Power Module Failure During Flight

This is a subject transfer from the developers forum where much discussion has been made regarding the separation of a small coil on the power module. It is our best guess that this small black cube located next to the two larger capacitors on the ESC board can vibrate loose during flight. We have three documented cases so far where this has occurred, and all have resulted in significant damage to the Iris. I would suggest that you apply a small mound of silicone around this black cube (coil) to prevent the thin copper strips from cracking loose. Two of these failures occurred on older “developers” models, but the third just occurred on a brand new “consumer” model.

Good to know thanks… do you have a photo of the black cube?

Any recommendations on what silicone to use?

It would be good if 3DR could step in and comment as to if this is a good idea…and if so, answer this question.

I have reviewed 3 crashes with this problem and in 2 out of the 3, there is clear evidence that the component was damaged in the crash, not the cause of the crash. The logs indicate that Ken F’s crash that you posted in was also pilot error, and not caused by any component failure.

Actually, if the component is coming loose or getting damaged from crashing (which is going to happen regardless) it might be a good idea to take some preventative measures and silicone the component down so that it wont get damaged or come loose in the event of future mishaps.

Just my 2 cents.

Sorry for the delay in getting back to your questions. Just a quick update on the cause of this crash is in ordersince Craig had a chance to review the video and tLog. It appears that this indeed was a secondary issue and not the cause of the crash. The best guess for this issue is a possible “brown out” of the pixhawk system and not the ESC system. In any case, I have attached two photos of the little black cube as found when it detached. The other photo is the fix that still remains to be seen if it will hold up over time. I used a regular GE Silicone 2 window caulk to tack it down. Nothing special as this part does not generate any significant heat.

[attachment=0]Fixed.jpg[/attachment][attachment=1]Broken Coil.jpg[/attachment]

What’s interesting is that the two capacitors on my IRIS have a slab of silicone around them to keep them in place. And looking at pictures of other folks, it looks like they have the same:
That silicone seems to be missing on yours.

I also tend to agree with CoastInternet: If that part comes loose on crashes, it probably should have gotten some silicone in the first place. And looking at the after-crash picture one can see that the coil is only soldered to the ground plate on one side with a very small soldering point. That piece is just asking for breaking off. Either after crashes or over time due to vibrations.

Craig…can you comment as to if it is ok to use silicone? I would prefer to use JB Weld…is it ok to use that? If ok to use either, how much should be used…how much of the board could be covered with either product?

Thanks.

To come back to this.
Two questions or asks for Craig or someone else from 3DR:

  • Without wanting to further go into the question of whether the damage was the cause or the result of a crash: Let’s assume we want to better tighten that component via silicone:
  • Is it ok to use silicone? Is it OK to use JB Weld instead?
  • If ok to use either, how much should be used…how much of the board could be covered with either product?
  • In order to see how well certain parts handle vibration and or shocks from crashed, did 3DR ever perform drop tests and vibration tests with IRIS?
    Here is what I mean with that:
  • Build a test-device that will assert force to IRIS via the legs to simulate a drop from 1 - 1.5 meters. Repea t that a couple hundred times and check after each x drops if pieces are still in place.
  • Place IRIS on a vibration pad and thereby introduce vibration from the outside. Let it sit on there for hours/days and check after x hours if pieces are still in place.

I don’t intent to replicate these tests myself, but knowing that 3DR did them would be very re-assuring.

  • Is it ok to use silicone? Is it OK to use JB Weld instead?
    Sure. I suggest something flexible
  • If ok to use either, how much should be used…how much of the board could be covered with either product?
    No more than 9 pounds … :slight_smile:
    I suggest a a blob about the size of a pea.

In order to see how well certain parts handle vibration and or shocks from crashed, did 3DR ever perform drop tests and vibration tests with IRIS?
Yes

The general procedure is to use a pendulum that swings the vehicle into a wall or something. It is easier to calculate and reproduce a consistent force

Thanks so much for the feedback! Really appreciated.

Now on to buying silicone. :slight_smile:
Does anyone have a link to Amazon for a recommended product?