Tail gains

[cjm180]

Hey

as mentioned, i am struggling getting rid of a sudden very fast oscillation in the tail. Last night i did a full collective punch out, at the top when i shut off bang its trying to shake its self to bits with a tail oscillation.

I lowered the gain and flew again tonight. Did the punch out and it was fine. Towards the end of the flight i was coming out of a loop quite fast. As it levelled it attacked again!

I am runing a low head speed on the tioblin 570. 2150rpm. I know usually u have higher gains for lower head speeds. I feel the gains are getting low so not sure what to do next.

Any help would be appreciated!

Attached is the screen shots of my screen:

[Grumpy]

Gain too high without a doubt, reduce it some more. There isn't a 'one size fits all' value for tail gains as it depends massivly on the mechanical setup, so dont read anything into it if the % number seems lower than expected. It's best to start low with gains, increase bit by bit until oscillation just starts to set in, then back it down to the last 'oscillation free' setting.

[cjm180]

Hey grumpy

i started low initially just flying round slow. I did some piros and sudden stops and dialed the gain there. As I have pushed a bit more u feel the gains are too high? Ok I will take them right down and start again.

Just for my info, the tail gain on the diagnose page is the value my Tx is outputting. On the head / tail page the tail gain quoted is not effected by my tx. Am I correct in thinking this gain is an overall gain and simply is used as a multiplier for the other gains?

As my tx gains are getting low, should I reduce the over all gain now then go from there? Or do I still carry on lowering the tx gain?

[Grumpy]

I'm not familiar with the BD Axon, but i'd suspect that if the values in the software aren't changing when you adjust in the Tx that the channel isn't correctly mapped so isn't actually doing anything. For sure with the Vbar when you adjust in the Tx you can see the value change instantly in the software. Better not take my word for it though just in case the Axon works differently.

You always need a bit lower gain value for aggressive and fast forward flight than you do for hovering and slow flying, so nothing unusual there.

[crash-test-tomo]

+1 mechanical setup. Cant say im familiar with BD either but if it has any form of precomp for the tail you usually need to be sure you have 0 pitch on the tail blades at center stick.
Got a nasty wag from my 700 comp doing an overspeed a couple of weeks ago. Hadnt changed any settings but had worked on the tail, blades were out at center stick so I adjusted the pushrod to suit, wag went away

On the diagnose, head/tail page is there an option for the gain to be set in software? That would mean you have to set the value in the BD software not the tx. I think this Is so it can be tuned on a dail rather than a menu in the transmitter?
If so, you should be able to untick the box and go back to the tx

[myxiplx]

Coming out of a loop is a definite sign of tail gain too high, you'll need a lower gain for high speed moves, so if you've previously tuned it in a hover it's likely too high.

At the top of a collective punch out is an odd place to see a tail gain issue though, I suspect that's more likely a governor gain too high. It'll be pushing hard to maintain headspeed and when you suddenly unloaded the head I'll bet if you had logging you would have seen the headspeed shoot up, and then aggressively drop back down. That'll cause a tail wag every time, but it's nothing to do with the tail gain.

[crash-test-tomo]

I suspect that's more likely a governor gain too high.

I was going to say this too.. Easiest way to check is just turn your governor off.

[p1andy]

Agree with the above, at the top of a punch out you would not expect the tail to oscillate due to tail gains

[Grumpy]

Though in reading Chris' first post he does say that reducing tail gain did stop the oscillation at the top of the collective punch out, so that one appears to be solved.

[cjm180]

Thanks for all the replies. I will do some checking I think.

I think I have got to the bottom of the gyro gain defined both in the diagnose page and the head / tail page. Page 8 of the axon manual. The axon works with a signal defined as -100 to +100. Anything less than 0 is normal mode, anything above 0 is heading hold.

So, the axon uses what ever is shown in the diagnose page, which is moving in the software for when I change the gain on the TX. EXCEPT, when zero is put out from the tx. When the TX puts out zero, the axon uses the value set on the head / tail page.

I did look at the logs for the castle esc, through out the flight the head speed did not change more than +- 50rpm. Its a similar story on the 600l and I don't have the issue on that. However, the gov gain in the castle software is set to medium on the 600l, but low on the goblin. I can up it to medium and see if it makes a difference.

I just felt the tail gain was getting low at 44! I will take it down to 40 and then 35 and see if it helps. Its been absolutely perfect in circuits.

I will check I have zero pitch at centre stick on the tail, I will have another flight with the lower gains and then pull the logs for the ESC. what am I looking for in the logs? over / under speeding? is +- 50rpm good?

[cjm180]

Hey Dan,

I cant see any pre-comp settings. I remember discussing the tail setup with DemonJoe who works for BD. If I remember correctly, the goblin does not have zero pitch on the tail at mid travel of the slider.

Joe said its best to have equal throw either side than have zero pitch in the centre. If I have no luck with the next round of adjustments I will set the tail at zero centre stick, I will just limit the bigger side to the value I am getting on the lower side if u know what I mean.

I will put the tail gain back on the dial and go from there. Will report back after next flight.

[crash-test-tomo]

I think its pretty common for helis to have a few degrees of tail pitch with the arm at 90? Though iv personally had better results on my goblins (flying brain dont forget) setting the tail at 0, end points set as far as the slider will travel each way without the servo binding. If you think about it the heli doesnt actually need as much left rudder anyway as it naturally wants to nose left

[Grumpy]

Chris,
Personal I'd tackle this thing one thing at a time. If reducing tail gain solves the problem (and I'm 99.9% sure it will) that the problem is solved, end of story. You don't need or want to go playing with other settings unless there is some remaining issue. Same goes for setting the tail centre position, I'd just leave it as it is providing the oscillation problem goes away when you adjust tail gain.

[myxiplx]

Yup, adjust one thing at a time, and don't worry about the actual value of the tail gain setting. The figure you need will be different on every model, tune it by how it flies.

[Crashalotjoe]

I did look at the logs for the castle esc, through out the flight the head speed did not change more than +- 50rpm. Its a similar story on the 600l and I don't have the issue on that. However, the gov gain in the castle software is set to medium on the 600l, but low on the goblin. I can up it to medium and see if it makes a difference.

I just felt the tail gain was getting low at 44! I will take it down to 40 and then 35 and see if it helps. Its been absolutely perfect in circuits.

I will check I have zero pitch at centre stick on the tail, I will have another flight with the lower gains and then pull the logs for the ESC. what am I looking for in the logs? over / under speeding? is +- 50rpm good?

You don't need to see a large change in head speed to cause this, it is the change in torque being applied by the motor that causes the tail to kick (after all, the governor is trying to keep the head speed constant). If torque is applied or reduced too quickly, the tail has to react to it. It does seem strange that you are already running with low governor gains though. This would normally be an indication of the governor gain being too high. Depending on which way the tail kicks, it could also be that the tail gain is too low and so not able to react to the torque changes appropriately.

The other possible scenario is that the castle logging sample rate is too slow to catch the head speed dip or increase (i.e. the change is happening between two samples and therefore not being seen). If this is the case, a higher governor gain may fix it but I think this is a bit of a long shot really.