Hello everyone,

A few years ago there was some discussion here about optimization of the Spirit system for F3C. At that time I can't say the results of the discussion led to much benefit with one user having to abandon the Spirit due to adverse behavior in aerobatics. I have also read about some issues duplicating a solid hover with the available settings. I am coming of extensive experience with the CGY760 which is current holder of the F3C world title and has held that title over many years besides regional and national championships. Thus we all know it works well, but I would like to try to get that same level of the performance out of my new Spirit GT. since flybarless models were allowed into competition.

Let's look at hover first. When the 760 is used, we classically run a very high style setting and very high cyclic gain, the latter as much as the model will tolerate without wobbles or oscillation. While aerobatic gain settings are typically in the usual 70-80% range, hover gains may be as high as 170% using the scaling functionality the gyro offers similar to what Spirit does for its rudder gain. This access to high gain in hover makes the gyro tremendously stable in hover. Regarding the high style value used, the bench behavior of this setting shows the swash decay to be very slow; higher style in general yields the more robotic feel. I have cyclic gain maxed out in low rpm hover on the Spirit @ 100%, but the model will clearly tolerate more. I need a gain scaler such as is present for the rudder (Futaba calls this Base Gain) to achieve this. Is there another way to achieve more gain in hover? The model still feels "loose" at 100% gain at a low rpm contest hover. The model is hovering ok, but it still feels loose and in gusty conditions is not as "locked in" as I am used to seeing before.

The explanations of the Style and Cyclic Feed forward are detailed, but the granularity of the range of adjustment makes it a bit vague to assess the effect of changes, especially in hover and when rates are already dialed back in the transmitter to soften response. The developer has suggested running a style setting maxed out at 8 because that is the smoothest, but it is unclear to me if doing that actually helps the gyro work in precision hovering. We know for a fact that running the Style setting in a 760 very high and for a very robotic feel dramatic improves hovering performance, but based on suggestion running the Spirit with a robotic setting is not ideal for precision hovering. The Cyclic Feed Forward is perhaps the most confounding to know how to tune based on explanation. Lower CFF values are supposed to yield a smoother and more delayed feel to stick input, while higher ones create and more aggressive feel. High Style setting is suggested for hover, yet there is also specific guidance that when style is there the stops become more aggressive. So now we have Style suggested to be high then l feel like I need even more elevator filter to fully suppress bounce, like an additional point; 1-4 is far too course for precise adjustment). Having suggestions like "if you like VBar or Bavarian Demon use this setting or that" are comparatively useless for guidance because those systems can themselves be tuned from one end of the spectrum to the other in terms of feel. Furthermore, stating the the pilot effectively has to juggle Style and CFF together to get what they like isn't as easy as talking because each can be tuned to almost contradict the other. I've tuned a many different flybarless systems over the past decade to good effect besides extensively flying the 760, and this is the first time that I'm rather lost about how to manipulate these two tuning values to optimize performance. Everything is like a shade of gray in effect across the entire tuning range. If I could get a more concrete cause and effect test, it would be much easier to know which way to go to optimize performance. Right now the system does not YET approach the 760 it replaced in hovering performance, but I know it could be excellent if the developer could give me further guidance on how to tune it particularly for high hovering stability. It really would like more cyclic gain in hover! I'm sure it will be asked, but the model in question is spot on set to 6 degrees with the adjust slider almost right in the center of the adjustment range. Swash is perfectly level with servo arms squared and perfectly 90 degrees. I have recheck basic set up setting no less that three times in as many days.

This was long winded I know, but I really want to make this thing fly as well as what it replaced. I just need to know how to go about it. Please offer me some guidance.

Kind regards,

Ben Minor

Update:

After flying today it does not appear that running style high has a benefit of improving stability and tends to make ele bounce behavior harder to suppress. It's more predictable to soften response around center stick with decreased cyclic rate and reduced transmitter rates. The feel of the model is a bit mor like I am accustomed to, but it's still begging for more P and I gain to really park in hover at low rpm with big, heavy blades. It still feels "loose" in the wind, but I appear to be out of things to adjust.

Here's a question: If cyclic gain is put on a channel for adjustment in the tx, the "channels" tab for gain my show the gain to reach 100% before maximum range of travel of the controlling channel is reached. This being the case, when the number shows 100% on the screen, does the gyro stop increasing internally as well or does the gain continue to increase after the screen say 100% and as long as the transmitter channel has "travel" remaining?

Hello,

Cyclic Gain will stop at 100% - at maximum value for this parameter when you will set P: Cyclic Gain in the Channels window.

Thank you for all the details. I would like to recommend to be not confused by similar parameter names between FBL systems.
For example Style parameter could work very differently on each controller. So it is better to try all range of the values in any case.

For example I prefer the highest Flight Style and recommend it for F3C together with higher Cyclic Feed Forward (6-10) but lower Rotation Rate (9-11).

_________________
Spirit System developer