single rotor fpv
- Thread starter kloner
- Start date
Rob
What I do with SRH drive train is to look at each stage from just the motor to the final assy. with the I phone vibration App. The tail rotor can be a problem but I don't use a conventional drive from the main gearwheel. It is just a motor fixed to the tail rotor assy. running at constant speed. That speed can be varied to avoid a harmonic. You are right the main rotor is not the obvious problem. Quality blades from Spin are the answer. All this good stuff seems to come from Germany.
On the video without camera mount what I did was land and change the camera angle when I needed to look in a different direction. What this teaches you is that camera movements from a constant angle are more natural to the viewer than having too many changes. A balance of the two concepts is about right in my experience. This is just a bit of fun stuff.
https://vimeo.com/44222343
The GoPro is just stuck on with Double sided rubber tape. I have since cured the jellow with a different mounting point. That model is built for power line inspection etc. it has to have full rotor protection. .
What I do with SRH drive train is to look at each stage from just the motor to the final assy. with the I phone vibration App. The tail rotor can be a problem but I don't use a conventional drive from the main gearwheel. It is just a motor fixed to the tail rotor assy. running at constant speed. That speed can be varied to avoid a harmonic. You are right the main rotor is not the obvious problem. Quality blades from Spin are the answer. All this good stuff seems to come from Germany.
On the video without camera mount what I did was land and change the camera angle when I needed to look in a different direction. What this teaches you is that camera movements from a constant angle are more natural to the viewer than having too many changes. A balance of the two concepts is about right in my experience. This is just a bit of fun stuff.
https://vimeo.com/44222343
The GoPro is just stuck on with Double sided rubber tape. I have since cured the jellow with a different mounting point. That model is built for power line inspection etc. it has to have full rotor protection. .
Last edited by a moderator:
kloner
Aerial DP
all the vibrations we'd see were from that big main gear under the rotor to the drive of the tail. you got a motor spinning 14-18k rpm spinning main blades 2000 rpm then behind all the spinning the other way is a tail drive trying to get 7k-8k rpm. seperate motor would be huge. seen it in scale alot
R_Lefebvre
Arducopter Developer
So you guys both found that going to an independent tail motor helped a lot? Interesting. I had been thinking of trying that but I'm not sure how much motor I need. I've got an extra 35-36 910kV lying around. And a 28-30 880kV, but I doubt that is enough. Looks like the 910kV would give me 15,000rpm on 4S, too fast. 11,500 on 3S. Still a little fast but getting better. Could govern it to the proper RPM from there. The problem with all this is of course the CG. I previously had my rudder servo mounted all the way at the back, but just converted back to normal because it didn't give me the benefit I wanted, and I had to strap ballast on the nose to offset the weight! (this problem would go away when I finally get a camera up there...)
Maybe I'll work on something. Should be easy, the only hard part is getting the correct size/length shaft.
I have also thought about changing to a belt drive on the main shaft, and this would certainly help that! Makes everything simpler. But you can only get about 6:1 reduction using common industrial timing belts, unless you go to a complicated dual-stage. However, this also might be curable by going to a really low KV motor, like 480. On 8S that I run, that would get me ~2400 RPM according to Dark Horse. Which could be throttled back. At 3.4V/cell, it would give ~1900 rpm, so still enough headroom to throttle back. This could work.
I'll have to see if an XL belt is enough, as they are easily available in alumium/acetal construction pulleys.
kloner
Aerial DP
usually it is a torque tube and usually it is just a shaft going to a coupler going on the motors shaft, couple set screws and flat spots, away you go
The funny part is the tail motor is a power hog. most guys run a couple of similar sized motors that i've seen, kv and cell count to match a decent amp draw/rpm
The kmax or coaxial though,........
stock drive products has a grip of anything, any size, etc.
like this
The funny part is the tail motor is a power hog. most guys run a couple of similar sized motors that i've seen, kv and cell count to match a decent amp draw/rpm
The kmax or coaxial though,........
stock drive products has a grip of anything, any size, etc.
like this
Last edited by a moderator:
R_Lefebvre
Arducopter Developer
I'm having difficulty following what you're saying.
But I was thinking just a single plate bolted onto the tail end piece (the part that clamps on the end of the tube), and the tail motor mounted to that plate, and then the tail rotor bolted directly onto the motor's shaft. But now that I think about it, use the existing tail rotor shaft, with that shaft simply coupled to the motor's shaft. Piece of cake.
But I was thinking just a single plate bolted onto the tail end piece (the part that clamps on the end of the tube), and the tail motor mounted to that plate, and then the tail rotor bolted directly onto the motor's shaft. But now that I think about it, use the existing tail rotor shaft, with that shaft simply coupled to the motor's shaft. Piece of cake.
I don't have a photo because I just changed the motor but it just attaches to the final drive shaft at the tail with a right angle fixing. I am trying a much smaller higher KV motor now (2500KV) so that I can power it from a 2 cell battery that drives the servos. Main power comes from two 11 amp 6s Max-amps. The main rotor speed is only in the range of 1000-1250 rpm. At 2000 rpm it would used three times more power just to hover. I am trying to keep all rpm's on the low side. Boom length is longer than standard to allow for 10mm larger tail blades. I think that a asymmetric (anti torque ) tail boom profile could be next to reduce power consumption of tail rotor .
Last edited by a moderator:
usually it is a torque tube and usually it is just a shaft going to a coupler going on the motors shaft, couple set screws and flat spots, away you go
The funny part is the tail motor is a power hog. most guys run a couple of similar sized motors that i've seen, kv and cell count to match a decent amp draw/rpm
The kmax or coaxial though,........
stock drive products has a grip of anything, any size, etc.
like this
Hey Kloner - dont suspose you have a pic of this setup a bit further out / away at all ?
Thxs if you have and you could post it..
Cheers
Thanks for that DennyR, Will look into that..
Friend has a Intrepid Bergan - Driving him up the wall with its vibes..
Cheers
R_Lefebvre
Arducopter Developer
I was doing so more testing on Arducopter yesterday. I've got the tail holding pretty well, and did some high speed passes. Sorry for the fixed camera angle, I don't have a cameraman. The last pass would have been much better if the camera actually caught the perigee. Also seems 60fps would be better for the high speed stuff.
http://www.youtube.com/watch?v=XFVfx90rVro&feature=youtube_gdata
http://www.youtube.com/watch?v=XFVfx90rVro&feature=youtube_gdata
Last edited by a moderator:
R_Lefebvre
Arducopter Developer
What do you mean about the tail holding? I don't think I did any pitch-pumps in the other video I posted? If you just mean the little waggle in a hover, it's still doing it, I just didn't do much hovering in this video so you probably didn't see it.
As for the angles, again not quite following. It definitely has an "atti" mode, that's what I was flying in. Center stick gives you a flat attitude. Full forward pitch gives you 45° nose down. Full right gives you 45° right roll. It makes it very easy to fly.
I'm still having some trouble getting the heli to hold attitude, the vibration is still messing it up. But I made some progress yesterday. I think a large part of the problem is the AHRS system is tuned to work with multi-copters, and helis have a different vibration profile than helis, so the tuning needs to change.
kloner
Aerial DP
i was talking about the wag, looks alot less what little you did hover.
when you fly across the screen in fff, it is diving down, going up, never makes a straight line from left to right or right to left. That is the DJI claim to fame, in 2008 when they first started it all there is an article about dji showing up to an ap show and did a 100 yard straight line ff that supposedly made em noticed, no other system had done it so consitently. I love that in fpv, takes all the collective management out of the equation
your vibes are cause it's an hk. I've messed with alot of cheap models (see who i was r&d for) and i know the gears don't have alot of lash adjustment, but you want to make it adjustable if any gears aren't something like .004" lash. if any gear gets loose then tight then loose, then tight as it rotates, get a different brand gear or try another. most important like mentioned int his thread is that tail drive gear at the back of the main. it spins so many rpm and is usually of such low quality they barely get by on a good day vibe free. if you take all your blades off and spool it up and feel the frame, that is generaly alof of what your feeling if the motor pinion to main gear lash is dead on and not binding. Lash vibes are what you near when you block out the blade noise and motor whine. can really hear it in autorotations
when you fly across the screen in fff, it is diving down, going up, never makes a straight line from left to right or right to left. That is the DJI claim to fame, in 2008 when they first started it all there is an article about dji showing up to an ap show and did a 100 yard straight line ff that supposedly made em noticed, no other system had done it so consitently. I love that in fpv, takes all the collective management out of the equation
your vibes are cause it's an hk. I've messed with alot of cheap models (see who i was r&d for) and i know the gears don't have alot of lash adjustment, but you want to make it adjustable if any gears aren't something like .004" lash. if any gear gets loose then tight then loose, then tight as it rotates, get a different brand gear or try another. most important like mentioned int his thread is that tail drive gear at the back of the main. it spins so many rpm and is usually of such low quality they barely get by on a good day vibe free. if you take all your blades off and spool it up and feel the frame, that is generaly alof of what your feeling if the motor pinion to main gear lash is dead on and not binding. Lash vibes are what you near when you block out the blade noise and motor whine. can really hear it in autorotations
R_Lefebvre
Arducopter Developer
Ok, now I understand what you're saying about the angles. I was in Stab mode, which only controls the attitude, but not the thrust/altitude. There is an Alt_Hold mode, which is the same, but also controls altitude. I could have used that but just didn't. I should do some test flying in that mode to see how well it works. To be honest, I'm still learning to fly a heli, and I want to make sure I know how to control it the old fashioned way without totally relying on the system. I have flown airplanes for years and am pretty good, but the transition to heli is a challenge. The one really low pass was much lower than I wanted, but I have to rewire my brain. I was pulling back on the "elevator", but it was still losing altitude. In an airplane you just keep pulling back. The heli, I should have increased thrust. On an airplane, if you're aimed at the ground and don't want to be, the last thing you would do is throttle up! 
I came within about 2 feet of the ground there. One of those "Heh, I meant to do that" moments!
Now, something else that I want to do is create a rudder control mode called "weathervane" where it intentionally controls the rudder to have the tail follow the ground-speed heading of the heli. This would simplify things and allow you to fly it more like an airplane. Put it in Alt_Hold and just cruise around. It would be really nice for FPV, or shooting a video where you're following a moving target.
Yes, it's slowly becoming less and less HK, and I'm getting out of that completely. It was necessary because I crashed a lot while developing the code. But now that things are working better, it's causing more problems. I am building up a heli that is basically a 550 stretched to 600, will probably go to 700 eventually. I want the lighter weight of the 550 frame with direct to swash servos. It's being built with all high quality stuff. Tarot 2mm frames, KDE parts, Heli Option, RJ head, etc. It'll be a bit heavy because everything is aluminum and thick CF, no plastic. But I'm trying to build something that is "industrial" using COTS components. At least as much as I can within my budget.
The tail drive shaft is actually a genuine Align already. And I've used their main gears but honestly didn't find them any better than the HK. I've been thinking of using this machined gear, what do you think?
http://www.kdedirect.com/TREX700MDGM07.html
I totally get what you're saying though. It's really obvious. This was a fresh, full rebuild, and the first flight was fine. Then the vibes got worse and worse. I think it's the gear mesh as you say, and I probably already blew the tail drive bearings. I suspect I'm putting too much tension on the belt because I'm going through these things really fast. Doesn't help that they're cheap bearings.
Still, I think that the system should function properly if the heli reasonable, which it is. I can fly it in full manual no problem at all. So if the system causes a crash that didn't need to happen, it sort of defeats the purpose.
I came within about 2 feet of the ground there. One of those "Heh, I meant to do that" moments!Now, something else that I want to do is create a rudder control mode called "weathervane" where it intentionally controls the rudder to have the tail follow the ground-speed heading of the heli. This would simplify things and allow you to fly it more like an airplane. Put it in Alt_Hold and just cruise around. It would be really nice for FPV, or shooting a video where you're following a moving target.
Yes, it's slowly becoming less and less HK, and I'm getting out of that completely. It was necessary because I crashed a lot while developing the code. But now that things are working better, it's causing more problems. I am building up a heli that is basically a 550 stretched to 600, will probably go to 700 eventually. I want the lighter weight of the 550 frame with direct to swash servos. It's being built with all high quality stuff. Tarot 2mm frames, KDE parts, Heli Option, RJ head, etc. It'll be a bit heavy because everything is aluminum and thick CF, no plastic. But I'm trying to build something that is "industrial" using COTS components. At least as much as I can within my budget.
The tail drive shaft is actually a genuine Align already. And I've used their main gears but honestly didn't find them any better than the HK. I've been thinking of using this machined gear, what do you think?
http://www.kdedirect.com/TREX700MDGM07.html
I totally get what you're saying though. It's really obvious. This was a fresh, full rebuild, and the first flight was fine. Then the vibes got worse and worse. I think it's the gear mesh as you say, and I probably already blew the tail drive bearings. I suspect I'm putting too much tension on the belt because I'm going through these things really fast. Doesn't help that they're cheap bearings.
Still, I think that the system should function properly if the heli reasonable, which it is. I can fly it in full manual no problem at all. So if the system causes a crash that didn't need to happen, it sort of defeats the purpose.
kloner
Aerial DP
the kde gears are good and usually cnc instead of cast so there perfect, no cooling/warping problems. but even better is the helical cut gears. align has em for everything
http://www.helifreak.com/showthread.php?t=409776
http://www.helifreak.com/showthread.php?t=409776
R_Lefebvre
Arducopter Developer
Yes, I'm just looking into this now. The problem is the 700 size helical main gear seems to be 112 teeth, resulting in too low of a ration and too high head speed. I need more ratio... Researching now. I need to solve this problem. (That's a great thread, thanks!)
Also, no helical cut tail drive gears? I'm using a belt on the tail. Not a fan of shaft drive.
Also, no helical cut tail drive gears? I'm using a belt on the tail. Not a fan of shaft drive.
R_Lefebvre
Arducopter Developer
I had bought the HobbyKing main gear cluster for the HK550GT, sort of just for SnG, but also I was hoping the quality was better, since most of the HK550 parts are better than the 600 parts. Nope. It's worse! It's an absolute joke. I think the run-out is close to 1/4" I'm not even sure the motor could turn it all the way around without losing mesh! Unreal.
Now I'm researching "slant" or helical gearing. The slant gearing is much quieter, and smoother than straight cut gears. Either a genuine Align parts, or the KDE gearing. I'm a little nervous about the rumor that slant gearing will generate axial thrust (well that part is not a rumor, it's true) which will then destroy the motor bearings. So I did a calculation today, got the motor power, calculated the torque at that speed, then the tangential force on the pinion. (9.6lbs at hover, 41.4lbs at full load) Then assuming the helix angle is 5°, the axial load would be 0.83lbs and 3.6lbs.
Either of those numbers seem pretty trivial for the motor bearings to take? If this motor were used in an airplane, the thrust loads from the motor would be quite a bit bigger than this. So the idea that the bearings will fail... I dunno, seems like bunk to me?
Now I'm researching "slant" or helical gearing. The slant gearing is much quieter, and smoother than straight cut gears. Either a genuine Align parts, or the KDE gearing. I'm a little nervous about the rumor that slant gearing will generate axial thrust (well that part is not a rumor, it's true) which will then destroy the motor bearings. So I did a calculation today, got the motor power, calculated the torque at that speed, then the tangential force on the pinion. (9.6lbs at hover, 41.4lbs at full load) Then assuming the helix angle is 5°, the axial load would be 0.83lbs and 3.6lbs.
Either of those numbers seem pretty trivial for the motor bearings to take? If this motor were used in an airplane, the thrust loads from the motor would be quite a bit bigger than this. So the idea that the bearings will fail... I dunno, seems like bunk to me?