Twinn Rexx Yaw Axis Heading Hold Mode

[joe@tech-mp]

I started this thread so we can keep all relevant info on this topic in one place.

We need to get HH mode on the yaw axis working by using a gyro or maybe some other approach not thought of yet. It's going to require experimentation and ingenuity. So put your thinking caps on ;)

There seems to be mixed results in the experiments done so far with gyros. More experiments are needed.....


Joe

www.techmodelproducts.com

[joe@tech-mp]

GY401 - Experiments

After researching with the search engines for HH gyro theory info and ending up unsatisfied.
I decided to try an experiment with my XL which has a GY401 and S9650 for yaw control. I mounted it to a bench on a turn-table and removed the main rotor blades. The goal was to characterize the gyro's operation.

My XL main rotor spins CW. The gyro gain was set to 50% HH mode with a delay of zero. Power set to 25%. I noted the following observations:

1. With no TX input, the gyro let's you rotate it CW.

2. With no TX input, the gyro fights you when trying to rotate it CCW.

3. With a CW TX input, The gyro fights to get there when restrained.

4. With a CCW TX input, The gyro remains idle (no fight) when restrained.

The same four tests were done with the gyro delay set to 100 %.

The results were the same except that the commands from the TX were delayed a little (slowed down).

From the tests above, it's obvious that the gyro operation is not symmetrical which makes sense for a single rotor heli.
It has much more control power in the direction that is fighting torque.

Here are a few questions that come to mind:

1. Does anyone make a gyro that works symmetrically ?

2. What if we bias a small yaw force mechanically to mimic the main rotor torque (so it fights). Will this help the gyro work better?

Feel free to chime in. Ideas are needed !



Joe

www.techmodelproducts.com

[chinookmark]

Why would the 401 not work symetrically? Interesting idea that it is working asymetrically, but that doesn't make as much sense when you consider that you can mount the gyro upright or inverted.

Maybe a test is in order on CCW rotating rotor blades?

Have you considered a Logictech 2100T? The Logictech needs to be mechanically trimmed for a no-drift hover in rate mode for it to work right (while the GY401 is setup for center of travel on tailshaft), so maybe it works more symetrically?

[Gra55h0pper]

Interesting experiments Joe! One thing I don't understand though. If the 401 works asymmetrically then it should only work (well) for either CW or CCW rotors, right? However I believe there are many scale helicopters where the main rotor rotates CCW (to mimic the full-size) and (I think) they're using 401's also. Also, the gyro can be installed right-sight-up or upside-down without any problems. If it works asymmetrically, then this shouldn't be possible, right?

You mention you removed the blades. This means you still had a considerable mass spinning CW. My guess would be that the gyro somehow is adaptive in that it senses that it needs to "actively fight" in one direction and only needs to "passively give" in another. I'm not convinced at this point it's asymmetrical.

[joe@tech-mp]

I think you missed the point. Or, I didn't explain it so well. CW or CCW dosen't matter. Of course the gyro works the same for either. It has a direction switch.

My observations show it works different for fighting torque and has much more control power in that direction. Try it !


Joe

www.techmodelproducts.com

[chinookmark]

I see what you're saying now. That still sounds wierd. With the rotors stopped, the gyro will fully deflect the servo either direction when manually turning the helicopter. Why would this be different with torque applied?

Also, assuming that the tail has equal pitch available in both directions, and some pitch travel is used up fighting torque, the model should have MORE available turning power in the direction of the torque. This would make for a much higher piro rate in one direction than the other. It's been a while since I used a 401. Does it piro faster in one direction? Or is it pretty much the same in both directions?

[Gra55h0pper]

Joe -- Even though you're right in that I hadn't taken the gyro direction switch into account yet, I'm still not completely sure from my side. If you would make the tail rotor spin the other way, you would have to change the direction of the gyro where the main rotor (i.e. torque direction) is still rotating the same way, right? Although I've never actually tried it, I'm pretty sure a 401 still would work fine in this case.

[joe@tech-mp]

So far, it seems to me that the GY401 is optimized for operation with single rotor helis. Again, this makes perfect sense. This is what I would expect after thinking about it more and from the observations.

The question I have is how can we get symmetrical operation. That is, the gyro fights the same when a force (external) acts in either direction. IMO ,this is what's needed for a tandem due to it's torque canceling configuration.

I have heard of special gyros made for airplanes by Futaba that stabilize roll and pitch etc. Not sure if they are just rate only devices.

We need to dig deeper.......



Joe

www.techmodelproducts.com

[v22chap]

Joe
It does make sense that a heli gyro would be designed to work better one way than the other . Although futaba's AVCS is suppose to just work to maintain straight flight either way the wind blows it .. I am wondering if the bench test are maybe misleading as I didn't notice any problem with the flying test I did with the 401 ... but then I was in the garage in a controlled atmosphere too and was starting to see the pitching glitch problem which I was watching more than the yaw controll . ;)

Anyway I have a GYA 351 ... which is futaba's airplane ail. gyro ,,which says it has the AVCS ... which is their HH .
I also have the ACT Europe airplane gyro and it is HH also .. I use these two on my V-22 before as they are great for dual control DCP on it as they both have dual input / output and reversing controls ... the ACT works the best of the two and has limit learn capibilities ... and it is also the only one still made the futaba ones are not made any more ... so I will work mostly with the ACT .
I will try to do some experimenting with these for ya .. but time right now is really tight and I still haven't found my glitch problem on the tandem .
Larry

[joe@tech-mp]

Larry,

The GYA350 may be our ticket to HH. I will try to get one ASAP and characterize it.

It's like a GY401 but for airplanes. I hope this means it operates symmetrically.

Here's the manual for it http://manuals.hobbico.com/fut/gya350-manual.pdf



Joe

www.techmodelproducts.com

[v22chap]

Joe
If you want ,, I will send both my 351 (about the same as the 350 only it has the capability of two servo ail control on it also instead of just one )and ACT gyros to you as you have the time and the test bed setup ... and that way you don't have to search for a dealer and don't have to pay for something that may not be used again.
Just say the word and they will be in the mail Monday morn.:thumbup:
Larry

[joe@tech-mp]

Thanks Larry. You have a PM. ;)


Joe

www.techmodelproducts.com

[v22chap]

You got it ,, both in the mail monday with instruction manuals. ;)

[Gra55h0pper]

Call me a non-believer (I also still believe the earth is flat... :D), but I'm still not convinced a 401 is asymmetrical. It certainly makes a lot of sense to me that a 401 is optimized for the dynamics of a tail rotor helicopter (it's marketed as such). It also makes sense that a gyro could "learn" the direction of the torque and then apply different techniques for "fighting" vs. "giving". However, if the 401 is asymmetrical from the get-go, I'd expect Futaba to publish which direction it's optimized to work in (put an arrow on it or something).

Joe - can you try your experiment again but now with the tail rotor on your XL spinning the other way and flipping the gyro's direction switch? As the torque direction is unchanged, according to the theory the gyro would then work suboptimal, right? :confused:
I guess my point is that although the 401 has a direction switch, this still doesn't tell it what the torque direction of the main rotor is and therefore in which direction is for "fighting" and which one is for "giving". :confused: :confused: :confused:

That said, I'm very hopeful on the GYA350! I think the Twinn-Rexx has in common with an airplane that it takes much more time for an input on the rudder channel to result in a heading change than it does for a tail-rotor heli which is super responsive on the tail (actually so responsive that without a gyro the heli is pretty much unflyable). The GYA350 may very well be "more patient" waiting for results of inputs applied (even more than the delay setting on the 401).

I'll try to post a video of my "401 HH oscillations" later today. I'll need a break in the weather though as the forecast is for rain pretty much all day... :(

[joe@tech-mp]

Gra55h0pper,

You will have to convince yourself with this one. Seeing is believing I guess ;)

Also, I don't follow your logic with the "direction" or the direction switch. See attached.

Moving on.....


Joe

www.techmodelproducts.com

[chinookmark]

I think I get what Gra55h0pper is saying ... but correct me if I'm wrong. The direction switch might not just flip clockwise to counterclockwise -- it might just turn the entire algorithm upside down.

If that doesn't make sense, I hope this does: Imagine a screwdriver pointed down. Lefty-loosey, turning CCW on the helicopter. Now flip that screwdriver upside down. Left is still loose, but now turning clockwise on the helicopter. Now maybe that reversing switch actually switches lefty-loosey to righty-tighty, but I think there's a possibility that it flips the screwdriver instead.

I think the only way to say for sure is to test with reverse blade rotation. Easy enough, since the Twinn Rexx already has a reversed one-way bearing, all you would need to do is swap main gears between that and the Trex.

[Gra55h0pper]

I'm not completely sure what I'm saying myself but something in me tells me that we we keep going west we'll fall of the earth... :lol:

Let me try to explain my confusion again. If the 401 were asymmetrical, then -for optimal performance- it would need to know whether it operates a CW or a CCW rotor. The direction switch doesn't give it this information as I can have a CW rotor with the direction switch one way or the other way depending on things like the rotation-direction of the tail rotor, the position of the tail rotor (left or right side), the orientation of the gyro (rightside up or upside down) etc.

Note that the GYA350 also has a direction switch to tell the gyro which input to give for left yaw and which one for right yaw, just like any gyro needs to know. For some gyros you have a direction switch where for others (e.g. the G190) you need flip them in case the direction is reversed.

Anyway, I'd be happy to be proven wrong but let's move on. This thread is about getting a gyro to work in HH and I think the GYA350 may very well be able to do so!

[chinookmark]

Joe, I'm not getting the same results on my RCT-G730 gyro. In rate mode it fights in both directions, and in HH mode, it fights and returns in both directions. With TX inputs, it does both slow and fast turns and snap stops in both directions. This is with 50% gain HH, 65% rate. I have only tested with clockwise rotor rotation and gyro mounted upright.

RCT-G730 gyro: http://www.rc-tek.com/us_store/rc-tek-rct-g730-digital-gyro-p1798.html

This gyro is supposed to use updated technology from the GY401. How updated, I don't know, but it is signifigantly shorter and lighter. I fashioned a makeshift turntable out of cardboard and the rollers from a microwave oven turntable.

[joe@tech-mp]

chinookmark,

Was this test done on the Twinn or a tail-rotor bird ?

My tests showed the GY401 only fights in one direction and relies on the torque moment to move it the other way. This works well for a single rotor heli with a tail-rotor.

For the Twinn, we need the gyro to fight both ways in HH mode.


Joe

www.techmodelproducts.com

[chinookmark]

Sorry, it was on my Diablo (Trex style helicopter). I put my Trex back together tonight, so I will try out the Logictech 2100T tomorrow. Just out of curiosity, have you tested more than one GY401, to rule out a bum gyro?

[tungym]

Forgive me if I get it wrong.

Do you reset the GY401 neutral by flipping the HH/NOR mode switch on the transmitter. According to the manual, it states it is necessary to do once by flippping the switch at least 3 times in a cople of seconds. This command the GY401 to reset the NOR/HH neutral position, so as the neutral to be consistent in the 2 modes.

[billyd]

It almost sounds like you would need two gyros for yaw (one set to normal and one set to reverse) and an onboard computer to process the data coming from them and send the resulting output to the linkages.

[v22chap]

you can according to the manual make it nuetral in both that way ... but most have found it best to get it neutral in normal mode by hovering and landing and adjusting linkages ...leaving the trim lever in neutral position ... then when you go to hh it will be in neutral position automactically ... and easier than trying to fly and flip the switch 3 X's fast and not crash ..:wink: :o

[v22chap]

Joe
Gyros and manuals in the USPS priority today :thumbup:
Larry

[Intrepid175]

I've read though this with interest and would like to add my 2 cents worth.

The issues stated with the 401 don't make sense to me. I have a hard time believing the gyro needs to "fight against" torque in any way. The system is designed to sense movement around a given axis and then apply a counter command to stop that motion. As has been stated before, the gyro doesn't care what direction the main rotor is turning. The reversing switch is simply there allow for different control configurations and nothing more. Rotor rotation direction is irrelevant and I seriously doubt it's reversing the control algorithms. It's just reversing the servo output. In heading hold mode, the gyro will do whatever it needs to in order to comply with the operators command. If that means going hard over on the tail rotor, that's what it will do.

Now, how that's going to work while dealing with the mixer system required on the Twinn, I don't know. Some of the previous posts show a setting of 50% on the gyro sensitivity. For a standard tail rotor machine, that would ususally be close to a maximum setting from my experience. Anything more will usually have the tail wagging itself all over the place. With the Twinn, I would wonder if you couldn't go significanty higher than that. Since it's not going to respond as quickly, I would think the Twinn would tolerate a higher gyro sensitivity setting.

FWIW!

Fly Safe,
Steve R.