2x Motor mount screws per arm: Design issue?

Archive
1

As Craig3DR stated in this post:
“On March 15th our motor supplier changed the motor base and now when the set screws for the legs are installed, the motor is pushed away from the base of the arm preventing the screws from fully engaging the motor. Customers have been informed and longer screws have been sent. You can also correct the issue by loosening the set screw on the legs and tightening the motor screws.”

As a result we have seen numerous people in this forum as well as over at DIYDrones, where the motor disintegrated from the arm, resulting in crashes of their IRIS as well as substantial damage.

Reading through Craig’s text would make one assume that IRIS manufactured before the given date shouldn’t have an issue.

Even though my IRIS was manufactured in February, I went ahead and swapped the screws. And I recommend to everyone to do the same. Here’s is why:

I went ahead measured the various dimensions of the arm and motor mount and here is what I found:
[attachment=0]3DR_IRIS_Motor_Screws.png[/attachment]

  • The motor mount screw has to go through approximately 6.5 mm of plastic arm before it touches the metal of the A2830 motor. With the standard M3x8 screw that was used with IRIS this means that the screw is only inside the thread by about 1.5 mm.
  • As the thread leads into the magnetic coils of the engines, the inserted screw needs to stay short of these coils. Thus the screws need to be shorter than 12 mm. (An M3x12 screw would touch the coils and therefore can’t be used).
  • You can see that with a M3x10 screw, about 3.5 mm of the screw are inserted into the motor thread. As there are no M3x11 screws, that’s the best you can do.

After taking measurement of the motor mount on my IRIS, I have to come to the conclusion that using an M3x8 screw for IRIS is a design mishap - even for quads build before March 15th. M3x10 screws should have been used from the beginning.

While this mishap is something that can easily be fixed with a longer screw, the next mishap is something that cannot be fixed as easily:
As this document clearly shows, the IRIS motor is only mounted to the arms by 2 screws. Yet, as the technical specs for the used A2830 motor show, there are 4 threads that can be used for mounting the motor to an arm or frame.

While the technical spec. doesn’t state how many screws should be used, keep in mind that the entire weight of the IRIS Quad including any additional payload rests on these motor mount screws. Quick ascents or longterm vibrations will place additional stress on these screws and will lead to quicker failure.
You are basically doubling the weight that each screw needs to support by only using 2 instead of the 4 screws.

As some of the after-crash pictures show, that when a single screw fails, the other screw is not able to support the weight of the Quad against the motor and will also fail. At least with the short M3x8 screws.
With that it’s in my opinion very short-sighted by 3DR to only use 2 screws for the IRIS motor mounts.

I invite everyone: Please go ahead and measure the dimension on your IRIS to confirm or refute what I measured and documented above. If you find out that my measurements make sense, please replace your motor screws (don’t forget the loctite).

As this is the official 3DR support forum, I would like to ask 3DR why your IRIS arm design only uses 2 screws per motor for mounting. Are you looking into re-designing the IRIS arm for a future version with the ability to use 4 screws per motor for mounting?

2

Thank for for this info! I just ran into this issue with the 8 mm screws.

For anyone else who is new to this like me, there’s an additional issue that compound what’s already been mentioned. The screws that lock the short legs in place do not have a stopper on them. Being that the motors are made of a soft metal, it’s possible to over tighten them, dent the motor, and put further pressure on the 2 screws holding the motor to the arm.

I discovered this today. I will say this is partially my fault as I over turned the screws. Other than that, I did have difficulty just getting the screw in the shortened legs and this further exacerbated me putting the screws too far in.

See the image attached.

3

I’m not 3DR but using 2 screws per motor is pretty much standard in the multicopter world. I only use 2 screws per motor in all my builds and I have built quads in the 70cm class with much higher AUW than an Iris.
And everybody I know or read from uses only 2 screws per motor.
I would guess that under normal circumstances, the motor mount would break before the screws - even CF plate motor mounts. At least that’s again my experience from my many crashes which resulted in many broken motor mounts but never in broken screws.
R/C motors usually have 4 (or more) holes because it’s not 4 holes but 2x 2 holes in different patterns so they can fit different mounts. In a 28-size motor that would normally be 16mm and 19mm patterns.
To illustrate this: My Turnigy Multistar 4822-390, which are rebranded HengLi motors have actually 7 (seven) mounting holes in 3 different patterns! That doesn’t mean, I’m supposed to use 7 screws with that motor. It only means, I have more options mounting it to different mounts. They also have 4 holes to mount a propeller yet normal multirotor propellers are mounted only with 2 screws.

4

That’s interesting. Thanks for your reply. Although I have two counter-examples for RFT models. :slight_smile:

These are two models that directly come to my mind, yet I don’t own them. But after having a closer look at them via available documentation, I discovered the following:

Again, the combination of 2x screws per arm and these screws being too short is probably the real culprit. With the correct length screws (M3x10mm) 2x screws per motor it might be sufficient. Only mechanical destruction tests could tell.
Yet why not play it safe and use all four. Cost difference would probably be below 5 USD.

5

Weight - or the lack thereof - is a precious commodity in multicopters! A few screws here, a few screws there quickly sum up to a few minutes less airtime, especially in smaller birds.
And - as I wrote - it’s not necessary. Before those 2 screws will break, the plastic arm will have broken much earlier - assumed that the screws have the correct length, of course. With the wrong length, even 4 screws won’t help you much, but the length problem seems addressed now, doesn’t it?

6

Your SteadyDrones assembly guide uses 3 + 4 screws for the roll and 4 +4 screws for the pitch motor of the brushless camera gimbal. For gimbals you do use more screws because of the leverage forces perpendicular to the motor plane. It does not, however, address any assembly of drive motors…

7

Yes, you are right! Sorry for that oversight. The product picture does show that all four holes of the mounting plate (as shown in the assembly guide) are used.

8

In the product picture I can’t see any motor screws clearly. Only the screws that clamp the motor mount halves together against the arms and for that, of course, 4 screws are needed as there are 2 rings for each motor mount :slight_smile:.