern themselves to some extent automatically; they always try to deliver the RPM that is commanded.  I would think that even the simple ESC approach will work reasonably well as long as conditions are pretty good - maybe even as well as a lot of IC.  I would catagorize it as still much easier to set-up and use than a IC engine.

...but my F2B birds will continue to use Phoenix ESC's!

Thanks for your comments so far, but nobody is telling how they actually set their chosen rpm. My question is only reffering to those esc's with governor mode. Do you use a tach, and set the timer's pulse width up and down until the rpm is what you need? Do you do this before each flight? I know that Pat McKenzie has a timer with a digital pot that is almost like a needle valve. The point that I'm trying to make is that most governor modes can still vary if the throttle stick, or in our case, the pulse width changes. Even if this pulse width is fixed and the motor fires up..........sorry........runs up with a different load or a battery that is not exactly the same as when you made the setting, then the governor will lock in at a different rpm.

This all "sort-of" works, but the motor still runs like a really bad glow engine, and the rpm varies about 1000 rpm as the load changes.
Sure the obvious question is why don't I just buy another Jeti, CC or Shultze ESC?  Flying stunt without a good governor really sucks, so I was just curious to hear from others as to how they get around this hassle. Thanks.

Keith R

Hi Keith,

I hate to see a friend struggling    I set my rpm by using a tach while turning the "Electronic Needle Valve" on our timer/processor. This sets the PW and it is sent to the ESC. The Schulze ESC will produce the requested RPM and will not vary by more than 20 rpm. There is no variance in PW or in RPM selected EVER! The voltage of the battery or the load on the motor is not relevant unless there is not enough volts or amps to satisfy the current demanded by the ESC. (This was not the case prior to the development of the Schulze 18.46k F2B control line specific ESC. At the 2006 W/C my rpm would vary by as much as +- 250-300 rpm) Any variance you see will be a result of a non control line specific ESC using programming designed for R/C where there is noise, potential signal interruptions, framing issues etc. and thus the filtering and other processing sets a different rpm.

Let me save you some time and grief. You will not build your own ESC with world class governor mode in your lifetime. Governor modes are as different as night and day. Chopsticks is easy, Beethoven takes years of work. Good governor mode needs Beethoven skill. Pat and I had a discussion about building our own and it lasted about 30 seconds. (And we have at least Mozart level capability)

This all "sort-of" works, but the motor still runs like a really bad glow engine, and the rpm varies about 1000 rpm as the load changes.
Sure the obvious question is why don't I just buy another Jeti, CC or Shultze ESC? Flying stunt without a good governor really sucks,

Duh!,

I feel your pain!!      


Let's keep this simple. IMHO a Schulze 18.46k F2B ESC with ANY timer is better than any other setup. Others would agree.


Kim.   

Hi Keith,

I hate to see a friend struggling    I set my rpm by using a tach while turning the "Electronic Needle Valve" on our timer/processor. This sets the PW and it is sent to the ESC. The Schulze ESC will produce the requested RPM and will not vary by more than 20 rpm. There is no variance in PW or in RPM selected EVER! The voltage of the battery or the load on the motor is not relevant unless there is not enough volts or amps to satisfy the current demanded by the ESC. (This was not the case prior to the development of the Schulze 18.46k F2B control line specific ESC. At the 2006 W/C my rpm would vary by as much as +- 250-300 rpm) Any variance you see will be a result of a non control line specific ESC using programming designed for R/C where there is noise, potential signal interruptions, framing issues etc. and thus the filtering and other processing sets a different rpm.

Let me save you some time and grief. You will not build your own ESC with world class governor mode in your lifetime. Governor modes are as different as night and day. Chopsticks is easy, Beethoven takes years of work. Good governor mode needs Beethoven skill. Pat and I had a discussion about building our own and it lasted about 30 seconds. (And we have at least Mozart level capability)

Duh!,

I feel your pain!!      


Let's keep this simple. IMHO a Schulze 18.46k F2B ESC with ANY timer is better than any other setup. Others would agree.


Kim.   

Kim please check you PM.

Keith,

The current firmware and software for CC Phoenix ESCs allow you to set the brake to 100% (hard) when you are in control line mode with the governor function working!

Sometime back I decided to bite the bullet and buy a CC Phoenix 45 ESC and I am sold on them. I picked up a couple of additional CC 45s on Ebay for between $55 and $65 and they also work perfectly.

I second what Will says. Having that functionality was very useful when I was setting up the Vector and the CC ESC. I knew roughly where I wanted to be, so I set up 3 rpm's , and then a 2 minute flight time. It made it easy to zero in on where I wanted to be. I like the set rpm mode of the CC because there is no ambiguity of trying to figure out pulse widths (although that is also a way to go).

If I wasn't so into having a takeoff rpm and then a cruise for my "normal" flying, then the functionality in his timer and the CC Set RPM mode easily lets you program a fair weather rpm, a modestly windy rpm, and finally NATS weather rpm!

O.K. so what you guys are saying, is that the CC is doing the work here and not the timer, which can only set a pulse width, or series of pulse widths. If the CC has 3 governor mode settings that lock onto those 3 rpm settings, the actual revs are then monitored by the internal hardware/software of the CC esc. So the timer only needs to set a start pulse width greater than 1.0 ms, and then go back to 1.0 ms after the preset time. Is this correct?

If you use Will's timers that offer various settings, then are you referring to the non-governor modes? The pulse width is then not an solid reference to rpm, and it will then vary with load and battery voltage. Unless the CC's governor mode follows the pulse width, but that seems counter productive.

Just by the way, I flew my system with feedback from the rpm sensor on a standard cheap Hobbywing esc with no governor mode. I sorted out my program so that it has no more hunting. It was rock solid and felt to me like my old Jeti governor. This is the best result that I have had so far. I stored the shut-of rpm in the PIC chip and it was within a few rpm of the take-off setting. I hammered the corners hard and the speed stayed solid up hill and downhill. I am encouraged to keep on with this thing. The neat thing is that I was using bad components. The MVVS 4,6/1120 in-out runner has a high internal resistance and so does the Evolite 2500 mAh battery pack. This would have showed up the system badly but it felt solid all the way, and I can enable the brake. I am going to call the timer/controller "Beethoven!"..........sorry Kim

Keith R

Thank again Alan, now I see how the CC works with Will's timer. My Align governor will not always lock onto the same rpm with the same pulse width. I have to do this with a tach before each flight. The Scorpion is much the same. I can certainly agree with you and Kim about a good governor being essential. This is why I miss the Jeti. I also believe that you get what you pay for..........but only to a certain degree. At some stage it can be overdone. The winners in serious competition have demonstrated clearly that you can win with more than one system. Richie Kornmeier just won the Euro champs with a golden oldie...ST 60, standard silencer, and wooden hand-made propeller. Then there were Igor and Jiri Vejmola with electric, and followed by everything from the tuned pipe brigade to the 4-bangers and the rest, and Rob Metkemeijer's new .77 glow with a long stroke and 6 ports. Igor used one pack of A123 that is built in to his model, others use Lipo's. The bottom line is that this amazing sport of ours shows that there is indeed more than one way to skin a cat.

I find that a constant speed through any attitude and maneuver is the best compromise to give the pilot a good confident performance. The trainer that I'm using for my experiments is a simple unflapped, very basic trainer that I called the EZE 40(Easy). I had a slightly different version for beginners to learn on years ago. It was powered by the old Tower Hobbies 40 and used a stock APC 11 x 5 prop. It flew well enough to learn on, but when I dieselled the Tower it became a contest winner for the kids. The real difference was that they all had total confidence with it in the wind. In nice conditions almost anything works. They knew that the diesel would thump out a solid rpm for the whole flight and would never flame out or go sick and slow down. They didn't understand why, but they knew that all they had to do was fly the plane, and it would go where they steered it.

Now my new Eze 40 had an Enya SS 40 in it, and it runs quite well. I modified the nose to play with electric motors, and I have tested plenty in it without concern about scratching or wrecking the thing. Without the governor it slows down in the climbs, speeds up on the downhills, and of course, slows down overall as the battery drops. I made my timer increase pulse width with time, and also monitored the battery voltage, but there is obviously no reference to the rpm, so the flight performance is not very good. When I added the rpm sensor this same basic model now flies really well, like it did with the Jeti. In fact the difference is like night and day.

I know that this kind of thing is not that interesting to the majority who just want to buy a plug & play system, but it is like Alan winding his own motors. What for, when you can just go and buy the "right" thing?? It is because we just like to experiment and learn a little more. Here in Darkest Africa (getting darker by the day) we pay too much for any hobby goods, so I always try to make the best of everything and also try to find ways of introducing newcomers into the hobby with the best possible price/performance ratio. To be real honest though, I just like tinkering around with gadgets

I'm sure that I can get around having the hall-effect sensor by using a wire from the motor wires like the Eagle Tree rpm sensor. I have tried this, but could not get the right filtering to get just the motor's back-emf pulses. I was more interested in trying to write the software to keep the rpm constant, and the hall-effect sensor gives a good clean pulse. I will most likely have another go at the motor wires. If anyone has a circuit that works like the Eagle Tree' brushless rpm sensor, I would love a copy. Thanks again for all the feedback.

Keith R

The first thing I considered was an optical pickup, because I've made made tachs through the years this way. The Hall-effect sensor works well in fact, but I'm sure that it can be done the way Eagle Tree does it with the wired rpm sensor. I checked it on my scope this morning, and it seems to have a clean square  wave that cycles quite a bit. Maybe its the recorder reading it a certain amount of times per second. If you look at one of the motor wires while the motor is running, there are a mess of waveforms, so it's not too easy to filter out just the pulses from the back-emf. For now, the hall-effect sensor works. It's a tiny plastic case like the small transistors. This one from Honeywell is a latching type so it halfs the pulses, but it has a nice clean square wave at the output.

I also have the optical sensor from Eagle Tree, and this works by shining a small infra-red light onto a reflective surface with black lines etc. It's made for the rotor shaft on heli's and I bought it to check rpm on my I/C engine models. I will post some more later when I've managed to record the rpm in flight.

Keith R

The optical sensor also puts out pulses, and it's connected to the same input on the recorder. The standard optical tachs use the flicker from light passing through the prop, but the Eagle Tree has a an infra-red light source plus a pickup sensor. The esc's do know the rpm, and they pick this up from one, or more of the coils when it is switched off. The designer's just don't think that most users need this sort of info, so I guess they don't output it anywhere.

My "governor" simply works by adding pulse width if the rpm is low, and subtracting it when it is high. It's fun to run the motor without the prop, and then hold the shaft......with a cloth......and then let it go again. My program will keep adding pulse width further than it needs because the motor has not followed it quickly enough. When I let the shaft go again, it pushes the rpm up much higher until it gets the pulse width down again. I'm going to play around with seeing if I can make it kick in when it is loaded more for a preset time and back-off a bit further when it it unloads. This is more or less how Igor's foamie worked before he used the gyro. He used the load from checking the current to do this. Seeing that I'm not clever enough to make my own esc, this is a fun way for me to fiddle with this.

Keith R

Hi Igor.........that's my problem......filtering out the mess. I had a look at the Eagle tree pc board, and they use quite a lot of tiny transistors, and also a small i/c (maybe) of some type. It's a compromise (like everything in stunt) between adding one simple small sensor and telling the end user to solder and extra wire or two on to his esc wiring. Most non-technical people can use a screw driver better than a soldering iron! My sensor at the moment is just held on to the casing of the MVVS in/out runner with a cable-tie (tie-wrap) because it does not have an exposed rotor. I have already figured out a simple mounting for the normal out-runner in my bigger model, so using the h-e sensor is not really a problem.

I wasn't going to post any photo's of this wreck, but anyway here goes. I took one of my basic "combat wing" trainers that had an Enya 40SS on it, and modified it with a simple bracket. This model has had quite a few prangs etc. so I did not mind abusing it to play with motors. It's had a boxful of motors on it already, including a Scorpion 3026-12, so it's been very useful so far. I put the sensor into a plastic housing with some epoxy, so it is a little bigger than necessary.

My comment is that you have two issues in your "quest" .

1) Make a governor
2) Get an rpm feedback for #1

I would solve (2) first by using the optical sensor. Then you can concentrate on the algorithm for the governor.

Once you have the governor working to your satisfaction, then tackle the electrical sensor pickup. As an aside, you might want to make a low pass filter to remove the known PWM frequency, which for most choices is higher than the commutation frequency--but not by much. For example, the frequency of the magnets passing by is 14 x the rotational frequency, or for 9000 rpm, => 2.1 kHz. I think the lowest brushless PWM frequencies are in the 8 kHz range (at least for Castle Creation ESC's)

I forgot you had the hall effect sensor. Sure use it! I was referring to the brushless sensor ET sells too.