HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Dec 08, 2012, 02:37 AM
Registered User
BigJimNZ's Avatar
Auckland New Zealand
Joined Sep 2010
1,296 Posts
Quote:
Originally Posted by jbalat View Post
Yep much better in the wind. Just tried it in 25 knots, had to keep the nose down about 45deg to hover in the one spot and it would only glitch a little when gusts came through.

Anyone else using smaller props with dt750's ?
Only went to 10 x 4.7 when my arms were under 360mm. Flew nice.
Think there's a comparison video on my blog here at RCG.
BigJimNZ is offline Find More Posts by BigJimNZ
Reply With Quote
Sign up now
to remove ads between posts
Old Dec 10, 2012, 05:31 PM
Registered User
Joined Feb 2011
706 Posts
Quote:
Originally Posted by BigJimNZ View Post
Only went to 10 x 4.7 when my arms were under 360mm. Flew nice.
Think there's a comparison video on my blog here at RCG.
Well before everyone goes out to buy some 9x4.7's I tried them on my H-Copter which is a lot heavier and it wouldnt even get off the ground
jbalat is online now Find More Posts by jbalat
Reply With Quote
Old Dec 11, 2012, 06:54 AM
Flying a Chipmunk in Portugal
jhsa's Avatar
NRW, Germany
Joined Sep 2010
3,404 Posts
My quad has 9x5 props and weights 1300g. It flies ok. well it used to...

Joo
jhsa is online now Find More Posts by jhsa
Reply With Quote
Old Dec 11, 2012, 11:37 AM
Registered User
BigJimNZ's Avatar
Auckland New Zealand
Joined Sep 2010
1,296 Posts
Quote:
Originally Posted by jbalat View Post
Well before everyone goes out to buy some 9x4.7's I tried them on my H-Copter which is a lot heavier and it wouldnt even get off the ground
Yeah but that was asking a bit John. 5kg is a bit heavy hehe ;-)
BigJimNZ is offline Find More Posts by BigJimNZ
Reply With Quote
Old Dec 14, 2012, 06:29 AM
An ordinary sort of incredible
mr.incredible's Avatar
United States, TN, Knoxville
Joined Dec 2004
931 Posts
Quote:
Originally Posted by jbalat View Post
Well before everyone goes out to buy some 9x4.7's I tried them on my H-Copter which is a lot heavier and it wouldnt even get off the ground
My all up weight is a little over 1200g, I had the same problem with anything less than 10x4.7s. I can fly with 9x4.7s but the motors are running a lot faster and noisier. I get really good punch with 10x4.7s and it runs quieter.
mr.incredible is offline Find More Posts by mr.incredible
Reply With Quote
Old Dec 14, 2012, 08:26 AM
Fly, crash, rebuild.
United States, NY, Shelter Island
Joined May 2011
877 Posts
Quote:
Originally Posted by jhsa View Post
My quad has 9x5 props and weights 1300g. It flies ok. well it used to...

Joo
interesting read. I have the RCexplorer tri with 500mm arms that works well on 11x4.7" props but I put them on my Talon Tri with 221mm arms and she's all over the place and twitchy in a bad way. 10x6x3 and 9x5x3 GWS were buttery smooth on the 221mm talon. I have 380mm and 320mm arms that I'm going to try and was hoping to go to 13-15" props with a low KV disk motor setup.
My talon weighs 850g w/o battery with 30A esc and NTM28-26 1000kv motors.
9x5x3 on 3S if anemic and on 4S its great.
Thanks for sharing that insight.
H
shelterock is offline Find More Posts by shelterock
Reply With Quote
Old Dec 14, 2012, 11:23 AM
Registered User
BigJimNZ's Avatar
Auckland New Zealand
Joined Sep 2010
1,296 Posts
Same tricopter. Arms about 360mm. One vid with 10 x 4.7 the other 11 x 4.7

Tricopter WiiTC10x47.AVI (1 min 0 sec)


Wii Tricopter11x47.AVI (0 min 33 sec)
BigJimNZ is offline Find More Posts by BigJimNZ
Reply With Quote
Old Dec 15, 2012, 06:17 AM
Lets make it remote control!!!
abletowinginc's Avatar
USA, FL, St Augustine
Joined Jun 2009
998 Posts
big difference on yours jim
abletowinginc is offline Find More Posts by abletowinginc
Reply With Quote
Old Dec 15, 2012, 06:19 AM
Lets make it remote control!!!
abletowinginc's Avatar
USA, FL, St Augustine
Joined Jun 2009
998 Posts
i currently run 11 inch on my t copter and tri but i am going to try 10 inch on the tri its light enough the t copter is going to be to big and heavy for 10 inch props. ill postt results when i try it on the tri
abletowinginc is offline Find More Posts by abletowinginc
Reply With Quote
Old Dec 17, 2012, 09:20 AM
Registered User
United States, MD, Ellicott City
Joined Feb 2010
191 Posts
Tricopter Masters Thesis

I just finished my masters thesis on a tricopter and figured the vibrational analysis my be of interest to some of you. Here is a small bit of it with the pictures and graphs on a PDF:

A test frame was constructed to dynamically isolate a motor. A vibration measuring device measured installed at the base of the motor measured vibrations through the whole throttle range. The vibration measuring device was made for an analog 3 channel data logger and a 3 axis accelerometer, both powered by a mini LiPo. Information was uploaded from the data logger to a computer for data analysis. The data provided was in Gs, which corresponds to one acceleration of gravity or 32.2 ft/s^2. By multiplying the mass of the motor by the accelerations, force is obtained. Since the mass of the logger was equal to the mass of the motors propeller adapter, the test was conducted without the propeller adapter without a need for correction of the mass of the measuring device in the analysis.

To ensure the motor was properly dynamically isolate in the test frame, a second vibration measuring device was used to observe that no vibrations from the motor was dissipated through the frame. The second vibration measuring device was an iphone 4 which contains a three axis gyro and accelerometer and with a vibration application that can measure, display, and log vibrations the phone experiences. A second data logger was used in the ESC to record voltage, RPMs, and throttle as the motor was run without a propeller through a range of throttle positions. Figure 14 shows the test frame used and the data logger.

Four motors were tested as described in the above method. The resultant accelerations from the X, Y, and Z axis of were combined and plotted against the % throttle which produced the graph in Figure 15.

The graph in Figure 15 shows the results plotted with a moving average trend line. The graph shows that the highest vibration spikes occurred at the 57% and 80-100% throttle positions. This information is useful to determine the proper throttle ranges to operating in a hover condition which may require minimized vibrations and to ensure that the RPMs related to the throttle position do not coincide with the natural frequency of the candidate materials to be used for the rotor arms.

Since the test was conducted with no propeller, ie a no load condition, the RPMs were used to correctly identify the occurrence of the vibration spikes. Different propellers on the motor will increase the loading while decreasing the RPMs. For example the no load max RPMs were approximately 12000 RPMs, while loaded the max was 8350 RPMs both corresponding to 100% throttle. Since the propellers are dynamically balanced, it is assumed little to no vibrations were caused by the propellers, the vibration spikes observed in the no load condition would occur at the same RPMs when loaded.

The relationship between motor voltage and RPMs is linear. With the operator transmitter having the ability to use programmable curves, the throttle programming normally used in RC helicopter applications was used as shown in Figure 16.

From Figures 15 and 16 it is determined that motor vibration spikes occur at 11200 RPMs and 12000 RPMs and above. These values of RPMs are outside the operating range or flight envelope of the loaded motor profile. Any vibration spikes observed on the frame will be a factor of the natural frequency and the length of the rotor arms for each of the candidate materials.
The natural frequency is a function of the rotor arm length and material properties of each of the candidate materials. To model the rotor arm analytically a vibration beam analysis was completed. The dimensions of the rotor arm were best modeled as a cantilever beam fixed on one end and the other free, as seen in Figure 17. The mass of the motor was neglected since it provides an upward force from the propellers. Torque was also neglected since it is considered to be small when compared to the thrust.

Figure 18 shows the area of constraint of the flight envelope, which is the operating RPM range of the loaded motor turning a 10 X 8 propeller. In this box it is undesirable to have the natural frequency of the material selected for the rotor arm to occur, which will cause excessive vibrations in flight. The rotor arm length was limited to 1 meter based on the availability of the material. The graph indicates that the favorable length and material for the rotor arm was 0.45 meters and carbon fiber. Aluminum was also acceptable for a length of .38 meters and wood .26 meters. At the lower end of the flight envelope the values were ignored since the corresponding lengths required would make the rotor arms easier to break.

The results of the analysis was tested by taking vibration measurements of a 0.45 meter length beam on its fixed end while running the motor, loaded with a propeller, at various throttle positions on the free end. The data from this experiment was used on conjunction with the data in Figure 15 to develop a vibration transmissibility profile for each of the candidate materials. A ratio was taken of the accelerations observed on the clamp end over the accelerations from the motor. A ratio less than one indicated that the material was absorbing the vibrations from the motor, where as a ratio greater than one indicated that the vibrations from the motor were being amplified by the material. The results of this test are on the graph in Figure 19. With the propeller being dynamically balance and beam properly clamped down, no extra vibrations appeared to come from the propeller except when running near 100% throttle and when excess vibrations were detected in the beam, in which the propeller seemed to amplify the effect. This was due to propeller blade flections from maximum loading on the blades while being hold stationary. A moving average curve was placed on the carbon fiber profile to better highlight the results.

As predicted from the beam analysis of the given length of 0.45 meters, the carbon fiber rotor arm was the best material. At a throttle range greater then 85% the carbon fiber began vibrating more to possible flections in the propeller blade as more thrust was produced. The vibrations can also correspond to the upper limits set in the beam analysis. The wood and aluminum rotor arms appeared to excessive vibrations through the throttle range greater than 60%.
MIDNCOCO is offline Find More Posts by MIDNCOCO
Reply With Quote
Old Dec 17, 2012, 09:25 AM
Registered User
United States, MD, Ellicott City
Joined Feb 2010
191 Posts
In short I measured the no load vibrations, figured out my loaded(with a prop) thrust curve, found my flight envelope, used beam theory to plot natural frequency vs rotor arm length for wood, AL, and CF, found my rotor arm length, and then tested it.

I used this for a tri on turnigy 1100 kv motor.
MIDNCOCO is offline Find More Posts by MIDNCOCO
Reply With Quote
Old Dec 17, 2012, 01:51 PM
Lets make it remote control!!!
abletowinginc's Avatar
USA, FL, St Augustine
Joined Jun 2009
998 Posts
very interesting and useful findings.... surprising even on the fact that wood didnt help reduce vibrations as well as carbon since the woods grains go in many different directions seems it would have helped reduce vibrations but you proved otherwise very interesting good work






Quote:
Originally Posted by MIDNCOCO View Post
I just finished my masters thesis on a tricopter and figured the vibrational analysis my be of interest to some of you. Here is a small bit of it with the pictures and graphs on a PDF:

A test frame was constructed to dynamically isolate a motor. A vibration measuring device measured installed at the base of the motor measured vibrations through the whole throttle range. The vibration measuring device was made for an analog 3 channel data logger and a 3 axis accelerometer, both powered by a mini LiPo. Information was uploaded from the data logger to a computer for data analysis. The data provided was in Gs, which corresponds to one acceleration of gravity or 32.2 ft/s^2. By multiplying the mass of the motor by the accelerations, force is obtained. Since the mass of the logger was equal to the mass of the motors propeller adapter, the test was conducted without the propeller adapter without a need for correction of the mass of the measuring device in the analysis.

To ensure the motor was properly dynamically isolate in the test frame, a second vibration measuring device was used to observe that no vibrations from the motor was dissipated through the frame. The second vibration measuring device was an iphone 4 which contains a three axis gyro and accelerometer and with a vibration application that can measure, display, and log vibrations the phone experiences. A second data logger was used in the ESC to record voltage, RPMs, and throttle as the motor was run without a propeller through a range of throttle positions. Figure 14 shows the test frame used and the data logger.

Four motors were tested as described in the above method. The resultant accelerations from the X, Y, and Z axis of were combined and plotted against the % throttle which produced the graph in Figure 15.

The graph in Figure 15 shows the results plotted with a moving average trend line. The graph shows that the highest vibration spikes occurred at the 57% and 80-100% throttle positions. This information is useful to determine the proper throttle ranges to operating in a hover condition which may require minimized vibrations and to ensure that the RPMs related to the throttle position do not coincide with the natural frequency of the candidate materials to be used for the rotor arms.

Since the test was conducted with no propeller, ie a no load condition, the RPMs were used to correctly identify the occurrence of the vibration spikes. Different propellers on the motor will increase the loading while decreasing the RPMs. For example the no load max RPMs were approximately 12000 RPMs, while loaded the max was 8350 RPMs both corresponding to 100% throttle. Since the propellers are dynamically balanced, it is assumed little to no vibrations were caused by the propellers, the vibration spikes observed in the no load condition would occur at the same RPMs when loaded.

The relationship between motor voltage and RPMs is linear. With the operator transmitter having the ability to use programmable curves, the throttle programming normally used in RC helicopter applications was used as shown in Figure 16.

From Figures 15 and 16 it is determined that motor vibration spikes occur at 11200 RPMs and 12000 RPMs and above. These values of RPMs are outside the operating range or flight envelope of the loaded motor profile. Any vibration spikes observed on the frame will be a factor of the natural frequency and the length of the rotor arms for each of the candidate materials.
The natural frequency is a function of the rotor arm length and material properties of each of the candidate materials. To model the rotor arm analytically a vibration beam analysis was completed. The dimensions of the rotor arm were best modeled as a cantilever beam fixed on one end and the other free, as seen in Figure 17. The mass of the motor was neglected since it provides an upward force from the propellers. Torque was also neglected since it is considered to be small when compared to the thrust.

Figure 18 shows the area of constraint of the flight envelope, which is the operating RPM range of the loaded motor turning a 10 X 8 propeller. In this box it is undesirable to have the natural frequency of the material selected for the rotor arm to occur, which will cause excessive vibrations in flight. The rotor arm length was limited to 1 meter based on the availability of the material. The graph indicates that the favorable length and material for the rotor arm was 0.45 meters and carbon fiber. Aluminum was also acceptable for a length of .38 meters and wood .26 meters. At the lower end of the flight envelope the values were ignored since the corresponding lengths required would make the rotor arms easier to break.

The results of the analysis was tested by taking vibration measurements of a 0.45 meter length beam on its fixed end while running the motor, loaded with a propeller, at various throttle positions on the free end. The data from this experiment was used on conjunction with the data in Figure 15 to develop a vibration transmissibility profile for each of the candidate materials. A ratio was taken of the accelerations observed on the clamp end over the accelerations from the motor. A ratio less than one indicated that the material was absorbing the vibrations from the motor, where as a ratio greater than one indicated that the vibrations from the motor were being amplified by the material. The results of this test are on the graph in Figure 19. With the propeller being dynamically balance and beam properly clamped down, no extra vibrations appeared to come from the propeller except when running near 100% throttle and when excess vibrations were detected in the beam, in which the propeller seemed to amplify the effect. This was due to propeller blade flections from maximum loading on the blades while being hold stationary. A moving average curve was placed on the carbon fiber profile to better highlight the results.

As predicted from the beam analysis of the given length of 0.45 meters, the carbon fiber rotor arm was the best material. At a throttle range greater then 85% the carbon fiber began vibrating more to possible flections in the propeller blade as more thrust was produced. The vibrations can also correspond to the upper limits set in the beam analysis. The wood and aluminum rotor arms appeared to excessive vibrations through the throttle range greater than 60%.
abletowinginc is offline Find More Posts by abletowinginc
Reply With Quote
Old Dec 17, 2012, 01:51 PM
Lets make it remote control!!!
abletowinginc's Avatar
USA, FL, St Augustine
Joined Jun 2009
998 Posts
i would like to see the isolation mount you made and used for this if youhave pictures




Quote:
Originally Posted by MIDNCOCO View Post
In short I measured the no load vibrations, figured out my loaded(with a prop) thrust curve, found my flight envelope, used beam theory to plot natural frequency vs rotor arm length for wood, AL, and CF, found my rotor arm length, and then tested it.

I used this for a tri on turnigy 1100 kv motor.
abletowinginc is offline Find More Posts by abletowinginc
Reply With Quote
Old Dec 17, 2012, 02:00 PM
Registered User
United States, MD, Ellicott City
Joined Feb 2010
191 Posts
Quote:
Originally Posted by abletowinginc View Post
i would like to see the isolation mount you made and used for this if youhave pictures
The mount is in the PDF, I made a square frame and used bugee cords to suspenend the motor in the middle of the frame. The cords would also disipate vibration transmission to the frame. The cords were attached to the screw loops on the motor mounting hardware. I tested the four motors and made an average fit on the graph. The three motors on the tri and one backup. I used a program on my iphone which converts it to a vibration meter and placed it on the frame to make sure the cords disipated the the vibs
MIDNCOCO is offline Find More Posts by MIDNCOCO
Reply With Quote
Old Dec 17, 2012, 02:03 PM
Registered User
BigJimNZ's Avatar
Auckland New Zealand
Joined Sep 2010
1,296 Posts
Quote:
Originally Posted by MIDNCOCO View Post
In short I measured the no load vibrations, figured out my loaded(with a prop) thrust curve, found my flight envelope, used beam theory to plot natural frequency vs rotor arm length for wood, AL, and CF, found my rotor arm length, and then tested it.

I used this for a tri on turnigy 1100 kv motor.
Great insight. Thanks for posting. Confirms what we all suspected.
Funnily enough I have arrived at arms of 0.46 metres for my tricopter through trial and error.
I clamp my motors to the arms with a silicon isolator. That helped mitigate the resonance.
Sparklet has been using lead on his craft to get the same effect for years ;-)
I going to start on designing the wheel next lol

Ian
BigJimNZ is offline Find More Posts by BigJimNZ
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Yet another tricopter build. !!Updated parts list!! SebastianJ Multirotor Talk 2669 Dec 30, 2011 05:48 PM
Discussion What to do? Failed tricopter build! Razor_Racer Coaxial Helicopters 9 Mar 07, 2011 09:20 AM
Discussion Tricopter Build Razor_Racer Coaxial Helicopters 21 Feb 20, 2010 08:24 PM
Build Log Tricopter build SebastianJ VTOLs 1 Dec 22, 2008 04:34 PM
Discussion Trying to build a mechanically actuated tricopter, Help! stefalo Coaxial Helicopters 3 Jan 09, 2008 04:10 PM