Settings for Mini Titan V2?

[gwright]

velcro isn't appropriate for mounting a fbl gyro. I've tried 2 different gel type mounting material (zeal was the worst of anything I tried), two white foams of different thicknesses, both 1/16" and 1/8" black foam tape, the clear 3M stuff, and the black stuff that comes with the gyro. The black stuff that comes with the gyro,..one layer, is worlds better than anything else. I know you want something to absorb vibes, but this is the best stuff for gyro operation. The gels were the worst (loose and floppy, probably like your velcro, although I think velcro would be worse). I think they give you 2 or 3 pieces of the black stuff with the gyro, but you can get it on a roll where the 3m tapes are at numerous places. I get it at meijers here but I've got it before at wal mart. There are a couple of the 3m tapes, a clear thick stuff and the thin black stuff. I was told once that you really want these sensors as close to being "welded to the frame" as possible. In other words, very securely mounted. Vibration dampening in the unit should filter out anything over a certain frequency which will be well above anything that would create erroneous inputs.

[seattleite09]

Thank you guys. I removed one layer of mounting tape, and mounted the gyro with just the tape that came with it (black with red backing). Big difference in flight characteristics!!!!

I flew three packs today, and tweaked the GT5 repeatedly. Here's what works best for me (so far):

Total gain: 100%
Swash P=80
Swash I=40
Swash D=0
Tail P=60
Tail I=50
Tail D=10

I also found myself having to use 50% D/R in the TX (which, I think, suggests I need to reduce the servo travel by about 30-50%? Mechanically or from the gyro?).

I also needed to change the Normal pitch curve in the TX (from its default 0-100%), to 0-80%, which suggests I need to reduce collective pitch by 20% somehow (mechanically or from the gyro)?... Basically decreasing the pitch/throttle ratio made a big difference in flight pattern, sound, and "tail kick" when ascending rapidly.

Another interesting bit: there is an ever so slight pull to the left from a hover position, which I suspect indicates that my swash is not 100% level, and it may require 1 or even 1/2 of a turn of one of the aileron servo links. Could anything else cause this?

The heli is tail-heavy in spite of the very heavy nose-mounted battery (TP 2700mah 3s). So I guess I will need to eliminate that issue too. All in due time.

So far I am happy. It's beginning to fly GREAT!

Unrelated -- Gary I saw some of your flights on YouTube. Awesome.

[seattleite09]

Oh, one other thing, though probably not the right forum: I previously had FBL blades on (Revolution). Today I put FB blades on (Thunder Tiger), and all I can say is -- wow! Why are people paying more for FBL blades? These FB blades are great and fly really well with the FBL Mini Titan V2.

[gwright]

I would suggest dropping the cyclic range on the servos (go in on servo arms) but that's going to get the pushrod geometries a bit off, so go into the gyro and simply reduce the ail and ele agility. It's that area where you have pitch, ail, ele but only the pitch numbers actually reduce the movement you see on the workbench. The ail/ele reduce the rates and feel dynamically but you can't actually measure a difference in pitch change statically. I know some people just hovering that have put those near zero and it gets sort of a really soft feel like a flybar with minimal movement.I'm not suggesting go to zero, just making an observation to let you know you can make DRASTIC changes in those numbers to get the feel you want.

Additionally, you can reduce the P gain on the swash, and maybe even increase the I gain,.. which may help with the slight lean under vertical acceleration.

You need enough gain to hold the machine and respond appropriately to inputs. From there, you can go a LOT higher, but if you do, you end up on the verge of little wobbles on occassion (to high gain) and the responsiveness to stick inputs gets pretty darn quick. Higher gain changes the swash inputs to undesired movement, but also to desired movement (gets quicker with higher gain, slower with lower gain). You may find that it's actually better at much lower P gain, like half of what you have now. Yes, I make BIG adjustments. I'm impatient and want things done quickly, so my rule of thumb is the same as they use in other industries to adjust PID control loops. I usually adjust halfway from the setpoint I'm at, to the max, in the direction I want to go in. If too much, I go halfway between there and the last setpoint,..etc. Gets me to my destination much quicker than little 2 or 3 point adjustments, that usually don't make enough difference to actually see.

[seattleite09]

Thanks Gary! I will keep on tweaking....

Regarding the P settings -- I started with lower total gain and much lower P (around 40), and got to T=100,P=80 by tweaking. What you described is known as "binary search" in my profession, and I practice it and apply it all the time. It gets you to the result in O(log(n)) time instead of O(n), which is generally much faster for larger 'n'.

[gwright]

Sorry, didn't mean to appear to be talking down,..but I don't generally know the background of people on here. I ran a pretty large IT staff and infrastructure for the last 20 years. Background was primarily network infrastructure but did a good bit of coding also many moons ago,..like before dirt was invented.

[seattleite09]

Oh, not a problem at all -- I was just enforcing the validity of your earlier point about binary-searching instead of linear-searching. An acknowledgement that it was a great point!

[Bladecpnitro]

I thought swash D Sense stayed at 100% for large helis and only sometimes lowered to as much as 80% for small helis and I see youve set yours to zero.

I set my tail D Sense to zeroing get rid of tail wag this morning and it worked.

[seattleite09]

Are you sure you are not actually referring to the I-D stick setting? That I have set to 100.

D-sense refers to the differential factor in PID. It specifies how much the gyro will "anticipate" motion based on past readings. I don't think it's supposed to be as big as 100%, neither for swash nor tail. A higher swash D factor will make the heli "floaty". My understanding is that P (the proportional factor) needs to account for most of the gyro compensation for best results. I started with D=25 earlier, and that felt weird. So I lowered gradually to 0, and that feels right for me.
I am not suggesting these are perfect numbers -- it's work in progress. Still tweaking...

[Bladecpnitro]

Originally Posted by seattleite09

Are you sure you are not actually referring to the I-D stick setting? That I have set to 100.

D-sense refers to the differential factor in PID. It specifies how much the gyro will "anticipate" motion based on past readings. I don't think it's supposed to be as big as 100%, neither for swash nor tail. A higher swash D factor will make the heli "floaty". My understanding is that P (the proportional factor) needs to account for most of the gyro compensation for best results. I started with D=25 earlier, and that felt weird. So I lowered gradually to 0, and that feels right for me.
I am not suggesting these are perfect numbers -- it's work in progress. Still tweaking...

I looked again and your right the manual does say that about Swash I-D stick settings for smaller heli's. Says it might be needed to lower ID Stick to 80 for smaller helis to get rid of head wobble.


My 450 was working great until I put a new Tarot Tail box on. I installed new bearings and shaft in the tail box as well as new grip bearings and now for the first time I get tail wag in a hover with the GT 5.1 at 3200 head speed. Head wobble has re surfaced as well, Its absurd! If anything I was expecting it to work better than ever with the new parts.

I've seen the tail wag in flips and high speed inverted backwards circuits with the Tail P gain set too high but never in a hover even with the gain maxed out. The only way I found to kill the wag was to zero out the D sense on the tail as suggested in this Forum in an older post.


I reduced I D stick to 80 yesterday and it did help a good bit to reduce head wobble. In fact there were only 2 packs out of my 5 that head wobble was still present. I guess because of higher head speed with my 2 newest packs.

Swash D Sense was set to 30, I'll try reducing that this morning to see if I can get the rest of that head wobble gone.


Higher head speed calls for lower gains. When you get your bird dialed in like you like it. If you have an idle up 2 for instance that is very high RMP like 3500, you can toggle to like an 80% total gain if needed to settle the heli down. High RPM reacts much quicker than low RPM so tail wagging and head wobble will be induced with higher head speeds. With a FB heli all you had to do was lower the over all tail gyro gain in the TX. Its a tad more complicated with FBL since you also have head wobble to deal with.


Also I don't believe the GT 5.1 has a rate mode like we used to have with the single axis tail gyros where the rudder would automatically return to center when no stick movement is seen. It may fly like its in rate mode though at lower gains, I don't know.

Rate mode was handy on the older gyros just for centering up the tail. And many gyros required 7 degrees of right rudder manually trimmed into the control rod in rate mode, Futaba probably still does. And thats how I set my GT 5.1 up as well, The tail centers one time at boot up and I can lay my pitch gauge under the tail and see when I have 7 degrees of right rudder mechanically adjusted into the blades at center stick on rudder. I'll adjust a few rounds and reboot it to check it until I get it dead on.

[Bladecpnitro]

Good luck with it!

[seattleite09]

Originally Posted by Bladecpnitro

Good luck with it!

Sorry you are having troubles! It sounds like your unit may be on its way to heaven?

[seattleite09]

Originally Posted by gwright

I would suggest dropping the cyclic range on the servos (go in on servo arms) but that's going to get the pushrod geometries a bit off, so go into the gyro and simply reduce the ail and ele agility. It's that area where you have pitch, ail, ele but only the pitch numbers actually reduce the movement you see on the workbench. The ail/ele reduce the rates and feel dynamically but you can't actually measure a difference in pitch change statically. I know some people just hovering that have put those near zero and it gets sort of a really soft feel like a flybar with minimal movement.I'm not suggesting go to zero, just making an observation to let you know you can make DRASTIC changes in those numbers to get the feel you want.

Additionally, you can reduce the P gain on the swash, and maybe even increase the I gain,.. which may help with the slight lean under vertical acceleration.

You need enough gain to hold the machine and respond appropriately to inputs. From there, you can go a LOT higher, but if you do, you end up on the verge of little wobbles on occassion (to high gain) and the responsiveness to stick inputs gets pretty darn quick. Higher gain changes the swash inputs to undesired movement, but also to desired movement (gets quicker with higher gain, slower with lower gain). You may find that it's actually better at much lower P gain, like half of what you have now. Yes, I make BIG adjustments. I'm impatient and want things done quickly, so my rule of thumb is the same as they use in other industries to adjust PID control loops. I usually adjust halfway from the setpoint I'm at, to the max, in the direction I want to go in. If too much, I go halfway between there and the last setpoint,..etc. Gets me to my destination much quicker than little 2 or 3 point adjustments, that usually don't make enough difference to actually see.

Ok, so I tweaked some more...

I lowered the cyclic servo travel numbers as well as the collective pitch number in GT5.1. I managed to bring the total collective pitch to 10 degrees in both directions. This basically requires more throttle/headspeed to achieve the same lift, but the heli feels a lot more stable. One consequence of this was a very slight tail wag, which I successfully ironed out by lowering the tail P gain in the gyro. It's now set at P=50, I believe.

Another tweak: the heli was tail heavy. I had to add a whopping 35g of tungsten into the very tip of the canopy to balance it. I am concerned about the possible impact of this additional weight on 3D performance (at some point in the future when I am ready). Any ideas here? A lighter heli is supposed to always fly better than a heavier heli...

The end goal of balancing the machine was to be able to apply full rudder and have the heli spin in place without destabilizing. I am getting closer to that, but still not there. I can spin 360 degrees and recover. Based on how it feels now, I don't think I can spin 2 x 360 degrees, unless I have twice the altitude to recover! Basically, when it comes to spinning in place, this is nowhere near the level at which my trex 250 with 3GX MRS is. It must be something in my GT5.1 setup?! Any ideas?

Thanks again guys.

[Bladecpnitro]

Originally Posted by seattleite09

Sorry you are having troubles! It sounds like your unit may be on its way to heaven?

Or possibly to hell, LOL! Brad Paisley uses that line in one of his new Album songs, its hilarious.

I've gotten a lot of flying out of that gyro. I flew it again this evening in fact and it still does well long as I stay off that 3500 RPM. It just might be fixing to kick the bucket, I'm hoping not though. I have a van in the shop for repair and don't need the added expense right now of a gyro.

I did bring it back in and in one last ditch effort to get it working better I moved all of the cyclic balls in one hole to see if lowering the mechanical gain will help. I was set up with 13 -/+ Pitch and 10 degrees cyclic and now I'm set up with 14 -/+ Pitch (GT Swash menu Pitch at 127 max) and 8 degrees of cyclic Pitch with the GT endpoints set back to 250 stock settings and 100% on TX endpoints. Gary says its best if you don't have to fool with those GT end points in the servo menu so I'll give that a shot, maybe it will help. I moved them from 200 back to 250. Well see how it does tomorrow hopefully. If I had some flood lights I'd try it now.

I was running my swash P gain at 50 to settle the head wobble down at 3200 RPM. I'm hoping I can run higher gain now in fact I'm anticipating it and have already bumped the P swash gain up to 70 to start with.

[Bladecpnitro]

Originally Posted by seattleite09

Ok, so I tweaked some more...

I lowered the cyclic servo travel numbers as well as the collective pitch number in GT5.1. I managed to bring the total collective pitch to 10 degrees in both directions. This basically requires more throttle/headspeed to achieve the same lift, but the heli feels a lot more stable. One consequence of this was a very slight tail wag, which I successfully ironed out by lowering the tail P gain in the gyro. It's now set at P=50, I believe.

Another tweak: the heli was tail heavy. I had to add a whopping 35g of tungsten into the very tip of the canopy to balance it. I am concerned about the possible impact of this additional weight on 3D performance (at some point in the future when I am ready). Any ideas here? A lighter heli is supposed to always fly better than a heavier heli...

The end goal of balancing the machine was to be able to apply full rudder and have the heli spin in place without destabilizing. I am getting closer to that, but still not there. I can spin 360 degrees and recover. Based on how it feels now, I don't think I can spin 2 x 360 degrees, unless I have twice the altitude to recover! Basically, when it comes to spinning in place, this is nowhere near the level at which my trex 250 with 3GX MRS is. It must be something in my GT5.1 setup?! Any ideas?

Thanks again guys.

You did get your Piro compensation set correctly didn't you?
If its set and your bird if fairly level you should be able to spin 4 or 5 times before it gets too out of sort.

[seattleite09]

Yeah, I did set the piro compensation, and tested it twice on the bench. Basically the gyro compensates and the swash maintains orientation when spinning the heli slowly on the bench. But up in the air, spinning the E325 feels a lot like spinning my Axe 100 CP.... one full rotation and the bird is all over the place, the rotor disc anywhere from 10 to 30 degrees off... requiring immediate correction. I was attributing that behavior to a cra**y Chinese gyro on the Axe 100, but the GT5 is in a different league.... so I am bit baffled.

[Bladecpnitro]

When in swash comp setup the swash tilts in one direction and if set correctly the swash stays tilted in that direction while spinning the heli.

If the head speed is very low it might cause that. Or if the P gain is low.

[gwright]

I've found higher I gain sometimes helps piro comp a lot . Can't get too high of course or you end up with bobbing in fast forward flight. Additionally, to high Derivative gain can show up in piro comp issues. however, If you have much more or less than 8 degrees cyclic range while in programming mode, that is more likely the culprit.

[mr_squiggle]

Have a listen to the RC Heli Nation podcast episodes 133 & 135. They explain very well the whole FBL tuning process & give real world examples. I think you'll find them invaluable.


Citizen #186

[seattleite09]

Originally Posted by gwright

I've found higher I gain sometimes helps piro comp a lot . Can't get too high of course or you end up with bobbing in fast forward flight. Additionally, to high Derivative gain can show up in piro comp issues. however, If you have much more or less than 8 degrees cyclic range while in programming mode, that is more likely the culprit.

OK, so lower swash D gain, increase swash I gain, and ensure 8 deg cyclic range.... I will try, and will report back.

With no wind today (or very mild wind), I experimented with piros a bit. I managed to spin 3 times in place (both directions), before I felt the heli needed leveling corrections. In forward flight, when tight-turning, I can keep pirouetting for a long time, and there is no destabilization of any kind. It flies really well. The only issue seems to be when spinning in place, while hovering.

I did stall turns, fast circuits & figure-8s today. Stopped short of entering loops but I felt I could have easily done that. It flies GREAT; I absolutely love it. It feels so much quieter yet more powerful than my Trex 250 -- it's just incredible. I can't wait to fly it again! I need more time and more batteries!!