New Cinestar 8 HL - Equipment failure caused bad crash on 3rd flight

Discussion in 'Cinestar 8' started by Philip Lima, Jun 12, 2014.

  1. Philip Lima

    Philip Lima Member

    Joined:
    Feb 18, 2014
    Messages:
    67
    Likes Received:
    9
    I have had a Cinestar 8 HL about a week now and I have flown it 3 times in total. Yesterday afternoon I was flying it at a local park, luckily no MoviMR/Epic was attached, I was roughly 7 minutes into the flight and the telemetry was telling me that I still had plenty of voltage and had burned through roughly 4500mah out of the 10000mah 6S pack that was only used a few times before. I was hovering roughly 20ft off the ground over the grass when the MZ24 voice spoke up and said motor error. The cinestar was still flying ok, though not perfectly, definitely less stable. I began to plan my next move to bring the copter back down and see what the error was all about it. Before I had a chance to do anything, about 3 seconds after the motor error voice, I saw a massive spark/flame on the cinestar and the MZ24 immediately said low voltage warning as the octo fell from the sky. It tumbled sideways and fell directly on it's side. Luckily there was no one around and I was flying directly over grass, otherwise it could of been much worse! After the shock passed I walked over to asses the damage. I immediately disconnected the battery, which was no longer functioning as the copter was no longer powered on and connected to the telemetry. I did not stop the motors with the MZ24, they stopped on their own immediately after the low voltage warning. I put the cinestar in the car and drove it back to our office.
    After taking everything apart, I now have a pretty good guess as to what the heck happened. As you can see from the pictures, motor #6's ground wire is no longer attached to the motor controller/PDB. My theory: I was flying, the solder failed and the ground wire came loose off the board, about three seconds later it moved just enough to meet up with another power source, which shorted out, and in tern shorted the battery and caused the octo to fall from the sky. The only other option that I can see is the the actual MK BL board failed and caused the short. I did all the standard pre-flight checks and calibrations on the battery and cinestar and everything checked out perfectly.
    I'm extremely happy that we were still in the testing process with this cinestar and had not started using it with our Movi MR and RED Epic on it yet. If those items were attached, we would be looking at a massive cost to get everything fixed. I have been flying multirotors for over 4 years, and have been flying RC helis and planes for another 10 years past that. I have never experienced a failure like this. I've made mistakes that have had consequences... But I've never experienced 100% equipment failure that causes the craft to go down without any chance of recovery. This makes me EXTREMELY nervous to put our Movi MR and RED Epic on this platform, or even just to fly it in a public park as I would any of my other multirotors for that matter.

    As I mentioned before, I had flown this cinestar twice before without any problems. It performed flawlessly, though the flights were much shorter(in the 3-5 minutes range between landing and giving it a roughly 5 minute break to verify voltages on the battery's cells and check the motors and all electronics for any heat problems). This was the first flight that lasted a little longer, and I'm guessing that is was caused things to either melt or fail.
    Has anyone else had a similar failure to this? We purchased the Cinestar HL as a built RTF and they did a quick test flight with it as well. I did a test flight with it being tethered, and then another one without anything and then this happened I have a few years experience with multirotors(and over 10 with other flying RC stuff), but have never worked with MK stuff before, so this is not helping boost my confidence in the reliability of the MK "kit". The Cinestar did not have navi/GPS when this happened, just the flight control, and I was flying in full manual mode(not altitude). The battery leads were damaged in the crash, so it is a throw-away, but I am able to see the voltages of the cells through the balance plug and everything is right where it should be voltage wise.
    Any input from those more familiar with the MK stack would be very much appreciated! I was flying this exact setup, that was built and setup by Quadrocopter: http://www.quadrocopter.com/CineStar-8-Heavy-Lift-MK-Basic-Kit_p_1152.html
    Here are some pictures of the damage...
    _MG_6970.jpeg _MG_6971.jpeg _MG_6973.jpeg _MG_6976.jpeg _MG_6979.jpeg _MG_6980.jpeg _MG_6981.jpeg _MG_6982.jpeg _MG_6983.jpeg
     
  2. Jason Smoker

    Jason Smoker Active Member

    Joined:
    Oct 23, 2012
    Messages:
    860
    Likes Received:
    67
    :( damn!!!

    pity you dont have gpx file to go with this. Contact Quadrocopter there good with customer service!!
     
  3. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    Sorry to hear this, thanks for posting. This looks like what happened to Steve Maller the first time he powered his original 2XL powere board with a lipo. Are you logging your flights with the GPX SD card? This is essential with the MK stuff to get a good idea of what is going on with anything that be an issue.

    My gut feeling is telling me this is not an install issue or poor solder joint. My guess is this is a board failure or an intermittant short in the wiring that caused a main power line on the BL circuit to short once the copter was up in the air and vibrating. I think an inspection of all the wiring is in order for sure. Check the motor wiring for the BL that failed really good. I would also check the motor itself to make sure that it didn't just have a hard failure.

    Another thing to look at is to if anything metal could have jarred loose that short circuited that BL. Any chance of significant moisture in the air or the slightest chance of rain drops that could have gotten on the electronics?

    Do you have any GPX files that we can look at that might help?
     
  4. Philip Lima

    Philip Lima Member

    Joined:
    Feb 18, 2014
    Messages:
    67
    Likes Received:
    9
    Thanks for your response. Unfortunately, I was flying with just the basic MK setup, I just received the Navi and GPS boards today, but since they were not installed at the time of the crash, I do not have any GPX files.

    I checked the wiring and other than the damage from the short/explosion there is no damage I can find. As for moisture, it was a beautiful sunny day, no chance of water on it at all. Unfortunately I have no way of testing the motor since this is my only large setup at the moment and it obviously wouldn't power on in it's current state! I also don't have any extra ESCs with enough amps to give the motor a push at the moment.
     
  5. Philip Lima

    Philip Lima Member

    Joined:
    Feb 18, 2014
    Messages:
    67
    Likes Received:
    9

    So far Quadrocopter has been good, I sent them a very long email explaining what happened with detailed photos. That was this morning, so hopefully I'll hear back from them tomorrow morning and we can get a game plan together to get the thing back in the air.
     
  6. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    I emailed Holger and alerted him to this thread for you. Any chance you think you had anything on the copter like a small screwdriver or screws/fasteners or anything metal wise that might have forgotten about that could have moved around and shorted the board? It would be curious to know what firmware you had with the board because I know that the most recent firmware has had some significant improvements.
     
  7. Gary Haynes

    Gary Haynes Administrator
    Staff Member

    Joined:
    Jul 6, 2012
    Messages:
    5,211
    Likes Received:
    460
    Perhaps the MOSFETs blew up resulting in the damage. The other visible solder joints look fine. I forget but when Steve had his failure did the MOSFETs show damage as in being totally evaporated?
     
  8. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    No not quite. Here's the damage.
    IMG_8915.jpg
     
  9. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,979
    Likes Received:
    807
    When mine failed it was on the bench with nothing attached to it (not even any motors). So it could be a similar issue, but in Philip’s case, he was in the air under load. But given the proximity of the damage to the main LiPo leads, I’m suspicious of a major board failure. Or, the MOSFETs could have failed, causing a chain reaction that shorted the main LiPo leads, which would cause a very large spark and would essentially arc-weld everything near them.

    Whatever did happen, I’m really sorry about it Philip. I know that I appreciate whenever anybody shares a bad experience like this, as it not only reinforces the strength of the community, but it makes us all more aware, and finally, will hopefully lead to better and more reliable products.

    Good luck, my friend. I hope you rebuild her better and stronger.
     
  10. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,711
    Likes Received:
    311
    I wonder if not having the latest update could have prevented the issue. Check out the email from Holger that he sent out about a month ago.

    We were analysing some defective boards from customers and were were able to simulate some effects that can cause extremly high current peaks that could even damage the board.
    Therefore we made some small, but very effective software-changes that eliminate these effects.
    We strongly recommend to update your V3-Boards.

    Here you can find the update and some examples of the connectors.
    http://www.mikrokopter.de/ucwiki/en/BL_V3Update
     
  11. Philip Lima

    Philip Lima Member

    Joined:
    Feb 18, 2014
    Messages:
    67
    Likes Received:
    9
    Thanks for letting him know. There was definitely nothing on the copter that shouldn't of been. I'm very careful and inspect everything before all flights, I have had over 500 flights on my other AP setups and have not had a single issue so far. As for the firmware, I am not sure what it was running since I had just received it and am waiting to receive a windows based machine so I can use the MK tools.
     
  12. Philip Lima

    Philip Lima Member

    Joined:
    Feb 18, 2014
    Messages:
    67
    Likes Received:
    9

    Thanks for all the info Steve. I was in the air under load as you mentioned, but I definitely wasn't pushing it at all. I had flow way more aggressively earlier in the flight, but when the chain of events started to unfold causing the crash, I was just hovering. I was actually talking to my friend(and creative director at the company I work for) who was standing next to me explaining the telemetry read outs on the MZ24, so I was literally just hovering, only giving slight input on the controls to keep it in one place since I don't have the navi/gps and was flying in manual mode.

    I definitely plan on rebuilding her, though it makes me a bit nervous to trust the MK stuff for this kind of setup. I know many people use and trust it, but our plan for this rig is to fly our Movi/Epic setup we currently use on the ground(just got the MR kit yesterday for it). So... It makes me extremely nervous to send that much of an investment out on a flight over the ocean.
     
  13. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,979
    Likes Received:
    807
    As you have a lot of multi-rotor experience, you probably know this, but from a heat perspective, hovering in one spot is actually sometimes more taxing on the BLs than flying around, even somewhat aggressively. That’s because there’s little or no airflow over the BLs, which is how they cool themselves. The MK BL 3.0 boards have very efficient cooling, but they’re not perfect, and it could have been the extended hover that did you in. Not that you did anything wrong, mind you. I’m just pointing something out. Obviously without the Navi board and its associated data logging capability, your BL temps were not recorded.

    But this may not have been heat-related at all...we may never know. Same for Dave’s comments above about potential firmware issues.
     
  14. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,979
    Likes Received:
    807
    But I do find it interesting that in both cases (yours and mine) the MOSFETs that imploded were right next to the main LiPo leads.
     
  15. Sean Haverstock

    Joined:
    Mar 6, 2013
    Messages:
    64
    Likes Received:
    1
    But a HL setup with no payload? Or did you have a test payload installed? Do the MK electronics heat up that bad? Never used them. Could be as simple as a cold/bad solder joint. After all, the telemetry indicated a motor failure - then the short ensued. Phil, I'm convinced of your original assessment.
     
  16. Philip Lima

    Philip Lima Member

    Joined:
    Feb 18, 2014
    Messages:
    67
    Likes Received:
    9

    Good info Steve, thanks for sharing your knowledge and experience! I still tend to think it wasn't heat related, just because this copter was designed to lift so much more.
     
  17. Philip Lima

    Philip Lima Member

    Joined:
    Feb 18, 2014
    Messages:
    67
    Likes Received:
    9

    There was no test payload, it was only it's 2nd un-tethered flight, so I was planning on at least another 10 or so flights before I started load testing with a test payload.
     
  18. Ozkan Erden

    Ozkan Erden Distributor

    Joined:
    Aug 3, 2012
    Messages:
    678
    Likes Received:
    131

    I'm very sorry about this crash.

    I can understand how and why you can't trust MK 6S system. It took me barely a month to figure out an stability problem and the worst part I'm still not sure what caused it.

    BL Ctrl 2.0 has been working flawlessly for many years; we have yet to see the same stability from BL Ctrl 3.0. But, you should gain your confidence asap and you will build your copter more reliable than before. When adopting new stuff, we should follow the updates & upgrades as close as possible since they are not as mature as the previous version.

    Just feel happy because your copter only damaged itself. No worries. You will be back very easily.
     
  19. Jeff Scholl

    Jeff Scholl Distributor

    Joined:
    Jun 29, 2012
    Messages:
    240
    Likes Received:
    50
    Hi Phillip,

    I understand you bought this CS8 from Ryan Sheets recently?
    When this was built back in April I believe we still had version BL vs3.0.


    Currently MK is shipping v3.2, but let me double check with the shop and see if Ryan ordered a newer board. Did he mention any details when he sold it to you?

    In the past I have seen mosfets blow on the bench which normally produces a little smoke puff and a cracked mosfet. This is the first where I have seen most of the components arc. (Except back when we used to run 12s on single rotors but that is another story)

    As far as heat goes I've been lifting 9-10 pounds with a clear hood and never has my British girlfriend ever reported anything over 60C.
    These new boards a much better job of cooling even at the higher loads. My CS6 with the older BL2.0 with the hoodinstalled recently gave me 90C temps so those that use the hood with the older BL2.0 please monitor the temps and perhaps cut a few holes off to the side.

    Best,
    Jeff
     
  20. Philip Lima

    Philip Lima Member

    Joined:
    Feb 18, 2014
    Messages:
    67
    Likes Received:
    9
    Hi Jeff,
    You are correct. Ryan had quadrocopter put it together and then had a change in plans and no longer needed it, so we purchased it from him and had it shipped directly from quadrocopter. Based on what Ryan told me, and the video you posted of quadropcopter testing it(published on Apr 25th), it was definitely a mid-April build. I am not sure if any firmware updates were done before it shipped(I assume no hardware updates would of been done) to us on June 4th, but I'm guessing not since Ryan wouldn't of requested and I wouldn't of known. Since Ryan did not mention any specifics about the board when we bought it from him, I can only assume it was not upgraded.

    When the crash happened, there was definitely a puff of smoke right when the failure occurred and before the voltage error came through on the MZ24. The puff of smoke was the first thing that happened following the motor error I heard through the MZ24. Is it possible that the mosfet blew in a more violent manner than you've seen before and somehow caused the motor wire to come disconnected from the board, then that motor wire is what caused the arc? Just another idea...

    I do remember looking through every telemetry screen on the MZ24 just about 30 seconds before the crash and nothing stood out, no high temps/weird voltages or anything like that... Also, wouldn't the telemetry of reported if the temps did get too high in a similar fashion to how it did when it let me know about the motor error? As I've mentioned, I'm no expert when it comes to the MK stuff so I could be way off...

    Thanks for your help Jeff!
     

Share This Page