Scale 500 sensitive to wind

[Andy01]

I am still struggling a bit with my UH-1N 500 size. The heli is quite big for a 500 (1150mm long &nearly 8lbs), and I tried flying it today at both 1900rpm and 2000rpm.

Initially the body was wiggling around under the rotor disc badly with no stick input, and I did;

Replaced dampers with soft grey ones
Made endless changes to gains etc
Tried headspeeds from 1900rpm to 2200rpm
Fitted a set of Spin asymmetrical 435mm wide chord blades.

I have now got it to the stage where in calm conditions it actually hovers and does circuits quite well, but if there is a light wind gust (around 12-15mph) it will wiggle around like a restless puppy, both hovering & FF.

I have attached the latest setup file & a recent log.

I would appreciate some ideas on what I can do to tame the wiggle.

Thanks
Colin

[georgi UK]

=] Colin hi, well it's getting somewhere at least! the vibe log is showing some spikes at the head, any play in the head bloc?

I really would drop the servo travels to 100%, then increase the servo ball distance to bring the swash mix down and closer to 60% (did we mention this last time?)

there is also high rates of acceleration showing in the log? I suspect this unit may have an issue? mail the log to SK for their opinion and keep us posted, please!

[Andy01]

Georgi

Thanks for the response. The head is a RJX FBL head, and with the soft grey Align dampers there is a bit of movement in the grips & feathering shaft assembly, but the head block is firm on the main shaft. When I first tried it, it had some KBDD orange dampers fitted, and the head was almost rigid.

The first vibe plot in your post above shows a small spike at roughly 1x headspeed, and the larger spike at 2x headspeed - is this a blade pass frequency or perhaps perhaps too much "teetering" in the head because of the soft damping ? The large at about 28000 rpm is the motor speed - it is a KDE 500 motor.

I could try to find some bigger servo arms, but the balls are currently in the outer hole - they are Futaba 9650 servos. I will see what I can find.

Perhaps a dumb question - could I increase servo travels to close to 150%, reduce cyclic swash mixes and collective mix to achieve the same overall pitch ranges, to achieve the same effect as longer servo arms ?

I will send a couple of logs to Art and see what he says.

If I try to swap a 720 from one of the larger Helis, any suggestions on what tape to use ? The black stuff that came with the latest 720 was hopeless. I have the 720 in the 500 mounted upside down and it just fell off under its own weight overnight. I peeled the black tape off undamaged and it really wasn't very sticky at all. I found some of the grey tape that came with my first 2 720s and used the last piece - it is impossible to remove in one piece. Would a double layer of standard yellow colored 3M double sided tape (from hardware shop) be flexible enough to damp out vibes ?

Cheers
Colin

[georgi UK]

=] Colin, yes the blades could be flapping! increasing the servo travels would amount to the same thing, ie, nothing mechanically will have changed. I mostly use align PU gel with a velcro strap. is it 3m 4011 that seems trendy these days?

[Luvmyhelis]

Originally Posted by gmcallister

=] Colin, yes the blades could be flapping! increasing the servo travels would amount to the same thing, ie, nothing mechanically will have changed. I mostly use align PU gel with a velcro strap. is it 3m 4011 that seems trendy these days?

Also in an upside down mount apply the thin 3m double sided sticky to the PU gel. The gyro will not fall off then. I still use a loosely fitted velcro safety strap just as insurance.

[Andy01]

Well I sent a bunch of logs etc to Art on Sunday (his time) but have not had a response yet. I hope he got it.

Colin

[Andy01]

Guys

An update on this issue.

I sent the logs etc to Art about 3 times and other from the initial responses, I have not heard from him since October.

I have now done about 60 flights/hovers and tried a VERY wide range of different settings - gains (Bell, cyclic, tail etc) from ±35 to 70, dampening settings from 13 to 22, collective & tail mixes from 5 to 25, headspeeds from 1800rpm to 2200rpm, 3 different main blades (most recently 465mm Spin asymmetrical), different tail blades (84mm plastic & 95mm CF).

Most importantly I tried a completely different SK720 from my 600 sized Seahawk (it work fine on the Seahawk).

I have re-done the head (added more shims with the soft grey dampers), and re-done the setup on the second SK720.

The symptoms I am seeing is the cyclic sensitivity to wind (same shudder / wobble) and a strange sideways shimmy in FF, even in calm conditions.

The shimmy is odd - I will be flying scale circuits doing a long pass (about 200 yards) and usually a couple of times it will wag the tail, or that it what it looks like, although it may be a sideways aileron shimmy. It only lasts a couple of seconds, and then goes for another 50-100 yards. It often seems to happen just before or a bit after a banked turn.

Any help would be appreciated.

I was going to upload the logs as .zip files and the setup file, but it now says these are invalid files - has this changed ?

Thanks
Colin

[georgi UK]

=] Colin, good to hear back from you, sometimes these threads dry up without knowing what the outcome was?

zip files are now invalid because of a trojan some time back, these guys are good for what we need in here = http://www.filedropper.com/ free, don't need to sign up or anything!

ok, the current state of play? are the servo's powerful enough for this type of loading, I wonder?......

[Andy01]

Georgi

Thanks for the response.

I am using Futaba s9650 cyclic servos, which are fairly standard fair for 500s. They are rated for 4.5kg (63oz) each which I would have been enough for a 3.5kg scale heli.

Obviously this heli is a fair bit heavier than a stock 500 pod and boom, but the rotor disc should be more lightly loaded than a full 3D 500 flown hard.

I have seen several scale 500s using these servos. I have just looked on Servo database and could find any correctly sized mini servos with more than 64oz.

Cheers
Colin

[georgi UK]

=] check the servo plugs use a gold and not silver connector, there were some fakes going around. my 500 tips the scales at 1.9 kg with 4.6 kg servo's.

it's not just the weight, the surface area is considerably more here, servos sprung to mind earlier, another log might help?

[Andy01]

Georgi

I will try to share the logs & setup using filedropper thanks - hope it works.

Vibe log
<a href=http://www.filedropper.com/log221><img src=http://www.filedropper.com/download_button.png width=127 height=145 border=0/></a><br /><div style=font-size:9px;font-family:Arial, Helvetica, sans-serif;width:127px;font-color:#44a854;> <a href=http://www.filedropper.com >file storage online</a></div>

Playback log
<a href=http://www.filedropper.com/log227><img src=http://www.filedropper.com/download_button.png width=127 height=145 border=0/></a><br /><div style=font-size:9px;font-family:Arial, Helvetica, sans-serif;width:127px;font-color:#44a854;> <a href=http://www.filedropper.com >upload files free</a></div>

Setup file
<a href=http://www.filedropper.com/uh-1nsetup161212><img src=http://www.filedropper.com/download_button.png width=127 height=145 border=0/></a><br /><div style=font-size:9px;font-family:Arial, Helvetica, sans-serif;width:127px;font-color:#44a854;> <a href=http://www.filedropper.com >file storage</a></div>

Do you know of any mini sized cyclic servos with more torque ? I have read threads here with a few guys flying this same model, but with different FBL unit, and no reported problems. Not sure what servos they use - there has been no activity on the threads for about 6-7 months now.

From memory the servos have gold plugs (I don't remember any of my servos having silver plugs) and were bought from a reputable shop, but it is a good tip, so I will check it again.

Cheers
Colin

Not sure that the "imbed" code has worked - here are the 3 links in the same order;

http://www.filedropper.com/log221
http://www.filedropper.com/log227
http://www.filedropper.com/uh-1nsetup161212

[georgi UK]

=] Colin hi, I take your point that others fly this heli well without your issues, the mental picture here is a 550 size machine using mini-servo's!

in calm conditions it seems fine, but with added wind loads it rebels is how I read it? hence the servo's came to mind. I have blown out 450 servo's which were under powered.

into the logs then, is the motor ok? looks like a bearing is going south.



are you running a 1900 headspeed? I think there is a lead / lag vibe going on the head, have you tried FBL blades?



the servo's are working quite hard but I have seen worse! the unit could be more firmly fitted, what is your interface here?



on your settings it's possible the servo ball needs to out a hole, the gyro gain and hiller gain look too high? go from here Colin, it's interesting! georgi.

[Andy01]

Georgi

Thanks. The heli is quite long for a 500 (angled tail boom, and pointy snout), but not as porky as a 550 scale heli. It looks small compared with my 600 sized Seahawk.

The vibes around 340-350Hz are around 22-23000rpm, which is around 1x motor speed (22.2V x 1450kV x 0.81 = 26000rpm). I did think that this may be a very slight balance issue with the motor because usually bearing fault frequencies are higher than 1x rotational speed. It was a brand new KDE 500XF motor (never been crashed) which is a solidly built motor with double bearings at the pinion end. It is not impossible that it has a problem, and in fact, Patrick (the owner of KDE) offered to swap it out but I had to post my motor back to the US etc, which I was trying to avoid if possible. Do you think that these vibes would be enough to interfere with the 720 ? They didn't seem high enough for that ?

I think I was running 1850rpm as set and measured with my Ice 75 ESC. I have noticed that the 720 seems to indicate a slightly higher speed. I should have mentioned before that after the early response from Art, I did try running the Ice 75 BEC at 6.5V, and if anything that seemed to make the cyclic shudder worse. I then tried running the 720 off a 2s A123 (I use these in my two larger 720 equipped scale helis), and it did not make any noticeable difference.

I have not tried FBL blades, but I have tried Mavrikk 430mm, Spin 435mm Asym & 465 Spin Asym, all of which are quality perfectly balanced blades. My "head" vibes seemed to be a bit worse previously with a noticeable 2x headspeed spike. I fitted an extra thin shim (from Atom 500) on each side of the headblock over the spindle shaft to tighten up the head a little (while still using the softer grey dampers), and it seems to have reduced the 2x vibe a bit to a point where it is not as pronounced a spike anymore, although your snapshot shows the 2x spike at about 3700rpm (as far as I can tell). It still seems to pop up intermittently, whereas before it was almost continuously the highest spike on the vibe plot.

I tried fitting the 720 using the Skookum supplied black tape, and it was completely useless - it actually fell off overnight. I then refitted the original 720 using a spare piece of Skookum supplied grey tape (from a year or two ago), and it was OK. When I refitted the second 720, I used some thin exterior tape and sandwiched a piece of thicker soft foam tape in between (as seen in this forum), and it gives a result very similar to the Skookum grey tape.

So, it is 720 - thin tape (Skookum) - metal plate - thick tape sandwich - solid 6mm aluminium gyro mounting plate on heli. It is mounted upside down without any velcro etc. I have removed any extra wires (like LCD interface etc) to minimise transferred vibrations.

My current servo arms (Futaba) are on the outside hole, so I could try to look for longer arms. By gyro gain, are you referring to the overall (locked) cyclic gain ? I have tried this from 35 to 70 without much difference, but perhaps not with the current combination of other settings (that is the only problem with these advanced FBL units - there is so many variables that there is millions of permutations of settings).

What would you suggest for gain and Hiller gain ? I have not adjusted Hiller gain too much previously, so I can certainly try that.

I assume that the additional plots you see are only available with the Pro viewer ? I have never seen anything more than the "sim-looking" flight in Playback mode.

Thanks again
Colin

[S76GUY]

Try dropping the head speed to 1700 and turn the bell hiller gain down 5 %.
Leave the gray soft dampers in the head.
Replace the gyro tape with the Gray Futaba tape if you have any or use some white
thick foam tape to see the results.
Are the mechanics secured at the dog house? with no side to side movement?
Tony T.

[npomeroy]

I always use Kyosho Zeal vibration mounting gel, which is green and about 5mm thick, with a metal plate attached to the SK directly with double sided carpet-mounting tape, the idea being to increase the inertial mass of the gyro unit and not to let the metal mass move independently of the gyro. Although mine are not inverted I usually arrange a rubber band to ensure the SK stays in place.

[georgi UK]

=] Colin, I just posted up what mainly caught my eye within the logs, the motor vibe was worth a mention, it certainly isn't affecting the 720, yes I was talking about the locked gain. the pro viewer enables deeper layers to be seen.

the software permutations are endless, indeed! how it should work is the SK has an out-the-box nicely tuned control loop, when we see things causing flight issues it's usually the odd setting that needs tweaking, or something could be out in the setup, the order for me on your heli would be to re-establish a known datum,

have a look at the servo arm distances as mentioned, right click on a cyclic bank to reset back to default scale settings, not sure how you approach the collective and cyclic throws for scale ships?, but I would setup for fully 3D and adjust the Tx pitch curves down to how you want it. then start with a locked gyro gain of 45%

keep going!

[npomeroy]

Hi Colin; I defer to Georgi on the software issues. However it may be worth reporting that I had an intermittent "wobble" with a 5 bladed 450 MD500, especially (but not exclusively) when a manouver interrupted smooth FF. It seems to have gone away by reducing cyclic gain. My thinking was that the inertia of the multi head reduced the heli's speed of response so the SK overcorrected when perturbed. You did say you had tested most software variables though.
The other thing I pay attention to is securing the fuse to the mechanics near the top to prevent sideways wobbles. I have used dense foam that either pressed firmly against the frame or some arrangement involving the original canopy securing posts.

[Andy01]

Thanks guys.

So, the summary of the above may be to try a "locked" cyclic gain of 45 or lower (I have been down there, but perhaps not with the current combination of settings.

And, I am a bit confused by S76GUY's suggestion with lowering the Bell Hiller gain to 5% - there is Bell gain, which my SK720 has self-tuned up to around 40-46, and there is Hiller Gain, which I have currently set at 55%. Is there another setting for a combination ?

I would have thought that reducing the Hiller gain to 5% may tend to make it unstable (like a FBL heli without a FBL controller) - is this correct ? The little mouse-over pop-up says "Flybar action, stabilises the pitch or roll action".

I can certainly try to reduce the Bell gain to 5%, but I would need to turn off the self tuning function to achieve this. Is this a good move ?

The mechanics in this UH-1N is a custom job supplied with the fuselage so that it all fits into the doghouse, and leaves the entire rear cabin open. This means that it is very short (the autorotation gear is only a few mm above the bottom plate of the mechanics), and there is not much to attach it too as the doghouse is attached with magnets. It is probably not as stiff as I would like, but there is not much I can do about it.

Georgi
I basically set the heli for a pitch range of ideally around 15-16° (total pitch range), but often can't get it much lower than 20° without going too low on the collective swash mix setting, so I reduce with the Tx pitch curves. I then set up the swash to blade grip links (and some minor pitch curve tweaking) to give me a range of -3° at low stick, around 5° at mid stick, and around 10-12° at high stick.

For cyclic pitch I usually try to get as close to the recommended Skookum ±10° as possible, which is why on this build I am sitting at a swash mix of 65 for both elevator and aileron. Arguably, I may not need this much movement for gentle scale flying, but the Skookum recommendation (and I think yours when I got my first 720) was ±10°, so I aim for that.

Hopefully I have explained it properly ?

Cheers
Colin

[georgi UK]

=] Colin, Don't turn anything down to 5%! reduce by 5% is what was being said, turn off self tune bell is fine.

[Andy01]

My bad

I will give it a try when I get out next.

Thanks
Colin