Settings for Mini Titan V2?

[seattleite09]

I flew my MTv2 with GT5.1 for the first time last week, three batteries, normal mode only.
Overall it flew really well, though I do have some questions:

1. Compared to my Trex 250, the MT felt "floaty". Hard to explain, though I suspect it has to do with the way the GT5.1 gyro stabilizes the larger heli. On the 250, tilting the heli will make the 3GX gyro compensate swiftly and heavily. The swash stays level at all times. On the MT, the GT5.1 compensates a bit, and the swash tilts in the opposite direction of heli's tilt, as expected; however the swash does not stay level. The compensation is nowhere near as heavy as the one induced by 3GX. Is that normal? Is there a way to set the GT5.1 to compensate more?

2. Can the floaty feel be attributed to the 450-size alone? On the simulator, a Mini Titan feels different from a 250, but not quite as floaty.

3. Tail blows a bit on a hard ascent. Can that be attributed to insufficient RPM? I am using the normal mode curves for thr+pitch as suggested by the MTv2 manual.

Here are my gt5 settings, for the reference..
P:120
I: 100
I D-stick:100
D-sense:30
Feed Forward:90
I-lim:25
Hover:4

Tail:
P:85
I:80
D-sense:10
Tail-D-DB:10
Tail Asym:0
Tail Dyn:20
DMA-cyclic:0
DMA-pitch:0
Smooth Stop:0

Thanks.

[seattleite09]

Flew some more today, and unfortunately after a few rounds of figure-8, I managed to crash it.... The "floaty" feeling was very strong. There was a breeze today, and the heli was destabilized a few times by it. I recovered every time -- but I would have expected the gyro to compensate and keep the bird locked no matter what. Something felt off.

I then climbed to 20m and applied FULL left rudder and no cyclic ==> the tail rotated slowly (surprising!), and the heli tilted backwards while the tail was rotating. I tried unsuccessfully to recover, but due to tamed down D/R & Expo, wind, and low altitude, I quickly lost orientation and control -- it happened so fast.

So now I am left scratching my head wondering what the heck happened:
1. Why didn't full rudder have a quick effect on the tail? I would have expected a full 360 rotation in <1 second.
2. More worrisome: Why was cyclic affected on full rudder input????

[Bladecpnitro]

How much Expo and DR are you using? I've never used DR on my 450 and the tail spins very fast. I have the rudder endpoints set to 150 in the transmitter. That's for rudder only though.

I have Expo turned off completely in the gyro and set to 30 in the TX.

Sounds like your bird is too tamed down to where it can't react fast enough to bail out of an issue.

Also how degrees of cyclic pitch are you getting during setup?
As well as pos/neg pitch?
What kind of rudder servo are you using?

[gwright]

the "floaty" feel you mention is sometimes (well, most of the time) due to too high derivative gain (d-sense on swash). Try dropping it from 30 down to 10 and try, then adjust from there. It will probably be pretty obvious which way to go,..in at least 5 point increments. By the way, it goes negative also, so 0 is not the minimum <G>.

[seattleite09]

Originally Posted by Bladecpnitro

How much Expo and DR are you using? I've never used DR on my 450 and the tail spins very fast. I have the rudder endpoints set to 150 in the transmitter. That's for rudder only though.

I have Expo turned off completely in the gyro and set to 30 in the TX.

Sounds like your bird is too tamed down to where it can't react fast enough to bail out of an issue.

Also how degrees of cyclic pitch are you getting during setup?
As well as pos/neg pitch?
What kind of rudder servo are you using?

Expo was set at about 40%, and d/r was 70%. Rudder d/r was 100% and expo 20%.
I am getting 10deg positive pitch, and -16deg negative pitch. Not sure how to set the two to be equal without using non-symmetrical servo travel numbers in gt5....
The swash is level, and blades aligned with 0deg pitch at mid stick.
Rudder servo is Futaba BLS257. Push rod in second hole of servo arm. I don't recall what the tail servo travel is set to, but I tweaked it to eliminate binding.

You may be right about d/r. I am reconfiguring everything to use 100% d/r. I generally need to use expo -- otherwise I find mid-stick too sensitive. Even on the simulator...

[seattleite09]

Originally Posted by gwright

the "floaty" feel you mention is sometimes (well, most of the time) due to too high derivative gain (d-sense on swash). Try dropping it from 30 down to 10 and try, then adjust from there. It will probably be pretty obvious which way to go,..in at least 5 point increments. By the way, it goes negative also, so 0 is not the minimum <G>.

Thanks Gary. That is very useful info... and btw your contributions on these forums are amazing; keep up the great work.
I reduced D to 10, and will try again. I just finished rebuilding. Hopefully the next flight will have a better outcome!

Another update+question. I figured out how to increase the swash's response to "tilting". Apparently bumping up the swash "I" factor goes a long way. In my original numbers, I had "I" set to 100. I bumped it up to 125, and the difference is substantial on the bench. Now the gyro reacts a lot closer to how the 3GX reacts on my 250. Question: now that I have "I" set to its maximum possible value, is that too much? I read that having large values for P and I may make the heli oscillate along one or more axes.

Also, the GT5 manual recommends setting up a gain switch, but I could never achieve that with the GT5.... I am connected via SBUS. Using the gain switch 50<->100% on my Futaba TX 8J does absolutely nothing. The gain number on GT5 stays at the same value. Is there anything special that I need to enable in GT5 to make the gyro "get" the value sent over by the TX?

Thanks again.

[gwright]

It should read channel 5 for gain,..or it may be channel 7, I forget the default. primary Gyro gain on futaba radios is usually channel 5. I don't know of an exception to that. If you set the gyro gain to a switch with one at 100 and the other direction at 0 you should see a difference in tilting the heli and swash response per gyro. No, don't fly it at 0, it's just a quick way to determine gyro gain is working from a switch <G>. If that doesnt' do anything, I'd try channel 7. On your transmitter, set channel 7 to a switch and try it at 100/0/-100 to see the difference in response (on the workbench, just to confirm the right channel is mapped to gyro overall gain). It sounds like it's reading channel 7 and that channel is set to 0 in your transmitter. Just go to the servo graph screen and look at that channel. I'll bet it's at zero (center), and needs to be at 100. You're basically telling the gyro the max P gain is there. If that channel were at say 10% movement, then the numbers in the gyro (0 to 125 on the gains) are adjusting from 0 to 10, in 125 steps. I'll be when you start adjusting gyro gain from channel 7, you'll see a big difference. Another clue to that is that your P gain is maxed and you've now maxed your I gain.
Think of P gain as the standard gyro gain adjustment we've had for years. I gain is the integral gain that averages out the response for samples taken over a period of time. It's for consistency (like for a tailrotor, if it "whips" doing pirouettes in forward flight and the mechanics are correct, the I gain is too low). Having I gain too high can create other problems. As you speed up into forward flight you can start to porpoise up and down, which can get rather nasty in a hurry. Imagine going high speed forward and then starting to oscillate the elevator stick from full down to full up, back and forth. That's an exageration, but thats the general effect of I gain being too high. As for the numbers, it generally needs to be at or below the P gain number. If it were me, I'd default to unit, set up servo freq/speed/directions/limits, set P and I both to 60 to start, and adjust from there. different units will respond drastically different in non-flying mode (tilting the heli) even when they fly the same, so that's not a good judge of the responsiveness of the unit, and definately nothing to compare by. I tilt the model on every axis and watch the swash, EVERY time I walk out to the flight line to fly. Call it paranoia <G>. On my CGY750 models you can almost hear the swash go "thunk thunk" hitting the limits. On my GT5/5.1/5.2 models you can barely see the swash move. flight performance of both is great.

[seattleite09]

Thanks Gary!

The 8J has gain settings ("gyro sensitivity") for rudder only, rudder+ele, rudder+ele+ail. Do people normally set these gains for just rudder, or for everything? Does it matter? (sounds like at the end of the day it's just one number that's being sent over to GT5).

Also, the numbers are either 0-100% in STD mode, or -100% to 100% in GY (AVCS?) mode, as shown on page 88 of the manual:
http://manuals.hobbico.com/fut/8j-manual.pdf

Is GT5 a standard gyro, or an AVCS gyro? When you say "0%" above, are you referring to 0% STD, or 0% GY? (they seem different)

[gwright]

The GT5 and 5.1 with original firmware (according to your numbers you don't have the new firmware and not sure it could be loaded into your unit), read a single channel from the radio and use that for overal gain control (both swash and tail). If you have it at 100, then the number you set in the gyro is your gain. If you have it at say 50%, then the gyro gain would be 50% of the number you set in the gyro. AVCS and NOR just mean which side of center that chanel is on. The futaba radios, 8ch and up, have provisions for the futaba CGY750 gyro, which can accept gains for rud/ele/ail seperately,.. using three channels. That's what you're seeing in the radio. The GT5 will only read ONE of those for overall gain. It's probably the one mapped to channel 7.

[seattleite09]

Originally Posted by gwright

The GT5 and 5.1 with original firmware (according to your numbers you don't have the new firmware and not sure it could be loaded into your unit), read a single channel from the radio and use that for overal gain control (both swash and tail). If you have it at 100, then the number you set in the gyro is your gain. If you have it at say 50%, then the gyro gain would be 50% of the number you set in the gyro. AVCS and NOR just mean which side of center that chanel is on. The futaba radios, 8ch and up, have provisions for the futaba CGY750 gyro, which can accept gains for rud/ele/ail seperately,.. using three channels. That's what you're seeing in the radio. The GT5 will only read ONE of those for overall gain. It's probably the one mapped to channel 7.

So does it matter which "100%" I choose, then?
100% (in STD mode)
100% AVC (in GY mode)
100% NOR (in GY mode)

The first two correspond to "all the way up" / "all the way to the right" on the graph in the manual. The third option corresponds to "all the way down" / "all the way to the left" on the graph in the manual.

The GT5.1 manual doesn't mention anything about AVCS vs. standard.

Thanks again.

[seattleite09]

Originally Posted by gwright

The GT5 and 5.1 with original firmware (according to your numbers you don't have the new firmware and not sure it could be loaded into your unit), read a single channel from the radio and use that for overal gain control (both swash and tail). If you have it at 100, then the number you set in the gyro is your gain. If you have it at say 50%, then the gyro gain would be 50% of the number you set in the gyro. AVCS and NOR just mean which side of center that chanel is on. The futaba radios, 8ch and up, have provisions for the futaba CGY750 gyro, which can accept gains for rud/ele/ail seperately,.. using three channels. That's what you're seeing in the radio. The GT5 will only read ONE of those for overall gain. It's probably the one mapped to channel 7.

So it turns out the gyro sensitivity / gain setting is locked to ch5 in the Futaba 8J TX, but the GT5 expects it on ch7. There is no way to change the channel mapping in either one. No wonder my TX gain switch had absolutely no effect....

I disabled that switch, and mapped a turn knob (VR) to ch7 instead. Et voila.... I have an adjustable gain knob that goes from -128% to 128% in increments of 1!

I can now see the gain number on the GT5 screen changing from 128 (black on white background) to 128 (white on black background), while turning the knob. The gain setting has a substantial effect on how much the swash reacts while tilting. The default value in GT5, without this gain setting in place seems to be 49 black on white bg.

So now the next question is.... should I be aiming for white on black bg, or black on white bg? Does it matter?

Next time I fly, I will probably lift off with the default (49), and tweak the gain slowly while in flight. Any concerns about that procedure?

I set the swash P/I gains to 60, and D to 10, as you suggested.

Thanks!

[mr_squiggle]

I started with the default of 49 as well until I figured out the gain adjustment. On a Trex450 that resulted in a right aileron roll when exiting a stall turn & the piro comp was crap. I suggest you set the gain at 60 for normal mode & switch to 100 for idle up 1 & 2. After testing this works pretty well.
The number on a white background means the tail is in rate mode. The gain number on a black background means the tail is in heading hold mode.


Citizen #186

[gwright]

I think terminology is confusing you a bit. AVCS (angular velocity control system) is Futaba's term for heading hold. In the gyro screen, if it says AVCS, that channel is on one side of center, if it says normal it's on the other side of center. Imagine a servo plugged into that channel. the servo would be centered if you had 0 gain, If you increase gain the servo will move to the extreme. Change from avcs to nor, the servo goes the other way. The three gyro settings on the gyro screen (rud/ail/ele) are merely three channels for this adjustment. the futaba gyro allows independant adjustment of gains for rud/ail/ele, hence this screen gives you that ability in a convenient location. I personally don't like adjustments for things like gain on knobs or slides. Too easy to bump them and change gain between flights. Gain is set initially and stays there so I like it a definitive number in the transmitter, not a knob, so I wouldn't put it on a knob. (personal opinion/preferance, you can leave it if you want it this way). I just checked with Josh in product support. I'm at my desk and don't have an 8fg here, but he does <G>. He said defaults are for the three gyros to be on channels 5/9/10, so you would want to remap one of them to channel 7 and use that gyro gain setting. This is the same as my 14SG so it's easy. simply go into the link menu then scroll down to "function", you can remap any function to a different channel. this way you could set it so that on the gyro screen the rudder gyro settings are on channel 7 instead of channel 5, and you'd have overall gain adjustment in the gyro screen (using the rudder gyro settings)

[gwright]

Just thought to look back,.. it's 8J, not 8FG,.. sorry. I checked with some folks here, and the 8J default channel 7 is governer, so you could use the governer screen if you wanted. Apparently you can't reassign all functions on the J like you can the FG. If it were me, I'd just use the governer screen.

[seattleite09]

Originally Posted by gwright

Just thought to look back,.. it's 8J, not 8FG,.. sorry. I checked with some folks here, and the 8J default channel 7 is governer, so you could use the governer screen if you wanted. Apparently you can't reassign all functions on the J like you can the FG. If it were me, I'd just use the governer screen.

Yes, it's 8J, and you are correct -- the gyro gain channel can't be reassigned. It's locked at ch5.

But hmmm.. using the governor setting instead makes a lot of sense. Not sure why I didn't think of that. I will give it a shot.

After I set things up yesterday, I started having the same concerns as you. How will I "lock" the gain settings once I determine the optimal ones? etc. Using the knob seems too easy to accidentally change the settings without realizing. So yeah, if I can use the governor menu & settings instead that would be fantastic.

Thanks again!

[seattleite09]

Originally Posted by mr_squiggle

I started with the default of 49 as well until I figured out the gain adjustment. On a Trex450 that resulted in a right aileron roll when exiting a stall turn & the piro comp was crap. I suggest you set the gain at 60 for normal mode & switch to 100 for idle up 1 & 2. After testing this works pretty well.
The number on a white background means the tail is in rate mode. The gain number on a black background means the tail is in heading hold mode.


186

Thanks @ms_squiggle. I will try that as well. Good to know someone already tried the default 49 and didn't quite work well. It will save me some time.

What's the difference between rate mode vs. head holding mode in actual gyro/stabilization functionality? Does the gyro stabilizes differently in the two modes? Which one is the preferred one for a 450-size heli such as the Mini Titan v2, in Normal mode? How about in IU1/IU2?

Thanks.

[gwright]

heading hold, or "AVCS" is ALWAYS the preferred mode,..doesn't matter which heli. You just may find someone flying scale where they want a bit of weather vane effect so they use rate mode,.. but that's probably 1 out of a few thousand.

Rate mode simply damps unwanted movement. It will also "fight against" you when you actually want the heli to rotate. It simply says "I'm sensing zero rate of rotation, everythings fine, oops now it's rotating, I must feed in some opposite control". If you do a bit of research on PID control loops, you'll see it's a "P" only mode. With heading hold, the gyro senses the rate of rotation it's experiencing but also compares that to what you are commanding from the transmitter,.. and computes what amount of servo movement it needs to input in order to give you what you want. since it samples this repetitively it can smooth out responses by comparing multiple samples over time and integrating that into it's calculations for consistency of rotational rate (Integral gain effects this). Since it knows what it's experiencing and what you are commanding, it kinda knows the future of where it needs to be in rotation, and by calculating it's acceleration/deceleration rate in rotation it can further affect the correction amplitude so that you "arrive" at the commanded rate of rotation quickly (Derivative gain, or D-sense in the GT5).

In heading hold (AVCS in futabaspeak) the control may not be centered, i.e. the swash may be tilted,.. and it just stays there although the stick is centered. It's sensing zero rotation, you're commanding zero rotation, so it doesn't think anything needs to be done. This sometimes confuses people since they turn everything on, move collective up and down to see that it's working, and the swash does strange gyrations all over the place. it looks like somethings wrong, but it's not <G>. Some units have a decay built into them so they will SLOWLY try to reach center on all the servos ;when the sticks aren't being moved. The GT5 has a decay to neutral like this but it diminishes as you go away from neutral stick (0 pitch), such that you may see the swash canted over like you have the stick in the corner. Don't be alarmed, it's all OK,.. and when it actually senses movement in the air it will work properly.

[mr_squiggle]

The governor in the tx works by assigning a constant value to a channel (ch 7 in your case) that corresponds with the target headspeed required. The governor unit itself on the heli is the part that makes the throttle changes to actually maintain the headspeed.
We can use this to our advantage for the GT5 by experimenting with the setting to get the gt5 gain number we want.
On Futaba the governor can be set differently based on flight mode (also known as Condition on some models). So set your governor number to get the gt5 gain number you want for the flight mode you are in. The actual governor number itself is not important.


Citizen #186

[seattleite09]

So... I tried lifting off with total gain = 60, P=60, I=60, D=10. The experience was weird. The heli was not capable of a stable hover. The most bothersome symptom was the fact that in the air the heli would tend to get out of a hover by itself, and to drift back and right -- or whatever other direction. I didn't see any consistency.

I also tried 50 total gain, not a big difference. I flicked the switch to 100 total gain, with the P/I/D params above; the flight felt even stranger. There was no stability. The heli wanted to eventually go in some direction by itself after carefully being placed in a hover. I basically could not count on the gyro to keep the heli even remotely stable for me. This is something I never experienced on any of the CP helis I have flown, 100- and 250- size, nor on the simulator. There's something wrong here....

I spooled up the heli on the bench, with the blades off, to take a look at the gyro at >0 rpm. I spooled up to about 70% throttle. The gyro was going bananas, the A/E/R bars were dancing within 50% off center in both directions, probably servos working non stop too.... This had me thinking that perhaps the gyro was experiencing too much vibration!... Could that account for the weird symptoms that I am seeing?

My GT5 was attached to the frame using thick hook&loop velcro. I replaced that tonight with a couple of (different) layers of mounting tape of different consistencies/materials. At the bottom, a red+black foam tape that came with the gyro; on top 3M transparent mounting tape (gel-like). Then the GT5 on top. Pressed in real well, to make sure the gyro is attached securely and there is no play.

After spooling up to 70% throttle again, I am seeing a dramatic reduction in the vibrations picked up by the gyro! Basically the bars now stay at 0, and occasionally fluctuate lightly within 5% off center.

I haven't had the chance to fly with the new dampening tapes on, but I'll report back as soon as I fly...

Also, could the heli be tail heavy, affecting the gyro's ability to keep the heli in a stable hover?


Thanks.

[mr_squiggle]

All gyros, flybarless units etc must be secured with the correct/proper gyro tape to work properly & I'm glad you discovered that without crashing! Vibration is the enemy of gyros & FBL units so you were spot on with your diagnosis on the bench. The bars on the display of the GT5 show you the correction that it thinks it needs to add to your heli based on your inputs vs the measured elements.
It's a good idea to use the special tape supplied with the FBL unit as the manufacturer has tested and calibrated with that tape for best results. Almost all manufacturers will give different mounting scenarios & suggested tape configurations depending on the type of heli (e.g. nitro helis shake & electric is smoother).
I assume you've balanced your main blades at least? If not, have a look in the forums or on John Salt's website rchelicopterfun.com for how to do it. It's important to minimise vibration everywhere on your heli for best results.
Your question about being tail heavy is a separate issue. You should always check your centre of gravity (CG) when you setup the heli. Have a look at www.rchelination.com and look I the Tech Tips area for an easy way to check this.




Citizen #186