Just checked my motors speed...

[NRJ]

You probably do this anyway, but if not, hold the stethoscope against one of the actual motor screws below to get the truest motor sound.

No, actually, I didn't think of that. See, more golden nuggets of experience. I'll try that. Thanks.

 

[flyNfrank]

WetDog and flyNfrank, I was able to get the DatCon and convert the files and get them in Dashware. I am attaching a pdf picture of part of the Dashware files that have the Motor information. Can you take a look at it and let me know if you need more information or if this is enough, Does it tell you anything? Thanks a lot.

HOLY~~~~ I can't believe I missed this! I'm so impressed that you have went this far in such a short amount of time. This is so cool that you have taken it to the point of learning to use dashware which is a bonus. I learned to use that program with out asking for anyones help. That meant I spent an unbelievable amount of time doing hands and trial & error type learning. I still have some to learn but currently waiting to see if any upgrades will takes place on the next version. A GoPro employee owns dashware and developed it before becoming employed by GoPro. It is in need of some serious upgrading now.

As far as your question above, I would really rather look at that same file using CsvView on my PC. You should be able to upload in this thread when creating a post.

 

[NRJ]

HOLY~~~~ I can't believe I missed this! I'm so impressed that you have went this far in such a short amount of time. This is so cool that you have taken it to the point of learning to use dashware which is a bonus. I learned to use that program with out asking for anyones help. That meant I spent an unbelievable amount of time doing hands and trial & error type learning. I still have some to learn but currently waiting to see if any upgrades will takes place on the next version. A GoPro employee owns dashware and developed it before becoming employed by GoPro. It is in need of some serious upgrading now.

As far as your question above, I would really rather look at that same file using CsvView on my PC. You should be able to upload in this thread when creating a post.

Hahaha, yes I just am not too good at reading the results of the guages. I do have the CsvView also. I would however like to take it out for a better flight and then get the dat files, convert them and upload them for you if you can help in the readings. Are you able to read the guages if I upload the files?

Actually I guess I could give you the same files. Do you want the dat file or the csv file? I can't upload the csv unless I change the suffix (.csv) to (.dat). Or I can upload the dat file. Let me know.

Also on post #30 BudWalker uploaded the MotorLoad.zip file that can also be used with the CsvView. You may already have this. If not, check it out. Thanks.

 

[flyNfrank]

The dat pretty much requires the use of dropbox. A csv can be uploaded in here.

 

[NRJ]

The dat pretty much requires the use of dropbox. A csv can be uploaded in here.

Here is the link for dropbox. Dropbox - FLY613.csv

Hope it works for you. Thanks.

 

[Numone]

FWIW, forget the Back Right motor running "much slower", cos your Back Left motor is running much higher than the others at 2948 RPM! I have to agree with FNF, go and fly the darn thing and enjoy it.
I noticed early on that the props run at slightly different speeds at idle.
NRJ..... You have waaaaay too much time on your hands and this non-existent problem will drive you mad matey.

 

[NRJ]

FWIW, forget the Back Right motor running "much slower", cos your Back Left motor is running much higher than the others at 2948 RPM! I have to agree with FNF, go and fly the darn thing and enjoy it.
I noticed early on that the props run at slightly different speeds at idle.
NRJ..... You have waaaaay too much time in your hands and this non-existent problem will drive you mad matey.

Nahhh. I love this hobby. It gives me a lot to do and also to learn. What other hobby can you take up where you learn about aerial photography, Aircraft flying and maintenance, map topography, latitude/longitude, AC telemetry, Remote control, Antenna's range and particulars of Wifi channels, the list goes on. It and this motor situation keeps me busy. My hands are busy with my time on my hands. Thanks for the observation though. Happy flying.

 

[Numone]

Nahhh. I love this hobby. It gives me a lot to do and also to learn. What other hobby can you take up where you learn about aerial photography, Aircraft flying and maintenance, map topography, latitude/longitude, AC telemetry, Remote control, Antenna's range and particulars of Wifi channels, the list goes on. It and this motor situation keeps me busy. My hands are busy with my time on my hands. Thanks for the observation though. Happy flying.

LOL! Fair enough.

 

[NRJ]

Last question I have before I actually "Do" something. Lol, I'm sure you guys are tired of this, but here goes.
My purpose for starting this post is to learn if I should replace my noisy motor. From that I wanted to be sure of what motor to replace.

I have attached a spreadsheet where I averaged the motor speeds and motor loads over 21,122 tick marks from the dat file from the AC. The graph gives a picture of what is going on within a 12 minute 20 second flight. I hear an unusual noise in my Left Back (LB) motor. All else seems fine. In this post it has been said that the important factors are; motor speed, motor loads. The interesting thing about the graph is:

1. The noisy motor is the LB at an average speed of 3327 RPM (the highest speed) with a load of 251.5.
2. The RF motor has the highest load at an average speed of 2664 RPM with a load of 418.5.

I know the ESC controls everything causing the motors to compensate wherever needed. But, I'm just curious if this graph says anything out of the ordinary to anyone of you. If not, I understand and will resort to my only three options.

Option 1. Do nothing and keep flying. (My least favorite.)
Option 2. Replace the noisy motor. (Leaning this way.)
Option 3. Replace all the motors. (Undecided at this time.)

I still haven't tested the motors with the infrared thermometer, so this may help in my decision when I receive it.

Thanks for any thoughts on the graph(attached) information.

 

[Dounin Front]

Last question I have before I actually "Do" something. Lol, I'm sure you guys are tired of this, but here goes.
My purpose for starting this post is to learn if I should replace my noisy motor. From that I wanted to be sure of what motor to replace.

I have attached a spreadsheet where I averaged the motor speeds and motor loads over 21,122 tick marks from the dat file from the AC. The graph gives a picture of what is going on within a 12 minute 20 second flight. I hear an unusual noise in my Left Back (LB) motor. All else seems fine. In this post it has been said that the important factors are; motor speed, motor loads. The interesting thing about the graph is:

1. The noisy motor is the LB at an average speed of 3327 RPM (the highest speed) with a load of 251.5.
2. The RF motor has the highest load at an average speed of 2664 RPM with a load of 418.5.

I know the ESC controls everything causing the motors to compensate wherever needed. But, I'm just curious if this graph says anything out of the ordinary to anyone of you. If not, I understand and will resort to my only three options.

Option 1. Do nothing and keep flying. (My least favorite.)
Option 2. Replace the noisy motor. (Leaning this way.)
Option 3. Replace all the motors. (Undecided at this time.)

I still haven't tested the motors with the infrared thermometer, so this may help in my decision when I receive it.

Thanks for any thoughts on the graph(attached) information.

This is fun - let's add to your misery.
Looks like RB is working very efficiently.
What if you performed another IMU calibration, but this time you "jacked" one of the corners to change the bias before you calibrate? Try it at several points and see how it affects the numbers. It's something I've wondered anyway. (Maybe you should get a new level (digital)).

 

[Mocho]

NRJ, don't forget the Phantoms motors are not sensored. So they will not control motor speed as well as a sensored motor/esc combination! Your deviation could me totally normal, especially at lower speeds.

Hey just another thought

 

[NRJ]

NRJ, don't forget the Phantoms motors are not sensored. So they will not control motor speed as well as a sensored motor/esc combination! Your deviation could me totally normal, especially at lower speeds.

Hey just another thought

Interesting info. I'm kind of getting the same thought from others out there. It may be true. It may be that I just have to replace the one noisy motor and that there is no electronic way to make that determination. Thanks for your input.

 

[N017RW]

NRJ, don't forget the Phantoms motors are not sensored. So they will not control motor speed as well as a sensored motor/esc combination! Your deviation could me totally normal, especially at lower speeds.

Hey just another thought

What sensors are you referring to?

One of the critical functions of an ESC is to commutate.

With brushed motors this is done with brushes. Located on the stator, the brushes contact the commutator, these dictate motor coil phasing and wear and often arc creating electrical and frictional loses. Eventually they fail resulting in motor failure.
Brushless motors however are void of these hence their name.
BLDC Commutation is controlled by the ESC sensing [back] EMF voltage generated in the coils by the permanent permanent magnets in the rotor. This provides the necessary coil energizing timing that brushes did.
Without some type of rotational sensory feedback a BLDC motor could not work.

 

[Mocho]

What sensors are you referring to?

One of the critical functions of an ESC is to commutate.

With brushed motors this is done with brushes. Located on the stator, the brushes contact the commutator, these dictate motor coil phasing and wear and often arc creating electrical and frictional loses. Eventually they fail resulting in motor failure.
Brushless motors however are void of these hence their name.
BLDC Commutation is controlled by the ESC sensing [back] EMF voltage generated in the coils by the permanent permanent magnets in the rotor. This provides the necessary coil energizing timing that brushes did.
Without some type of rotational sensory feedback a BLDC motor could not work.

Basically they add some electronics to the motor and send a signal back to the ESC for better control of speed, especially at slower speeds.

Sensored Motors uses the "Hall Effect" (named after Edwin Hall, an American Physics scientist) to detect the rotor (magnetic poles with different polarity) inside the motor and transmit the data to the ESC through a sensor wire. It allows the ESC to control the rotation and the speed of the motors precisely. In general, it is smoother compared to a Sensorless motor with the wire connected, and the signals are sent and received correctly. It is noticeable that Sensored motors are smoother especially in the lower end of the motors operations.

 

[David_Cambridge]

This is a fascinating discussion. My approach would be to find a way to attach an oscilloscope to the pulse wire to the motor. I am fairly sure that the motor speed is controlled by a pulse stream to it. The speed would be changed either by the pulse widths, or the pulse rate. In either case, the scope pix could be used to correlate with the rpm. I suspect that this would be the quickest way to tell what's going on. Current flow has nothing to do with the motor speed.

Let us know what you discover if you decide to try it.


Sent from my iPad using PhantomPilots

 

[NRJ]

This is a fascinating discussion. My approach would be to find a way to attach an oscilloscope to the pulse wire to the motor. I am fairly sure that the motor speed is controlled by a pulse stream to it. The speed would be changed either by the pulse widths, or the pulse rate. In either case, the scope pix could be used to correlate with the rpm. I suspect that this would be the quickest way to tell what's going on. Current flow has nothing to do with the motor speed.

Let us know what you discover if you decide to try it.


Sent from my iPad using PhantomPilots

It sounds like a fine idea indeed. However, I'm not adept enough at electronics to attempt such a feat and I don't have an oscilloscope. The best I have been able to do is determine that there is definitely an abnormal noise in the Left Back motor. There is however a couple of charts of the motor speeds and motor loads from the aircraft dat file. It is interesting, but I am unable to understand the meaning of those files. I will attach them below. Possibly you will have some understanding of them. If not, that is alright too. Thanks.

 

[David_Cambridge]

The difficulty in getting data from the raw speed, it that the electronic speed controller is changing the speed many times per second under normal conditions. If you look at the motor loading graph, which is the green one I think, the motor loading is the lowest. This would indicate that there is nothing dragging the motor down more than any of the other motors. Are the props on or off? If they are on, it would probably mean that the prop has slightly less air drag.

I don't see anything abnormal in these traces. The sound files are not definitive. I'm with the others in thinking that if the ac flys normally, with no problems with hover or tracking, that the systems are ok, for the times that that is true. The shift in sound is concerning, although I gouldn't get anything from the sound clips, that told me anything. If the noise happened during the time the traces were plotted, I would expect any drag on the motor from damage to show up as a sharp increase in the motor loading, as the ESC worked to keep the motor turning at its required instantaneous speed. To be safe, you could replace the noisy motor, although I see no indication that it is required from the data available.

 

[David_Cambridge]

A further note on the tach readings. You can't draw any conclusions about variations in rotational speed differences, other than they are occurring. The ESC appears to be ordering them, as a previous poster has asserted. I would give up on it being a meaningful diagnostic tool if the bird flys correctly.

 

[NRJ]

The difficulty in getting data from the raw speed, it that the electronic speed controller is changing the speed many times per second under normal conditions. If you look at the motor loading graph, which is the green one I think, the motor loading is the lowest. This would indicate that there is nothing dragging the motor down more than any of the other motors. Are the props on or off? If they are on, it would probably mean that the prop has slightly less air drag.

I don't see anything abnormal in these traces. The sound files are not definitive. I'm with the others in thinking that if the ac flys normally, with no problems with hover or tracking, that the systems are ok, for the times that that is true. The shift in sound is concerning, although I gouldn't get anything from the sound clips, that told me anything. If the noise happened during the time the traces were plotted, I would expect any drag on the motor from damage to show up as a sharp increase in the motor loading, as the ESC worked to keep the motor turning at its required instantaneous speed. To be safe, you could replace the noisy motor, although I see no indication that it is required from the data available.

Your analysis seems straight forward and logical. I think you are right about the load being the most definitive item. The two graphs I provided was motor load the top graph with the colors for each motor, the blue color being the load on the noisy motor. The other graph on the bottom was the motor speeds for each motor and again blue being for the noisy motor the LBack. But even when I average the data from the dat file (30,651 ticks) of speed and the loads for each of the motors there does not seem to show any overly large amount of load on the motors. You are right, I don't have good recording equipment especially since I was using a mechanics stethoscope to get the sound file. But, using it directly in my ear, I definitely hear a continuous hit on what I can only imagine might be a bearing. Hah! I don't even know if this has bearings. And listening to the other 3 motors that are smooth in their sounds tells me something is wrong and possibly a pending problem which I so desire to avoid.

To answer your question, I had noticed the noise in a motor while my drone comes in for a landing and also when I have it hover. That is what started all of this. Yes I was in flight for all of this data which was a 17 minute flight. Below is the flight path I took. I was actually trying to run a battery down to 8-10%. Then are the graphs again which if you look at the blue color on the load of the top graph, that is the noisy motor. Then I have included my averaging graph of 30,651 ticks from the dat file showing the noisy motor. This is a more complete graph as the previous data on a previous graph only had 21,122 ticks (data points) in it. This is an average of each motors speed and load. The left back is the noisy motor.

The blue "MotorSpeed:LBack" in blue is the noisy motor in quesiton above.

I like what you said. It is logical and reasonable. Do you think using a infrared thermometer on each motor will help my analysis? Determining if one motor is hotter than another makes sense especially if the noisy motor is hotter? Thanks for you help and analysis.

 

[David_Cambridge]

I'm sorry. I thought that your issue was with the slowest motor. The blue line is the fastest motor. That is also one of the most heaviest loaded. That could mean a bearing in it's very earliest stages. It could also be something about the prop, since you mostly notice it when flying. For some reason the ESC want's to spin the prop faster than the others, which would suggest some prop abnormality. What do you hear with no props?

If you move the props around, does it follow the prop? What about trying a new prop?

The other thing to consider, is that is the motor going fastest. It will probably sound different for that reason.