Well...It Crashed! Tarot 650

[dmetz]

So I thought I had built the titanic...well...in MR terms, an un-crashable ship. I studied this and other forums and felt like I had built a fairly meticulous craft. In fact, earlier today, following many, many hours of build time, I launched her on the maiden voyage; no spectators, no cameras. I had faith and hope, among the reservations, that all would go well....and it did! The Tarot 650 quad with Quadrocopter QC 3328 motors, Carbon Fiber 1365 props, AKE 30 amp ESC, Wild Scorpion 4S Lipo 5500 30C and DJI Naza M V2 FC lifted and flew superbly.

It seemed perfectly balanced, held its position and moved nicely between GPS, ATTI and Manual modes. The first flight was 9 minutes and the second (same battery but very low altitude) was around 6 minutes prior to the lipo alarm sounding at 3.4 volts per cell. I later charged the battery. Feeling confident now, I invited the family out to the front driveway to watch. I had previously launched the craft on this fully charged battery for only one minute. I lowered the craft and disconnected the battery. I failed to put a battery monitor on the lipo but knew I would only be in the air a few minutes. Having recorded 15 minutes of flight time on this battery earlier in the day, I was not worried. So...the Tarot 650 launches, GPS mode, all green....hands off the controller and it is steady as a rock for about 1.5 minutes and then all hell breaks loose. The LED flashes RED, the Tarot banks to the left and all I could do from ten feet up is reduce the throttle and let her topple over onto the brick paver driveway. The damaged seems minimal; a scuffed motor shaft and a few scratches on the DJI GPS disk. I recalibrated the GPS only to get the flashing red indication once again.

I trouble shot the system to no avail until I placed the lipo voltage monitor on the battery. Holly cow....2.7 per cell!?! The battery was depleted....but how? Wondering now if I have a short or something? I was meticulous with wiring this thing. Why would a freshly charged battery die so quickly? Anyway....just wanted t share my experience. Obviously I have some work to do. This is unacceptable. I tried so hard to create a craft and environment where this would not happen but still failed by "assuming" that my lipo voltage was alright and by not verifying with a voltage monitor prior to and during that short flight. Live and learn. I have a Tarot 960 essentially built and just waiting a FC and final setup but crashing it WILL NOT be an option. I need to get any kinks out now. A big thanks to Bart and others on this forum that have helped answer questions and provide information through previous posts and communications.

 

[Bartman]

Sorry to hear it, maybe there is a misunderstanding going on with the charger? It sounds like the battery wasn't charged even though you thought it was.

 

[Motopreserve]

Dmetz,

Welcome to the forum and sorry your first post had to be under these circumstances. Bart has a good point about the charger. What make/brand is it?

Don't be discouraged , but I think crashing is an inherent part of this hobby. Even the best builds use so many components that we don't make ourselves (ESCs, batteries, motors etc), and we rely on the quality control of manufacturers that of course are not perfect. So that leaves us at the mercy of outside influences.

Having realistic expectations has become part of my mindset for this hobby - and although I try to build as meticulously as possible, I always know in the back of my mind that some of these components can fail. Even the pricier, supposed high-quality gear is not infallible. Personally, I think the best we can do is plan and execute well, double check our work and take it slow for initial testing. The rest comes down to a little luck from the MR gods.

I would run a few charge/storage cycles on the charger through that battery to see if anything pops up to give an indication of a bad battery before risking it on a craft again.

 

[dmetz]

Thanks Bart and Moto for your words of encouragement and suggestions. I do wonder if the battery didn't take the charge even though the charger indicated that it was. The charger is the Imax B6, balance charger.

I thought the DJI Naza V2 PMU and FC LED would indicate a low voltage situation well before the battery was depleted and the craft just dumped. I'll have to check my FC/PMU settings via the setup software.

At least I was at 10 feet and not at 100. For the break end period, that just makes sense. Guess I'll be going over components today and never a flight again without a separate voltage monitor on the battery.

 

[Motopreserve]

Good call. Double check your low voltage settings in the Naza assistant - and of course a low voltage alarm.

 

[tstrike]

Hi Dmetz, I see you're using AKE 30 amps for your esc. Did you disconnect all but one red esc wire going in to the flight controller? I see on a site selling them they state:

User Note 1: When the system is installed with more than one ESC, please keep only the red one of several ESC signal cables is in leading through state, then disconnect the rest ESC red signal cable, the way can improve ESC inner BEC working efficiency, and also can avoid unnecessary power consumption.
User Note 2: Highly recommend using AKE quarter cable, six cable or nine cable to transfer power for the ESC.

just a thought, sorry your masterpiece went down, if you're not crashin, you're not flyin...

 

[dmetz]

http://www.cnchelicopter.com/servlet/the-2476/multiaxis-dedicated-high-dsh-speed-high/Detail

Thanks for that advice Tstrike. Actually, I purchased the 30 amp model of the AKE ESC referenced above which is supposed to be an OPTO version so no internal BEC. These particular ESC's are rated for up to 6S as are the motors so I wanted to be able to use either a 4s battery OR a 6S. Good consideration though and thanks for checking.

I have disabled the voltage protection on the NAZA M V2. I've read on this forum where others recommend that? I feel better, I think, trusting a voltage monitor attached to the battery and would prefer the FC not shut me down.

I've got my LIPO voltage monitor warning indicator set to 3.6 volt/cell now but that seems high to me?? I only managed about 5 minutes of flight time with this setting but perhaps that's what I can expect when I'm not literally depleting the battery (albeit unintentionally). I'm going to search more on the forum to see what the convention is on low voltage with Lipo alarms.

I'm currently running a 4s 30 C lipo but I built the system to accommodate a 6S Lipo. I had planned that I might need to reduce the prop size when using my 6S Lipos but I'm curious as to how the system might perform with the current 13.5 inch props and the 6S battery.

I appreciate everyone's input and support...a really great community here.

 

[fltundra]

http://www.cnchelicopter.com/servlet/the-2476/multiaxis-dedicated-high-dsh-speed-high/Detail

Thanks for that advice Tstrike. Actually, I purchased the 30 amp model of the AKE ESC referenced above which is supposed to be an OPTO version so no internal BEC. These particular ESC's are rated for up to 6S as are the motors so I wanted to be able to use either a 4s battery OR a 6S. Good consideration though and thanks for checking.

I have disabled the voltage protection on the NAZA M V2. I've read on this forum where others recommend that? I feel better, I think, trusting a voltage monitor attached to the battery and would prefer the FC not shut me down.

I've got my LIPO voltage monitor warning indicator set to 3.6 volt/cell now but that seems high to me?? I only managed about 5 minutes of flight time with this setting but perhaps that's what I can expect when I'm not literally depleting the battery (albeit unintentionally). I'm going to search more on the forum to see what the convention is on low voltage with Lipo alarms.

I'm currently running a 4s 30 C lipo but I built the system to accommodate a 6S Lipo. I had planned that I might need to reduce the prop size when using my 6S Lipos but I'm curious as to how the system might perform with the current 13.5 inch props and the 6S battery.

I appreciate everyone's input and support...a really great community here.

What radio system are you using? Telemetry is the only way to fly now.
Unisens-E and the Frsky lipo sensor, wouldn't fly anything of value without both. pack volts, cell volts, watts, current, mah used, height, rpm, and alarms galore with the Taranis.

 

[tstrike]

Yeah, shut the voltage protection off, 3.6 is alright, fly until the alarm goes off and see how much mah you put back in the battery and adjust from there. Once you have an idea, set a timer on your Tx so you have a rough idea how much flight time you have left instead of waiting for a buzzer or a red light. I know I can get a clean 8 minutes with my 550 quad carrying a gopro and bg on a 4k 4s batt, so I set my tx timer to 7 minutes knowing when the timer goes off I have a minute or so to get back on the ground and not stress my batteries.


Thing is, if I can keep it up for 8 minutes, I'm a porn star in my mind...metaphorically speaking of course...

 

[dmetz]

What radio system are you using? Telemetry is the only way to fly now.
Unisens-E and the Frsky lipo sensor, wouldn't fly anything of value without both. pack volts, cell volts, watts, current, mah used, height, rpm, and alarms galore with the Taranis.

Yes....I'm rapidly coming to this conclusion! I have a Spektrum DX9 transmitter and AR8000 receiver so I should be able to add a TM1000 telemetry module. In fact, I just purchased two while writing this response. Thanks for the "push"

 

[Motopreserve]

Yes....I'm rapidly coming to this conclusion! I have a Spektrum DX9 transmitter and AR8000 receiver so I should be able to add a TM1000 telemetry module. In fact, I just purchased two while writing this response. Thanks for the "push"

im with Tundra. Although I used to do all the calculations before having telemetry, and still use a timer, I now feel like I can't fly without telemetry warnings. The taranis also allows for setting up voice alerts (not sure if Spectrum does this) which triggers a young woman to warn me that it's time to come home

i keep an excel sheet on my phone for logging the batteries - so that I can track how many mah I put back in and compare to the telemetry (frsky is off by 100mah or less) and monitor the health of the batteries.

 

[fltundra]

I don't discharge more then 80% ever, or no less then 3.6 volts under load. I fly my 8000mah packs to 6000 mah occasionally, 5000 mah on a regular basis on the Unisens-E readings, and it's always within a few percent for ma back in with my cellpro charger. It's a perfect combination to monitor Lipo health.

 

[cootertwo]

Well, the first post says you pulled the battery down to 2.6 volts per cell. That's too low. It may have permanently killed the battery. At least do a good balance charge, and then see if what the readings are for all the cells. And I do shut off the Naza voltage warning on everything I own. 3.6 on a beeper will work, but when it goes off, don't hesitate to land. Funny/strange how some MC's react to the voltage dropping. If I push mine (all Naza Lite), past the voltage warning, most all will fly till I notice I have to push the throttle past half, to hover, then they just settle slowly down to a soft landing, and not enough left to even lift off the ground. (2.7/3.2). But I do have 2 that will fly fine, and when the battery gets too low, they just fall out of the air like you shut them off.

 

[dmetz]

im with Tundra. Although I used to do all the calculations before having telemetry, and still use a timer, I now feel like I can't fly without telemetry warnings. The taranis also allows for setting up voice alerts (not sure if Spectrum does this) which triggers a young woman to warn me that it's time to come home

i keep an excel sheet on my phone for logging the batteries - so that I can track how many mah I put back in and compare to the telemetry (frsky is off by 100mah or less) and monitor the health of the batteries.

Moto....yes the Spektrum DX9 does have voice alert. A spry young English girl who sometimes proclaims "Throttle High" on power up!!! ;-) The excel sheet for logging batteries is an excellent idea. I'm sure it will be mandatory for commercial work once the new regulations hit. I have a Commercial Pilot License (thought not current), Flight Instructor, Ground Instructor and Amateur Radio License (General Class), so I imagine I have a good leg up for commercial shooting. Now I just have to understand and implement all the "Best Practices" of MR building and operating and I will feel more confident. BTW....Is there a "Best Practices" publication of sorts for Multirotor construction and operations? If not, I bet there will be when the legislation hits. Regulation almost always requires standards.

Thanks for all the help!

 

[dmetz]

Well, the first post says you pulled the battery down to 2.6 volts per cell. That's too low. It may have permanently killed the battery. At least do a good balance charge, and then see if what the readings are for all the cells. And I do shut off the Naza voltage warning on everything I own. 3.6 on a beeper will work, but when it goes off, don't hesitate to land. Funny/strange how some MC's react to the voltage dropping. If I push mine (all Naza Lite), past the voltage warning, most all will fly till I notice I have to push the throttle past half, to hover, then they just settle slowly down to a soft landing, and not enough left to even lift off the ground. (2.7/3.2). But I do have 2 that will fly fine, and when the battery gets too low, they just fall out of the air like you shut them off.

Yes....unfortunately I realized I may have damaged this battery. It was of course unintentional and a good lesson....will NOT ever happen again unless my redundant systems all fail. I should have not had to repeat this lesson as I'd seen videos of this happening on the internet. I was confident in my battery charge but dumb not to confirm it. I hear ya on the FC response from a dead battery. It surprised the heck out of me when she just feel out of the sky. Wish I had been over grass but glad I was just 10 feet AGL in a hover. Surprisingly little damage fortunately. The props look in fair shape with just a few scratches and it flies fine now but I do plan to replace them to be sure.

 

[deluge2]

Well when I got to the end of the section of your first post excepted below, you may not have been worried, but I was! Before entering the world of multirotors, I was fortunate to have transitioned to Lipo batteries with RC trucks where the consequences of loss of power are far less dramatic!

You have obviously spent a lot of time assembling your craft and are taking a thorough, thoughtful approach to learning from this incident. As mentioned in later posts, logging of both flight info and battery charge details is probably a good idea! You will want to know how much energy you are putting into your batteries after flights of different durations, different loads, different flight plans. Whether you use time, voltage alarms, and/or telemetry, these are your fuel gauges. As you know from your FS flight training, fuel issues are important on the ground and paramount in the air. Especially with a craft that simply won't glide or autogyro. Think of charging as filling the tank. Clearly you want to know how much fuel (energy) you're putting in. The energy input is directly related to net mAh delivered so monitoring and recording that is a useful measure of energy consumption.

I'm glad the damage was minimal, except perhaps for the battery. Also, might as well get several batteries so you're not charging 'in a hurry'.

Steve

<SNIP>...The first flight was 9 minutes and the second (same battery but very low altitude) was around 6 minutes prior to the lipo alarm sounding at 3.4 volts per cell. I later charged the battery. Feeling confident now, I invited the family out to the front driveway to watch. I had previously launched the craft on this fully charged battery for only one minute. I lowered the craft and disconnected the battery. I failed to put a battery monitor on the lipo but knew I would only be in the air a few minutes. Having recorded 15 minutes of flight time on this battery earlier in the day, I was not worried...<SNIP>

 

[Motopreserve]

I'm not sure there is a "best practices" guideline for building these things. At least I haven't seen one. They are still relatively new and the technology is advancing so rapidly - and like most hobbies, offer a wide variety of opinion.

Although more about build design than assembly: I've become somewhat frustrated by the lack of good info (or worse, misinformation) on some products - Sunnysky motors for example. I ran into a situation where I ended up spending a bunch of time and $ chasing down the perfect ESC to work with "pancake" motors, only to find that the pole count listed for my motors was incorrect. I found myself wanting more expert opinion and empirical data.

So I went and contacted the folks at KDE Direct and asked them to explain what defines pancake, what is high/low pole count, advantages and disadvantages etc. I just heard back today and will be posting their responses soon. Hopefully this can become a trend and we can begin to compile threads of similar feedback and data.

As I said in my initial post, we are at the mercy of these manufacturers. The best we can do is have the best data available to piece together these machines with the hope of eliminating as many pitfalls as possible. Then we build them as meticulously as possible, and hope for the best.

 

[dmetz]

Well stated Deluge. I would never "trust" a fuel gauge in the air so I just simply need to better understand the nuances of LIPO management and capacity. After cycling this 4S up a couple of times, I was able to get 12 minutes of flight time again this evening without even reaching the alarm threshold of 3.6 volts per cell so I guess I have more work to do to figure out what the real capacity of the battery is (with safe reserve) and what reliable and consistent results I can achieve on the MR. I'll be recording the mA during the charges to get a good feel for what I'm burning and "refueling".


I configured this craft to work with either 4S and 6S batteries. I wonder if anyone else goes back and forth? Does the ESC throttle calibration that I did with the 4S battery carry forward to a 6S battery? Perhaps the calibration is only concerned with the receiver and throttle and not battery dependent? I did put the 6S battery on for a very short flight today and noticed that the throttle was VERY touchy with the extra voltage. Perhaps I need to dial down the throttle influence with DJI assistant to find a happy medium for both the 4S and the 6S; probably a little slow for the 4S and a bit fast for the 6S but not too severe on either spectrum? Any thoughts on this?

 

[Motopreserve]

I don't think you can rely on only one method either. The telemetry (or fuel gauge) is another tool in the arsenal to help you avoid what happens when you don't have a handle on the state of the battery. There is no substitute for doing the math. With one of the frsky sensors I have there is a direct readout of the battery (per cell up to 6S), just like the voltage checker you keep in your toolbox, but showing real time changes on your Tx screen. It's like having the voltage alarm - but on the ground with you instead of only on the craft. A multimeter can quickly determine it's accuracy of whether it's a trusted voltage reading.

As far as the 4S vs 6S batteries: I think the battery choice needs to be made based on the craft's features and components. If the MR weighs enough to warrant a jump to 6S, go for it. But if it's flying well on 4S, and seems to be in the "pocket" hovering at 50% and has good maneuverability, you've probably already found your sweet spot for the current weight. I thin you'd be better off trying to find a more appropriate prop for the 6S as opposed to upsetting good flying characteristics that you have with 4S. Efficiency and flight times can be tweaked based on these choices, but it can go both ways. You could overpower and lose efficiency too.

 

[dmetz]

I don't think you can rely on only one method either. The telemetry (or fuel gauge) is another tool in the arsenal to help you avoid what happens when you don't have a handle on the state of the battery. There is no substitute for doing the math. With one of the frsky sensors I have there is a direct readout of the battery (per cell up to 6S), just like the voltage checker you keep in your toolbox, but showing real time changes on your Tx screen. It's like having the voltage alarm - but on the ground with you instead of only on the craft. A multimeter can quickly determine it's accuracy of whether it's a trusted voltage reading.

As far as the 4S vs 6S batteries: I think the battery choice needs to be made based on the craft's features and components. If the MR weighs enough to warrant a jump to 6S, go for it. But if it's flying well on 4S, and seems to be in the "pocket" hovering at 50% and has good maneuverability, you've probably already found your sweet spot for the current weight. I thin you'd be better off trying to find a more appropriate prop for the 6S as opposed to upsetting good flying characteristics that you have with 4S. Efficiency and flight times can be tweaked based on these choices, but it can go both ways. You could overpower and lose efficiency too.

Really great advice on power management and power monitoring. I'm overwhelmed by the responses and great information relative to my little "mishap". Thank you!

You know...I think you are right on target with your comments about the 4S vs 6S dilemma. The craft does seem to be well tuned and configured with the 4S battery and with a full charge it looks like 10 minutes flight time will be the norm. The quick fight I did with the 6S battery just seemed like the MR was over powered...definitely not the smooth response of the 4S.