Andy here is the link to the first of two videos i took yesterday... this one had 4 Cinestar dampers on a Hex. and dampers between cam mount and cam http://youtu.be/Zl4MZUty6CQ Ramesh
Ramesh, I have a few pieces of magnetic balansers tried out and NON of them was consistent, the props that I balansed on them came out worse than before the balancing. Try this, balans a prop and then turn the locking cone from the magnetic balanser on other positions on the prop hub and look at the results. Mario
Andy this is the second video of yesterday it has been taken with only the CineStar 4 dampers http://youtu.be/bHP9m2AkJGM i removed the dampers from between the cam and cam mount so it does look like the dampers between cam and cam mount were really not required But still cant figure out the shaking. And at least with the CineStar damper there is no jelloing effect. But is it really possible to get a smoother video from the mcopter. Ramesh
ok guys totally forgot.. my apologies my experiment with Futaba S Bus worked perfect to control Wookong M and Radians i am using the 8FG S bus this is the channel assignment and connections for Single pilot operation i put a S bus hub into Rx S bus port then took one output to Wooking M S bus port and the other line to first radian pin 1 ch1 to 4 to fly multicopter ch 5 - IOC - switch SB ch 6 - Fail Safe - switch SF ch 7 - WKM Mode - switch SE ch 8 ch 9 - Radian Mode - Switch SG ch 10 - Radian Tilt - Dial RD ch 11 - Radian Click - switch SD NOW. for two pilot operation, for the time being , till i lay my hands on Spectrum DX18 on my JR 11 radio, this is what i plan to do ch 9 - Radian Mode - 3 way sw assigned ch 10 - Radian tilt - trim lever Ch 11 - Radian click - 3 way switch ONCE i get the Spectrum DX18, i have the option to assign control sticks of Aileron and Elevator to any channel SO on the Futaba i will assign ch 12 to Pan AND now on DX 18 i will use the same ch of 9 to 12 that r on Futaba on DX 18 only now i can assign ch 10 Tilt to elevator stick and ch 12 which is Pan to aileron stick so the Radian channels will match on Futaba and DX 18 and there will be no conflict On 27th a friend of mine is getting a DX18, so will have a look at channel hardware reasignment options and then order a DX18 for the second pilot, How does that sound Ramesh
Andy just one doubt all my motors r mounted vertically upwards does it help if they r inclined inwards slights OR if the clockwise motors r inclined one way and the anticlock motors r inclined the other way. does this help in better stability of Multicopter or only for yaw purpose. cause i really dont need any more yaw power that what i am getting Ramesh
I can borrow a MK USB programer from a friend, is it possible to use this for programing the Radian's? and how do I do this? Mario
Keep us posted, please, Mario. I wish there was a way to do a loopback test with the Radian USB Adapter in circuit so that we could easily eliminate the data path from the Radian software out to the connector that plugs into the Radians --- then we could exonerate that whole data path. Andy.
Hi Ramesh: It seems to me that you're doing all the right things to figure out the cause. I wish there was something more that I could suggest -- it's the classic problem of tele-diagnosis -- namely that Brad, Gary, and I can only see, hear, and know a limited amount of data about the problem. The other barrier we face is that you're using a custom set up and so you're operating out of the envelope of our experience, I'm afraid. Right now my concerns are that the booms themselves might be flexing and turning the copter into more of a tuning fork than it should be. I would be tempted to strap the iPhone to the hub, set Vibrations up for a long sample, and then hover just above ground effect and hope that you can see the "shiver" happen and capture data about it's frequencies. I seem to recall that Vibrations allows you to do a Fourier transform so you can see what frequency components are present. I suspect you're talking about a frequency down around 3-7 Hz -- which, when you know that the motors are spinning orders of magnitudes faster than that has to tell you something about the beat frequency. I wish there was more that we (and I mean Brad, Gary and I) could do to help, but... Andy.
Ramesh looking at the videos and your posts about the motors here are my thoughts. If the motors without props are well balanced you should almost have flat lines on all three of the graphing axis. I don't have any motors with the amount of vibration that you are showing in your pics. All of the XY's look pretty good but not the Z axis. And I have found that using a single small piece of tape the size you have, i mount it to the motor, test and then move it a tape width counter clockwise, test and repeat all the way around the motor until I find a spot that is near perfect. Then I test moving the tape a 1/2 width either side of that spot. Some of the other approaches, like using a wire tie to me doesn't make sense. To much overall weight. We are talking about single digit grams of weight to get it in balance. If that doesn't do it then the motors are probably inferior since it shouldn't take that much to balance a good motor. My last batch of 9 motors only two needed to be balanced. And a single piece of tape did the trick. After all of the balancing of motors and props that leaves the overall frame and gimbal. Checking vibration on the bench will give a totally different result from actual flight since you don't have any flight loading on the copter. Looking at your last two videos and assuming that your props and motors were balanced then that leaves the frame/gimbal. Since it isn't a CineStar frame perhaps the frame and arms are flexing leading to overall vibrations that no amount of gimbal vibration dampers or balancing will be able to handle.
Gary: I share your observations about the Z axis. If I'm understanding the data correctly, that would be vibration parallel to the prop shaft and without the props on that's likely to be a function of the quality of the motors and their bearings, I suspect. Andy.
Andy and Gary thanks a ton for that discussion i am beginning to understand what u guys r says. As regards the Z axis vibration, i did wonder why it was highest in all the motors, cause it is the vertical axis along the motor shaft ... so it could be the quality for sure, and so the bearing becomes suspect The reason why i used the square arms, was cause i need it to fold without removing the arms totally, so i can send in into check-in baggage while travelling by air.. other wise i will have to send it a day in advance by cargo, with the round tube arms.. there would be too much of dismantelling and then assembling at the other end or would need a read big box to carry it safely and cargo handling here is bad they will even destroy a steel box hahahaha so looks like i have fresh problems on my hand to solve And i think i will do that balancing thing once more... i was moving the tape 90 deg at a time around the circumference each time. i think will do it 45 deg at a time. And next flight will be with the iphone strapped onto the machine to see the shiver the max acquisition time on the i phone is 52 sec so looks like i will have to land every minute and start again is there any way to make it 2 to 3 minutes And i really appreciate all the help. this distance learning is difficult, and even more diff to solve problems...but in its own way it is helping a lot in some way. oh,, any idea,, is there a folding Hex or Octa which has round arms or i could make the square arms i have that r 15x15 mm to 20x20 or even 25x25 looks like still loads of experiments lines up ahead Ramesh
Andy i may just have found the problem my arms go into the central body only by 40 mm and this may be the cause they should go in all the way in which would not be about 80mm and this would give the central structure more rigidity and more strength. and then less flexing cause in my hand i can see flexing and it should not be there. the image shows how little the arms go into the central body so work started on making the arms long enough to right inside all the way. Hopefully tomorrow i should have good news... i guess Ramesh
Ramesh is the hub made of standard fibreglass sheeting? If so I think you will find that there is to much flexibility in fibreglass. That's why you see so many copters where the bodies, arms, etc are made from carbon fibre. Stiffness and weight.
As Gary says, Ramesh, the rigidity of the hub/boom assembly is critical -- otherwise you have created a flying tuning fork that will vibrate like crazy at certain resonant frequencies. Andy.
Sorry for the delay in responding to this, but the answer is that the MK USB Adapter will not work correctly with the Radians. Thanks Andy.
Andy and Gary u guys hit the nail on the head i am now cutting longer arms so that they go in as deep as possible and my central body is 3mm fiberglass i will do just this one more change in length of arm so that they go in far deeper into the body, if this improves things,,, great... if not will order 3mm carbon sheets Ramesh
hi hahahaha.. .i think Andy was apt in saying i was flying a tuning fork... the more i think about it the more i am convinced that, that was exactly was i saw as was trying to find the correct word and was calling it "shiver " it was like when u get suddenly hit by a cold breeze and sends a shiver down ur body,,, lasts barely a second So now ready for a test flight Ramesh
sorry I did not see your reply earlier.. Actually I am using two Spektrums for two pilot ops. I describe it here Things you did not know about the Radian One of the benefits with having extra channels in the dx18 is the pilot is controlling the radian mode, not the camera man.
Mario: I exchanged email with JohnC. One thing to try, with the Radian USB adapter connected to your PC (but without necessarily being connected to a Radian sensor), when you click on Connect in the Radian Configuration software, you should see a Yellow LED flash briefly on the Radian USB Adapter. Do you (or did you) see that? Andy.