Oct 31, 2012

Flybarless Tail Tuning with V-Bar/Logo Example

Some of you may recognize this tech tip from my original tail tuning guide in the V-Bar forums on Helifreak. I originally wrote this in response to requests for a sticky regarding Logo/V-Bar tail issues, but I’d like to emphasize that this is applicable to pretty much any heli or flybarless system. For troubleshooting purposes and since this was originally motivated by questions in the Mikado Logo Helicopter forum, I’ll stick with the stock Logo plastic tail assembly as the detailed example below, but if you have an aftermarket tail on your Logo or if you’re reading this and have another heli entirely, you can apply it without a problem and should just look for the same issues, but on the specific component details for that heli design. Please feel free to post your feedback, experience with this or other methods, or suggestions for modifications or improvements.

It’s of critical importance on any heli to make sure that your tail is behaving itself, as few things can more readily rob a pilot of confidence than a bad tail. The most common tail issue is a wag, which can manifest itself in may different forms. 


NOTE: Before performing any tail tuning on a new or troublesome helicopter, turn off your headspeed governor!

The governor control loop can cause tail wags very similar to the issues discussed below if it is not tuned correctly. Once you get your tail tuned on a on a throttle curve, you can enable your governor and tune it knowing that you’re not fighting your tail mechanics or the flybarless system.

The first thing that I tell everyone is to check the smoothness of your tail slider assembly. To be clear, this means your entire tail control assembly from the servo all the way to the blade grips. I’ve seen tons of people say it’s smooth, but then when I feel it, it’s chunky, rough, and just bad. This is not to say anything about you or your building skills, but a lot of people suffer from not having perspective that is only gained from building a lot and flying around others. Trust me, I’ve built a TON of reallllyyyyy “smooth” tails… 


1. Checking the Tail Slider Mechanics:

First, pop the control rod link off the belcrank and move it over it’s travel range. Is it less smooth at certain points along the shaft? Right rudder (slider towards the tail case) seems to be the most common portion of the travel to be rough. This is usually due to a combination of the smoothness of the plastic balls or other linking mechanism on the blade grips and the sizing of the pitch slider links.

 Right Rudder

The plastic Logo grips come out of a mold and the mold lines run right through the center of the balls. Clean them up with some sand paper and give it another go. This is usually an iterative process and you should be mindful of not over-sanding the balls. Once they appear to lack any sign of mold lines, you should check the smoothness again, but you may still need to go over the actual plastic link arms with a ball link sizer. Anytime you’re using a sizer, less is more…you cannot put plastic back if you screw up, so be patient and careful!

For other helis or tail slider linkage designs, check to make sure that your linkages are not binding in the right rudder region of travel. This could happen due to too tight links on metal balls or some roughness between the link and bushing in the bushing-type designs. Again this applies to all helis, but you’ll just have to think about your particular design to figure out what specifically to address.

Left Rudder

If the binding occurs in the “left rudder” portion of the travel range, it’s more often then not due to the plastic cup or other mechanism on the bell crank by which the pitch slider is actuated. The height of the cup can sometimes be excessive, which causes it to rub on the underside of the pitch slider housing. Sand the cup from the top side slowly so as to not take too much material off and recheck.

Full Travel Range

If the binding/roughness occurs over the entire range of travel, then it’s most likely your shaft/bushing interface. I like to use Mothers Aluminum/Magnesium polish with a micro-fiber cloth and polish the shaft a bit so that it’s silky smooth. To do this, remove your main blades and get a nice microfiber cloth with some Mothers on it. Spin the head while holding pressure on the tail output shaft with the cloth. Give it a good 30 seconds of spinning and then find a clean portion of the cloth and repeat. The end result should be a super-polished output shaft. Clean it off with a bit of alcohol and then re-grease it with Tri-Flow and re-assemble.  


W.O.D. Check

WARNING: Only attempt this if you are fully confident in your ability to control your helicopter in any situation! If you get spooked easily, are not confident in all orientations, or are not comfortable with controlling the helicopter during full-speed pirouettes, then DO NOT attempt this!

A good check for whether binding is the issue is to get your heli in a hover at an altitude that you’re comfortable with and do a few full-throw piros with hard stops (“dropping the stick”) in each direction. If the heli enters the dreaded Wag of Death (W.O.D. is characterized by a mild-to-wild 15 to 90-degree tail wag that comes on upon stopping the piro and makes a loud, angry buzzing), then remain calm and simply give rudder input in the opposite direction and it should stop. Now you know you’ve got a binding issue… 


2. V-Bar Tail Fundamentals:

If you believe that you’ve exonerated all of your potential mechanical causes, then move on to the V-Bar. The V-Bar software gives you the choice to tune using two standard parameters on the front panel called yaw rate and gyro (hereafter referred to as “common”) gain. Yaw rate is just what it sounds like. How fast do you like your full-throw piros? Not fast enough for you? Increase it and have fun! Common gain is your box-standard tail gyro gain knob and will affect how well your tail holds through various maneuvers.

Now, that’s not all that you can play with… As most of you know, there’s an expert menu as well. Contained within is a sometimes-intimidating list of parameters that most people don’t understand completely unless they’ve had the opportunity to take a control theory class in engineering school. For now, I’m only going to focus on a few of them:

P Gain

Known as proportional gain, this parameter affects the responsiveness or sensitivity of your tail control loop. P gain can sometimes be the culprit behind what most would consider a fast tail wag. 

I Gain

The integral gain parameter is what most know as “heading-hold” gain. This is the little guy that’s responsible for making sure that your tail stays where it’s supposed to throughout the constant thrashing that you’re giving it! For those of you familiar with the BeastX flybarless system, Parameter Menu D (Tail – Heading Lock Gain) is the same exact thing, except BeastX doesn’t allow you to change P gain separately.

Precomp (Collective and Cyclic Feed-Forward)

The precomps are parameters that allow your flybarless system precompensate for torque or load changes in such a way that it minimizes the disturbance to your tail control loop. In other words, when you’re changing collective or cyclic, you’re varying the load on the machine, which results in changes in torque. The tail control loop will attempt to minimize the effect of these changes, which you see as tail kicking or deflection in certain maneuvers. I gain and precomps works together, but generally must be tuned inversely.

Common Gain

This parameter actually varies total gyro gain as a combination of both P and I gains. This is why MrMel suggests that you set your P and I gains to equal values (i.e. 75/75 or similar). By doing this, you can then vary your common gain slider and get equal action from the proportional and integral parameters. I actually like to think of it as a multiplier, if you will… In other words, the common gain slider sets the coarse level of your P and I gains and the actual P and I gains act as fine adjustments about the common gain value.

Real-World Example

Ok, so before we start messing with the heli, let’s  try to bring this crazy stuff down to Earth. Consider a situation where you’re driving behind one of those awesome people who ride their brakes and you’re doing your best to keep your car some constant distance away from this person. Say that we’re trying for 5 car lengths or the distance at which you can just make out the license plate number of the other car.

Your internal P gain is what determines how quickly you apply your brakes when you see the other car’s brake lights go on and subsequently how fast you apply the gas after the braking event to get back to your 5-car length separation. So, if you have too high of a P gain, then every time you see the brake lights go on and visually sense that the other car is slowing, you slam on your brakes and your tires screech and then once the lights go off, you slam on the gas. In other words, you’re way too sensitive to slight changes. Now think of what this would look like if the other car’s brakes are going on and off rapidly…your car is going to be bouncing all over the damned road, right? Conversely, if you’re P gain is too low, you won’t react quickly enough and may actually run into the back of the other car or fall far behind it.

Your I gain acts to maintain your 5-car length separation distance during the braking or speeding-up events by looking at the response of your car over time and changing the brake or gas pedal input accordingly. So, for example, you’ve just applied the brake and have now fallen behind farther than your desired distance from the other car. You’ve now got your foot on the gas and as you see that you’re approaching the point where you can begin to make out the license plate numbers, you progressively lessen pressure on the gas pedal until you reach the exact 5-car length distance. Nice and smooth!

Now, for precomps…let’s say that the person in the other car is actually a friend of yours and you are on your cell phones talking to each other. What happens if your buddy tells you “I’m about to slow down”…with this knowledge, you’re now able to anticipate the braking event and begin ever slightly slowing to prepare for it. If you do it correctly, then you’ll have slowed just enough so that when he applies his brakes, you haven’t lost much ground on your 5-car length separation. If you’re not at your exact 5-car length distance, then it won’t take you much additional input to get there.


3. V-Bar Tail Tuning:

Ok, you’re an expert on tail control theory now, so let’s tune up your heli. There are a few things that you need to do to be able to tune your tail correctly. Once you get this algorithm down, you should be able to tune any tail in just a few dedicated and focused flights. In the past, I have suggested starting with stock V-Bar tail settings (P = 80, I = 60), but lately I’ve found quite a bit of utility in setting these values equal per MrMel’s suggestion. In the interest of remaining consistent with Frederik, let’s use 75/75 for P/I gains.

Before going further, a quick word on precomps…

My general rule of thumb is to lower your collective and cyclic precomps by 3 and 2 points, respectively per 5 points of I-gain increase. This means that with stock settings, if you increase your I-gain to 75, then you ought to be running collective and cyclic precomps somewhere in the range of 13 and 3, respectively. Please keep in mind that this is simply a guideline that I’ve found to work in most cases. You will need to fine-tune once you get in the ballpark and as you well know, not all birds are the same.

Always remember I Gain’s effect on precomp because if you change your I-gain later during tuning, you’ll need to readjust your precomps as well. I-gain and precomps are inversely proportional in effect. In other words, if you lower your I-gain (reduce the amount of overshoot you allow), then you need more precomp to get the same effect. Conversely, if you increase you I-gain, you may need to lower your precomp. Remember to do the same for other flybarless systems as well. So an increase in the BeastX Parameter Menu D (Tail – Heading Lock Gain), should be followed by a reduction in the revo mixing (precomp) parameter.

Back to the tuning…

Common Gain

The easiest way to set your common gain is to do funnels. If you can’t do funnels yet though, then fly in fast-forward or fast-backward flight and throw a hard turn in there. What you’re looking for is usually pretty obvious… If the common gain is too high then you’re going to get a hard wag or a buzzing where the tail sounds pissed. In this case, start off by dropping it about 5 points at a time while repeating the funnel. My rule of thumb is to find the gain value where you can JUST BARELY hear the tail get angry in a hard funnel and then back it off by a few points or percent, depending on your flybarless system. If, on the other hand, the tail is not holding or altogether blowing out, then you need to increase your common gain. Find the right common gain value and once you’re confident that the tail will hold and not over-gain, move to the next step… 

I Gain

To get your I gain dialed in, you should go into fast forward flight and do full-throw piros. If the piro rate seems inconsistent, then your I gain may not be set correctly. Too high of an I gain can result in what most would call whipping, where the tail whips around the heli as it piros and too low of an I Gain can result in sluggish or mushy looking piros. You want your heli to be able to piro at a constant rate throughout the forward-flight pass. Once you’ve found an I Gain that gives consistent piro rate, re-run the common gain check as described above and then recheck precomps. Again, the same goes for the BeastX Parameter Menu D (Tail – Heading Lock Gain) setting. Usually, one of the first two lights is on this parameter is sufficient to provide good tail holding control without excessive whipping.

Fast Wags

Too high of a P gain can cause what most would consider to be a “fast” (>2-3Hz), but small amplitude (a few degrees) tail wag. I like to decrease P gain on the V-Bar by 5 points from 75 until I get down to around 60. If it doesn’t change the behavior at all, then I would automatically assume that I’ve missed something mechanically…Go back to #1 and try again. If a lower P gain does fix it, then re-run the common gain optimization as described in the above paragraph and you should be set.

Slow Wags

“Slow” (<1Hz) wags can be caused by too low or too high common gain and as such, I suggest that you play with that first before messing with anything in the advanced parameter sheet. If the wag still exists, then this could point to a number of issues…Your tail servo, your tail servo arm ball distance, or your mechanical setup could all be at fault. These usually have to be dealt with on a situational basis and aren’t always easy to diagnose. On the mechanical setup side of things, recheck what we’ve already talked about, but this time if the tail linkage mechanics pass the test from #1 above, then move to on to the next step.


4. Advanced Mechanical Checkouts:

On the tail, check the tail output shaft and tail rotor hub…are they bent? It only takes a little bit to aggravate things when the tail is spinning at several thousand RPM. Is there any excessive play in the tail grips either in the radial (along the rotor hub shaft) or axial (along the tail output shaft) directions? Disassemble the tail rotor and make sure that you’ve installed the radial and thrust bearings and their associated spacers correctly.

If you have radial play, then you may need to resize one of both of the rotor hub shafts.See if you can match the grips to the correct shaft by finding the combination that gives you the least radial play. If you still have some play, you can very lightly and patiently sand the end of the rotor hub shaft that is longer than necessary until, when matched with the appropriate grip, you have a nice snug feel. Do not over-tighten the screws, as they’re too small to handle you kung-fu death grip. Degrease and apply a thread locking agent and then tighten them hand-tight. 

Next move on to the main rotor head and drivetrain. Worn dampers, bent shafts, bad bearings, an out-of-round main gear, over-tight gear mesh and other similar issues can manifest themselves as tail wags on flybarless systems, so be thorough. Of all the flybarless helis I’ve seen with these issues, I think I’ve only ever seen one of them with a head issue that was causing a tail wag and it was due to a slightly bent main shaft. By “slightly” I mean that the shaft had approximately .001″ runout…it doesn’t take much!

Once you’re satisfied that your advanced mechanics are in order, go back to #3 and tune your tail as described above. At this point, you should have a heli with a rock-solid tail and minimal if any tail wags. Keep in mind that nothing is perfect, so don’t kill yourself trying to get that last millimeter of wag out of your machine. I’ve flown helis before with extremely slight tail wags in a hover that hold like a rock and fly wonderfully! Depending on the flybarless system, it’s actually not as uncommon as you might think to see a very mild wag in a hover. It’s when the wag starts affecting your tail control and confidence in flight that you need to take notice and fix it.

 Good luck!

– Justin