Swash plate setup changing by itself Trad heli using F/W 3.4.6

Geoff,
Did you reduce the FILT params to 4. I didn’t see it in the second log file that you posted. I see that you reduced VFF. I think you could get a lot of benefit out of reducing FILT to 4 for pitch and roll. Your instability frequency seems pretty high and the low pass filter can do a lot to attenuate that frequency since it is much higher than the cutoff frequency I’m suggesting.

Chris & Bill
Got a good test flight in this morning, before the wind got up.It is flying much better with the latest Param’s file. Tried it in AltHold, again very good. I’ve attached the bin file and Param file. Was going to try Loiter with my next flight, but the wind got up.
Again thank you both for the help you have given me. I am the only Heli flier in our club, so any help is very welcomed.

2017-04-11 09-57-29.bin (3.9 MB)

2017-04-11 09-57-29.log.param (10.8 KB)

Geoff,
Your param file shows Roll VFF to be 0.3. Was that intentional? I’m guessing you meant to put in 0.03 like you have in pitch.

Bill
Accident, sorry about that. Which is the better setting?

You want to keep VFF as low as possible and use P gain instead. It looks like having the Pitch axis at 0.03 was acceptable and I noticed that you were able to increase the P gain a little more. Hopefully you aren’t seeing any signs of the instability even in winds. Put the Roll VFF to 0.03 and see how it feels. Increase it if you have to. I’m guessing 0.048 was as high as you could get the P gain. Once you set VFF to 0.03, try to see if you can get the P gain higher. If not then use VFF to get the desired response.

I was just looking at your yaw axis. It appears that you may not have the hover pitch set on the tail rotor. When you set up your tail rotor, did you set mid servo position to flat pitch on the tail rotor blades. That is the way it appears from the data. You are using over 50 percent of the servo throw in one direction to hover. You need to set the mid position of the tailrotor servo to the tail rotor pitch required for hover. If not I’d be afraid that you would lose control authority in one direction.

I do have pitch on the tail blades, There is no swing on the tail when taking off. Checking the tail setup I’ve noticed that there is play in the linkage from the crank to the slide, in the ball joint. I can add a bit more pitch to see if that will balance it out. Also I had a bit off drift back and to the left, looking from the rear. that I can trim out.

so if you go into mission planner under the data flight tab. In the display region below the hud there are several tabs available. Go into the status tab. There you will see a lot of parameters with digital readouts of the value of that parameter. scroll to the right until you see ch4out. push your rudder to the right until you see that value become 1750. That is the tail rotor pitch required for you to hold a stable hover. Change your mechanical set up so you see that tail rotor pitch angle at a value of 1500 for ch4out. I hope that makes sense.

Geoff, I would recommend fixing that. Chances are it’s just the hole worn out in the crank arm and it either needs to be bushed or replace the crank arm. Play in the tail linkage can cause a lot “hunting” back and forth on yaw, which is undesirable if you use a fixed camera mount. If you use a 3-axis gimbal probably won’t be a big deal. But it does make the servo work harder in wind to take up the play.

If it’s hard to get a crank arm for your heli I’ve fixed those before by drilling the bore on it and press in a small brass bushing that just fits the ball on the slider. I’ve also found a couple interesting cases on Align clone heli’s where the crank arm goes over the ball and the ball doesn’t actually pivot in the bore of the crank arm. It goes thru it and the ball is extending outside the bore, which causes play in it. On those I’ve shimmed the bearings on the crank arm to move it down a little and get the ball inside the bore of the crank, and the play is gone.

Your heli is actually looking really good, once you get it trimmed out and try it Loiter!

Hi Chris & Bill
I got a couple of test flights in to-day. I had the P value at 0.49, but I had altered the Param’s to your suggestions, beforehand. I’ve attached the bin file. As you can see i did get into Loiter and it just sat there, lovely. I then tried Land & RTL again with success. I still need to setup the Trim as it’s drifting to the left, from the back, and drifting back as well in stablize.
Since tten I did somw test flight in the garden, getting the P value up 0.090. I’ve enclosed the latest Param’s File

2017-04-21 10-08-58.bin (3.9 MB)

21-04-17 Heli.param (10.9 KB)

1 Like

Geoff, your setup is looking very nice. That should be a nice handling and well performing heli! So, evidently, turning down the low pass filter helped to get your P gains up more in the rate controller? When the helicopter holds position nicely in Loiter and it can nail an auto landing, you are very, very close. Now you can start having some fun with it!

When I stop shaking, Chris

Geoff,
I took a closer look at your log file. The P gains in this log file don’t match the P gains in your Param file. You had tuning selected for RATE ROLL kP (49) and it was set to 0.051 and the pitch gain was set to 0.058. Did you send the correct bin file? Or is the previous flight with the lower gains which is the bin file and since then you flew and were able to raise the gains but you did not include the bin file from that flight?

Bill
I raised the P value in my garden, doing short tests. the ones in the params file were after these tests. I was also trying to adjust the trims to hold flight position better in Stablise.

I.ve attached the Bin file when I tried Loiter in the garden (ZIp to fit it on) as well other things

2017-04-21 16-25-17.zip (2.4 MB)

I was wondering if I could get one of you guys to check out this log from my Protos 500.
I’m having a lot of issues with the heli drifting right (and a little forwards) in stabilize, if I level the heli on a known flat surface.
In order to stop it from drifting, I’m having to increase the AHRS_TRIM_X all the way to 0.175., which is nearly a 10 degree offset.
I’ve re-leveled the swash plate. and while there’s a little offset in the front servo at the min & max, the ‘roll’ servos are level.

I’m also getting a lot of toilet-bowl effect in loiter, but I’m not sure if it’s because of the rather extreme trim offset? Especially yawing will get it drifting a good 2-3 meters before it catches up.
2017-04-22 12-03-33.log.param (10.8 KB)

Log:

This log I start with the AHRS re-leveled and the increase AHRS_TRIM_X to a almost hand-off level.

Hi Jacob,
I’m not able to check your log file until I get home later tonight. Based on your param file and your description of the aircraft behavior, I have some thoughts to help improve the heli behavior.
From your param, I see the pitch and roll P gains are around 0.05 which is rather low. In my opinion, the toilet bowling in loiter is being caused by the low P gains. I would suggest lowering the FILT param for pitch and roll to 4 and see is you can get the P gains higher.
As far as the drifting and not holding attitude, I think that the integrator could be leaking off too quickly. So I would suggest making the pitch and roll ILMI param 0.2. Be aware that this will cause the aircraft to want to lean on liftoff if you are on uneven ground or if you make any stick inputs prior to liftoff.
One other thing you could try that could fix both of these issues is larger I gain values, like 0.3 or 0.4. In this case, keep the ILMI term zero. The only thing I don’t like with large I gains is that it introduces a lagged response to your inputs. Still flyable but not to my liking. Hopefully this is helpful.

Jakob, is your AHRS_ORIENTAION correct? I see you have postiive values for the AHRS trims, which will trim the heli into a roll right, nose up condition. If the heli is already drifting right, you should have to reduce the TRIM_X, maybe even to a negative value to trim it left.

It doesn’t make sense that if the heli drifts right, that you would have to increase the X trim value to get it to stop doing that.

Late night brainfart. It’s drifting left.

No worry - that happens to me during the day. All the time. :grinning:

Left drift is normal for a clockwise turning heli due to tail rotor side blow countering rotor torque. There’s a param called ATC_HOVR_ROL_TRM which is set to 300 centidegrees by default. I forgot to mention that before. What I do is put the heli on a level surface of some sort, set that param to zero, and then use my digital pitch gauge for setting blades to level up the heli on the table so the mainshaft is perfectly vertical. Then I adjust the AHRS trims until it shows zeo pitch and roll in the ground station.

Then take it out and fly it with no wind just in Stabilize and make it hover perfectly stationary for about a minute. And while you’re doing that you can observe the right-side tilt of the heli frame required to hold it there. Then pull the logs out of it and look to see what the actual degress of correction there was that was required to hover the heli in one spot. Adjust the AHRS pitch trim to match that, and add in the amount of ATC_HOVR_ROL_TRM in centidegrees required to hold it. Then fly it again and fine tune it with the AHRS trim.

I don’t know if that’s the right way to do it. But it’s the way I’ve always done it, and always ended up with good results. I typically end up at somewhere between 300-400 for the ATC_HOVR_ROL_TRM. But on my nitro 700, which carries a Sony camera on the right gear leg, hanging off the side of the heli so the weight is not balanced, it was only 220.

Here’s a photo of how I do that. You’ll notice the paint mixing sticks under the landing gear to shim it up until the digital pitch gauge shows the heli is level and mainshaft perfectly vertical. And then I adjust the AHRS trims to get zero pitch and roll shown in APM Planner2.

New photo by Chris & Kristin Olson

I agree with Chris’s method. That is a good way to set up those parameters

Thanks guys!. Looks like I have a limited weather window, so I’ll better get charging!