Fourier Analysis of in-flight RPM---Brodak 40

[Alan Hahn]

Well here is another post in the continuing effort to "quantify" what our engines are doing.

I think the data is pretty good, however my interpretation of what a "Stunt Motor" is gets fuzzier with each new engine I test.

So this is a setup I must have been running some years back--an "original" TopFlite Nobler Arf with a Brodak 40. This was put together within the first month (or even faster) of when this first, of the modern era of Arf's, became available. I actually wrote a review of this plane and engine and posted it on SSW many years ago. Unfortunately the post was lost in one of the crashes SSW suffered back in that era. I don't even recall how long ago that was! Fortunately someone picked up the review and put it in Stunt News. I remember how surprised I was when I was reading SN and thinking how familiar the review was. When I saw my name at the end was when I realized it was my post!

That's the lead in for this data. After my difficulty with the side-mounted Fox35, I thought (haa!) that the B40 might provide a better example of a 4-2-4 engine, simply because it was mounted in a full fuse plane, and because it was always a good runner. And actually this was a pretty easy subject to resurrect from my "Boneyard"--up in the attic. Although the thought crossed my mind whether the hot-melt glue of this Chinese ARF was still up to the job after a number of summers in the attic (just as I   was getting ready to pull out of the wingover!!).

So the actual numbers---this is also an early Brodak 40, that I am running with a Tongue muffler to keep the weight down. All up weight is a little shy of 40oz (no fuel). I am running a un-obtanium Black Tornado Magnum 11-4 prop. Fuel tank is a 4 oz round clunk tank. Fuel is some %Nitro Sig 50-50 castor/synthetic fuel. I am not sure of the % because the label has fallen off the old fuel bottle. I am guessing it is 10% or so.

What follows is a selection of maneuvers. The technique to get the rpm is described here. http://stunthanger.com/smf/index.php?topic=14196.0

The B40 was pretty clean in terms of the sound--especially with the tongue muffler. Wow that sounded loud--I guess I am really use to electric power these days. Only issue was that a full scale plane flew overhead during the inside loops and interfered with the data. As a result, I reflew the inside loops at the end after the clover.

Also I note that the wind was almost non-existent.

[Alan Hahn]

Here is the plot of the release and level laps. The engine is turning the 11-4 prop at 10000 rpm, and increases to ~11600 rpm after takeoff. The marking of the level laps is just a convenience--when the plane is 180 degrees opposite the video camera (which is setup near where the judges would stand), I mark that time. I use that for lap timing. By the way, the lap times are 5.0s, and the same inverted.

I modified the look of the plots a bit. Just to make it easier to follow (I hope), I draw horizontal yellow lines which indicate the limits of where the motor is running in 4 stroke or 2 stroke mode--that is shown by the green trace. It actually seems to work reasonably well. If the engine was running a perfect 4 stroke, it would be on the yellow 4-stroke line. Likewise a perfect 2 stroke would be on the yellow 2 stroke line.

You can see the engine goes into a pretty good 2 stroke right after it breaks ground and accelerates. On the ground before release it is in (audibly at least) a pretty throaty 4 stroke. I am not sure why it goes into a 2 stroke as it climbs, because on my electric, the prop is unloading at that point--static on the ground is usually the biggest load. So I guess it may have something to do with the fuel mixture changing as the plane picks up speed--but before the engine can get up to final flying speed and cool off.

[Alan Hahn]

Here is the Wingover.

The engine goes into a strong 2 stroke as it nears the top of the wingover. Notice that the rpm does not really increase. This seems really to be different than the Fox 35, where I see the rpm jump when the engine goes into a 2 stroke. Now the power may be higher in a 2 stroke, but the Brodak 40 doesn't speed up. I am not sure if this is a really desirable trait or not.
The engine drops back into a 4 stroke on the downward led and inverted half lap.

What I mean by "entry", is that I try to time the point as near as possible to when I pull the corner. The "exits" are timed just after the plane enters level flight after the maneuver.

[Alan Hahn]

Here are the inside loops (as I mentioned,  I re-flew the loops after the clover).

Audibly, I can just hear the B40 breaking into what most of us would identify as a 2 stroke mode during the climb. But notice that the rpm still is dropping during the climb. That implies to me that the airspeed is dropping (prop load is increasing). Of course I am biased, because I have run an airspeed indicator on my Electrified Nobler, and that indeed is what is happening--and there the rpm  even stays constant, not dropping.

[Alan Hahn]

Triangles

I am not too sure what comment to make here--except I wish I could fly these better!

[Alan Hahn]

Vertical 8's

I put this in because there is a lot of effort keeping the plane climbing and turning during this maneuver.

[Alan Hahn]

Overhead 8's

At least on my electric setup, the overhead 8 is where my setup puts out the peak power. So it is useful to look at how a glow engine handles the effort.

[Alan Hahn]

I do have more, but I bet that is enough for now.

[Dean Pappas]

Verrry interestink!
Thanks Alan,
   Dean P.

[tom hampshire]

Alan - THANK YOU!  I'm the development/test/repair guy for Brodak, and I'd love it if we could see multiple data sets for a high power maneuver, say either the square or overhead eights.  I'm very curious to see how repeatable the breaks and overall power regulation are.  It's amazin to see what you guessed at by the seat of pants shim and tweak method laid out in real time data.  Thanks again.  Tom H.

[Alan Hahn]

Hi Tom,
I can certainly try. However one thing I didn't really mention but has become apparent to me during these tests---I am a pretty lousy pilot!

I sure wish I could consistently fly similar sized maneuvers (would help my scores too!). But I might be able to draft a better pilot than I to do a series of the same maneuvers, like the overhead 8's or square loops.

We are going through a patch of crummy weather in the next few days, so I am not sure when the next opportunity will arise.

[Dean Pappas]

Hi Tom,
And the more carefully you look at the RPM data, the more you'll be convinced that electric is the way to go.
What I have found interesting is that even the corners like the upper right in the inside squares and the upper left in the outsides and the triangle tops result in a brief RPM drop. I suspect that even as a dive is initiated, the cornering drag loads the system. That would suggest that a burst of power any time the handle is moving away from neutral would be useful.
Hmmm,
Dean Pappas

[tom hampshire]

Hi DeanO - I freely admit you're right, electric regulation is so much more accurate and controllable than IC.  But what's the thrill in flying a vacuum cleaner?  Yeah, keeps fields.  But remember the time a gas turbine ran at Indy?  What fun is whoosh when you can go boom?  Too bad, nostalgia ain't what it used to be.  Tom

[Steve Fitton]

... That would suggest that a burst of power any time the handle is moving away from neutral would be useful.
Hmmm,
Dean Pappas

I take it you are suggesting having an accelerometer on the ESC that would throw a few more electrons at the motor when it sees a maneuver of some sort?

[Alan Hahn]

I take it you are suggesting having an accelerometer on the ESC that would throw a few more electrons at the motor when it sees a maneuver of some sort?

Well let me point out that I am not planning it, but it would probably be "easy" to incorporate something like that into the Timer/Throttle setup--especially in you have access to the PIC programming. That way you can leave the ESC stuff alone.

Of course you could do the same on a glow setup if you had an rc throttle and a servo which would plug into a timer.

However this is starting to get a little too "techie" for my tastes (in CL), but that's just me. I like measuring to see what is going on. From what I am seeing here, electric is already superior to glow, simply in holding the rpm constant. In some way I think that perhaps some PA rules should be set in order to stop an "arms" race  .

Hope I am not disbarred from the electric forum! 

[phil c]

Might be "quicker" to incorporate a sensor on the bellcrank to detect when and how fast the bellcrank moves away from neutral.  The best servo I've seen has a 9 millisecond response time, you can probably figure about the same time to read a sensor and output a response to an motor controller.  It's probably useless to try it with an IC engine.  It takes much longer than the servo for the motor to start responding to a throttle change.  An electric motor might just respond before the plane got past the midpoint of a square corner.  Even then, there is a lag between when the motor applies more torque to the prop and the prop actually starts spinning faster and the air actually starts moving faster to apply sufficient force to accelerate the plane.  Just going from RC experience and watching carrier planes, the total loop with an IC engine takes about 2 seconds from the pilot moving the throttle until the plane starts to respond.

[Alan Hahn]

Might be "quicker" to incorporate a sensor on the bellcrank to detect when and how fast the bellcrank moves away from neutral.  The best servo I've seen has a 9 millisecond response time, you can probably figure about the same time to read a sensor and output a response to an motor controller.  It's probably useless to try it with an IC engine.  It takes much longer than the servo for the motor to start responding to a throttle change.  An electric motor might just respond before the plane got past the midpoint of a square corner.  Even then, there is a lag between when the motor applies more torque to the prop and the prop actually starts spinning faster and the air actually starts moving faster to apply sufficient force to accelerate the plane.  Just going from RC experience and watching carrier planes, the total loop with an IC engine takes about 2 seconds from the pilot moving the throttle until the plane starts to respond.

Electrics can be pretty darn fast since as soon as the current flows, the torque jumps up. Since the prop is so light, the acceleration is pretty fast. I have actually slowed down the throttle response of my electric setup--because I takeoff at 7600 rpm and transition to 8000 just before the wingover. I didn't want the rpm to jump too fast.